1
The Science of Aphasia (X)
SOA
2009
Antalya (Turkey)
September 28 – October 3 2009
Program and abstracts
2
The science of aphasia X 2009
September 28 – October 3
Location: Sherwood Resort Antalya-Kundu
Organization:
Dr. Ilknur Maviş, Anadolu University, Turkey
PD Dr. Frank Burchert Potsdam University, Germany
Dr. Roel Jonkers University Groningen, The Netherlands
Website:www.soa-online.com
3
Program
Monday Sep 28, 2009
21.00 Welcome drinks
Tuesday Sep 29, 2009
09.00-11.00 Language and Math
Organizer: Carlo Semenza (University of Padova)
Philippe Pinel (CEA, INSERM, Orsay, France) –
Impact of the superior temporal cortex onto arithmetics and reading
cognitive functions
Frank Dohmas (Universitätsklinikum der RWTH Aachen, Germany) -
Number words are special.
Elena Salillas ( University of Texas, San Antonio, USA) -
The verbal code for Arithmetic: data from TMS and bilinguals.
11.00 Coffee break
11.30-12.45 Contributed papers (Syntax)
- Anjarningsih, Bamyaci, Hsu & Bastiaanse - The comprehension of time reference in
agrammatic aphasia: A comparison of languages with Tense and languages with
Aspectual adverbs
- Yarbay, Altınok, Özgirgin & Bastiaanse - Sentence Comprehension in Turkish Broca’s
Aphasia: An integration problem
- Bamyaci & Bastiaanse - Time reference in Turkish agrammatic speakers
- Ho & Bastiaanse -Verb retrieval and Aspect in Chinese agrammatic aphasia
- Vasiliki Koukoulioti & Stavroula Stavrakaki - Verb retrieval, argument structure and
inflection in aphasia: Insights from different verb classes in Greek
- Wieczorek, Huber & Darkow - Tense/Aspect Category in Fluent and Nonfluent German
Aphasia
4
13.00-14.00 Lunch
16.30-18.30 Workshop 1 Statistical tests using R
Shravan Vasishth – University of Potsdam
19.00-20.00 Poster session 1. (Syntax, Morphology and Auditory
Comprehension)
- Borgwaldt & Bose - The production of novel and lexicalized compounds in aphasia
- Grauel, Cupples & Law - <no titile >
- Güven, Eroglu, Kaçar & Zencir - Language Characteristics of a Child with Landau-Kleffner
Syndrome: A 4-Year Outcome Study
- Martínez-Ferreiro - Object clitics and reflexive pronouns in agrammatic Ibero-Romance
- Morawski – Idiom comprehension in aphasic populations
- Razak, Aziz, & Hassan - Complexity Difficulties in Sentence Comprehension Among Malay
Adults with Aphasia
- Schmidt – Artificial grammar learning by Borca’s agrammatic aphasics (a pilot study)
20.00 Dinner
Wednesday Sep 30, 2009
09.00-11.00. Motor Speech Impairments
Organizer: Wolfram Ziegler (EKN, Klinikum Bogenhausen, München)
Angela Morgan (University of Melbourne) -
Genes, brain and behaviour in developmental apraxia of speech
Wolfram Ziegler (EKN, Klinikum Bogenhausen, München) -
Apraxia of speech and the architecture of phonetic plans
Juliane Kappes (EKN, Klinikum Bogenhausen, München) –
Auditory influences on speaking: The role of imitation
5
11.00 Coffee break
11.30-12.30 Contributed papers (Motor Speech Impairments)
- Feiken, Jonkers, Meilof, Exter den, Sijbinga & Schönherr - How to diagnose Apraxia of
Speech? Towards the development of a diagnostic instrument
- Laganaro, Bagou & Croisier - Syllable frequency effect in progressive apraxia of speech: a
case study
- Varley, Cowell, Dyson, Roper, Inglis& Whiteside - Lexical therapy for apraxia of speech
- Maassen & Pasman - Clinical Use of Auditory Event-Related Potentials (AERPs) in
Children with Speech Output Disorders
13.00-14.00 Lunch
16.30-18.30 Workshop 2 Spontaneous speech analysis
Roelien Bastiaanse – University of Groningen
19.00-20.00 Poster session 2 (motor speech impairment; semantics and
discourse)
- Hurkmans, Jonkers, Feiken, Boonstra, Bruijn de, Arendzen, Reinders-Messelink – Oral
Diadochokinesis in Apraxia of Speech
- Matthias & Karanth – Conversational analysis: A comparison between non barin-damaged
geriatric and middle-aged adults
- Bergelson & Dragoy - Inefficient Fluent and Proficient Non-fluent: A dissociation between
micro- and macrostructure of aphasic discourse
- Ertan, Ünal, Sat, Sakar & Dikmen - Determining verbal fluency of Turkish speakers with
literate and illiterate educational background
- Mengotti, Negri, Corradi dell’Acqua, Trincia, Tessari, Lunardelli, Zadini & Rumiati –
Mapping language and action in the brain: Evidence from aphasia and apraxia.
- Nilipour & Yadegari - Evidence of Different Patterns of Semantic deficit in fluent and non-
fluent Aphasic Patients
- Tunçer, Demirkan, Şan & Maviş - “Frog where are you?” Narration Analysis in Turkish
Participants with and without Aphasia
20.00 Dinner
6
Thursday October 1, 2009
09.30–10.30 Keynote address:
Genetics of Language: Insights from developmental dyslexia
Franck Ramus (Laboratoire de Sciences Cognitives et
Psycholinguistique, Ecole Normale Supérieure, Paris).
10.30 Coffee break
11.00-12.30 Contributed papers (auditory comprehension, reading and
semantics)
- Albyn Davis & Alarie-Bibeau - Rapid Strategic Processing in Word Comprehension with
Aphasia
- Heide, Lorenz, De Bleser & Burchert - Effects of semantic transparency in reading German
prefixed verbs: Evidence from acquired dyslexia
- Hessler, Jonkers & Bastiaanse - The influence of phonetic features on aphasic speech
perception
- Kadyamusuma, De Bleser & Mayer - Perceptual discrimination of Shona lexical tones and
hums by Left and Right hemisphere damaged patients
- Soroli, Boudre, Hickmann, Nespoulous & Tran - Spatial cognition and language in aphasia:
crosslinguistic perspectives
- Dragoy, Iskra, Mannova, Kuzmicheva & Masalova - Verb retrieval in anomic aphasia:
effects of instrumentality and transitivity
13.00-14.00 Lunch
Afternoon: Excursion and social dinner
7
Friday, October 2, 2009
09.00–11.15 Crosslinguistic and multilingual aphasiology
Organizer: Roelien Bastiaanse(University of Groningen)
Speakers: Ilknur Maviş (Anadolu University, DİLKOM); Isabell
Wartenburger (University of Potsdam); Loraine Obler (Graduate
School and University Center, City University, New York); Roelien
Bastiaanse (University of Groningen)
11.15 Coffee break
11.45- 13.00 Contributed papers (language and math; diagnosis and
therapy)
- Benn, Zheng, Wilkinson, Cohen Kadosh, Siegal & Varley - Language and calculation: Brain
activations in healthy participants and in patients with aphasia but preserved
calculation
- Hogrefe, Weidinger, Ziegler & Goldenberg - Capturing nonverbal information in narrations
of aphasic patients with reduced verbal output: Does limb apraxia make a difference?
- De Jong-Hagelstein,Van de Sandt-Koenderman, Prins, Dippel, Koudstaal & Visch-Brink -
RATS-2: Cognitive-linguistic therapy and communicative therapy equally effective in
early aphasia: a randomized controlled trial.
- Nault, Baayen & Libben - Is all morphology morphology: Evidence from therapeutic
intervention
- Nickels, Best, Williamson, Burmester & Etzien - An examination of sub-types of semantic
errors in word production in aphasia.
13.00-14.00 Lunch
19.00-20.00 Poster session 3 (Bilingualism and Psychosocial aspects of
aphasia)
- Ghafar Samar & Akbari – A language teacher in the maze of bilingual aphasia: A Kurdish-
Persian case
- Akyıldız, Maviş, Olgun & Eylikeder – Public awareness among a sample of Turkish
population about traumatic brain injury
- Yasar, Akyıldız, Bilge & Çoşkun - Spouse Attitudes in Turkey Toward the Person With
8
Aphasia
- Ünal, Düzenli Öztürk, Ölmez, Bal & Kılıç - Determining the Amount of Social-Interactional
Participation in People with Aphasia through CAL
- Johari & Ashayeri - Paradoxical successive recovery of Bilingual Aphasia (Turkish –
Persian) after stroke
- Karimi, Falahi & Parhizgar - A Descriptive study of Aphasia-Apraxia's features'
Correspondence and conformity of a Trilingual Broca's Aphasic Patient (A Case Study)
- Munarriz & Ezeizabarrena - Lexical and post-lexical deficits in bilingual anomia
- Özdemir, Toğram, Alev, Aksoy & Gülpınar - A Retrospective Study: The characteristics of
Patients with Aphasia at DİLKOM
- Toğram, Maviş, Aksu & Yargan - Assessing the Use of Speech Acts in Participants with
Aphasia through ADD
20.00 Dinner
9
Abstracts
10
Rapid Strategic Processing in Word Comprehension with
Aphasia
G. Albyn Davis, Ph.D.1
City University of New York
&
Lynne Alarie-Bibeau, Ph.D.
University of Massachusetts
1 G. Albyn Davis,545 West End Ave. 12E, New York, NY 10024 USA; phone 212-874-5922; e-mail: [email protected]
Recently Wilshire (2008) reviewed certain limitations in the “box-and-arrow” approach to
understanding aphasic impairments and suggested that psycholinguistics can fill in some of
the blanks. One limitation has to do with understanding how a process works and whether a
deficit can be understood with respect to automatic and controlled levels of processing
within each component or “box.”
Our proposal deals with word comprehension. People comprehend everyday words with the
instantaneous activation of a concept stored in a semantic network. This activation spreads
automatically to nearby concepts. Furthermore, processing may continue in a capacity-
limited strategic or controlled fashion. Semantic priming is one procedure that enables us to
distinguish automatic and controlled processing.
Researchers have focused on manipulation of stimulus onset asynchrony (SOA) or the time
between a lexical prime and lexical target in a semantic priming task. Relatively short SOAs
(< 250 ms) detect automatic processing, while longer SOAs (> 250 ms) allow time for
controlled/attentional processing. Clinical researchers have reported that aphasia includes a
deficiency in automatic processing; others have not. This apparent contradiction is difficult
to interpret because of inconsistent and questionable experimental procedures.
When expectations are high for a prime-target relationship, controlled processing is more
likely and is most evident when there is sufficient time for initiating strategies (i.e., longer
SOAs). Semantic relatedness of a prime-target pair, which is knowledge fixed in long-term
memory, influences target recognition with very short SOAs. In a study of Alzheimer’s
dementia, both expectancy and relatedness were manipulated but only at a long SOA of 1250
msec (Bell, et al., 2000).
The purpose of our study was to determine whether chronic mild to moderate aphasia
includes a deficit in controlled processing, compared to automatic processing. Following
11
Neely (1977), we manipulated expectancy through instructions about the nature of prime-
target pairs. We also relied on the use of a neutral prime to detect instruction-induced
facilitation by related and unrelated primes.
Method
Ten individuals with mild to moderate aphasia caused by stroke (AQs 74-99) and ten adult
controls participated in the experiment. The participants with aphasia were chronic (1 to 8
years postonset) and mixed aphasias (7 fluent, 3 nonfluent). Their accuracy with lexical
decision averaged 96 percent, which is important for the standard analysis of response times
only for accurate responses.
In two sessions, participants viewed a total of 576 prime-target word pairs presented on a
computer screen. To detect overall priming, one fourth of the primes were categorically
related to the targets (e.g., BIRD-robin) and one-fourth were categorically unrelated (e.g.,
SPORT-door). To detect facilitative priming, half the primes were baseline neutral (e.g.,
BLANK-bluejay). Nonword targets followed an equal number of primes from each category
condition. To induce expectancy, participants were instructed to expect specific category
exemplar targets following a specified prime (e.g., a type of bird after BIRD or part of a
building after SPORT). The latter is of particular importance to the study, because
expectations could be created for targets with unrelated primes and well as related primes.
This approach detects resistance of automatic processing to conscious strategies. Short and
long SOAs (250 ms and 2000 ms) were distributed equally between each prime condition
within each experiment
Results and conclusions
For the sake of time, pertinent comparisons will be shown with graphs. Because this report
emphasizes strategic expectancy induced by instructions, performance with unrelated
primes is of most interest. The methodology was validated for inducing the expectancy
effect, because the controls were facilitated by unrelated primes only with the 2000 SOA (by
66 ms). The group with aphasia was facilitated similarly by unrelated primes with the long
SOA (by 80 ms). Controlled or strategic processing is more likely with SOAs that are long
enough to allow it to occur, which is not so with short SOAs. Thus, individuals with chronic
mild to moderate aphasia are capable, within two seconds, of using processing strategies.
With respect to the short SOA, the controls showed both overall and facilitative relatedness
priming (68 and 72 ms, respectively) indicative of automatic spreading activation in the
semantic network. Overall priming, in this case, demonstrated that automatic spreading
activation was not influenced by the slower acting expectancy of the unrelated prime.
Controls also reinforced common result patterns with related primes at the long SOA.
12
They had no overall priming (because unrelated primes were expected) and showed
facilitative priming relative to neutral primes (67 ms). The group with aphasia was similar to
the controls, with absent overall priming but the presence of facilitative priming of 93
milliseconds.
Unlike controls with related primes at the short SOA, the group with aphasia had no overall
priming but did demonstrate facilitative priming (53 ms). Thus, for aphasic participants,
overall priming was similar for long and short SOAs. This lack of overall priming at the short
SOA might lead us to conclude that the aphasic group was impaired in automatic spreading
activation. However, this is not the only possibility. The slowness of response allowed for
post-lexical strategic processes to fill the delay and to speed up processing of unrelated
primes (i.e., see comparison with neutral primes). These post-lexical processes are thought
to consist of backward target-prime comparisons, and these comparisons may be modulated
by instructed expectancy. Therefore, when individuals with aphasia have long response
times and no priming effects with short SOAs, they may not necessarily be impaired in
automatic processing.
References
Bell, E.E., Chenery, H.J., & Ingram, J.C.L. (2000). Strategy-based semantic priming in Alzheimer’s
dementia. Aphasiology, 14, 949-965.
Neely, J.H. (1977). Semantic priming and retrieval from lexical memory: Roles of inhibitionless
spreading activation and limited-capacity attention. Journal of Experimental Psychology: General,
106, 226-254.
Wilshire, C.E. (2008). Cognitive neuropsychological approaches to word production in aphasia:
Beyond boxes and arrows. Aphasiology, 22, 1019-1053.
13
A language teacher in the maze of bilingual aphasia:
A Kurdish-Persian case
Reza Ghafar Samar, Ph.D
Mohsen Akbari, M.A.
ELT Department, Tarbiat Modares University, Tehran, Iran
To broaden the scope of bilingual aphasia research, it is unquestionably required to reflect
on it in different contexts (Fabbro, 2001; Abutalebi & Green, 2007). As well, owing to the
scarcity of aphasia-related studies in Iran, the present study aims: (1) to shed light on the
impaired language skills; (2) to identify the recovery pattern of languages; (3) to provide the
main variables affecting the recovery pattern of languages in a bilingual Kurdish-Persian
aphasic speaker; moreover (4) to associate the main findings to language acquisition
theories.
The case recruited in the present study was MP who was born in Ilam (a Kurdish province)
in Western Iran. She was a 40-year-old right-handed bilingual suffering from Broca’s
aphasia following a cerebral hemorrhage 26 months before the present study. She had no
history of other diseases that would affect her communicative ability. She acquired Kurdish
(L1) as her mother tongue from childhood. Regarding Persian (L2), she learned it as a second
language at primary school and also as a formal language used in her bilingual hometown.
An interesting point to mention was that the patient used to teach Persian language and
literature for about 18 years.
To assess the language skills in both languages, with pre-morbid language proficiency and
history of bilingualism; the equivalent versions of the Bilingual Aphasia Test (BAT; Paradis,
1987) in Kurdish (Paradis & Maniyi, 1989) and Persian (Paradis, Paribakht, Nilipour, 1987)
were administered, respectively. Then, the performances were described and compared
qualitatively and quantitatively in detail.
The overall findings revealed preserved comprehension but reduced non-fluent speech in
both languages. The final recovery pattern was non-parallel recovery with L2 (Persian; as a
formal language in Iran) better recovered and less impaired than L1 (Kurdish; as a minority
spoken language in Iran). The non-parallel recovery of languages was associated with several
variables including type of aphasic syndrome, pre-morbid language proficiency; language-
specific features; along with age, context and manner of language acquisition. Besides, the
experience of teaching L2, here teaching Persian language and literature, was incorporated
in this respect to the findings.
14
To put it briefly, the study revealed that a combination of “Critical Variables” (Paradis, 2001)
may clarify the deficits and the way to determine the recovery patterns in bilingual aphasic
speakers. It was concluded that teaching a language as an explicit automatic process
(Kainz,1983; Paradis, 2000) most likely affects the recovery pattern of languages post-troke,
and the consciously learned and taught language is better recovered and less impaired in
bilingual aphasic speakers. Furthermore, the findings were implicated to language
acquisition theories and their association with the recovery pattern of languages in bilingual
aphasic speakers.
References
Abutalebi, J. & Green, D. (2007).Bilingual language production: The neurocognition of language
representation and control. Journal of Neurolinguistics, 20, 242–275.
Fabbro, F. (2001). The bilingual brain: Bilingual aphasia. Brain and language, 79, 201–210.
Kainz, F. (1983). Speech pathology I: Aphasic speech. In M. Paradis (Ed.), Readings on aphasia in
bilinguals and polyglots (pp. 636–640). Montreal: Didier.
Paradis, M. (1987). The assessment of bilingual aphasia. Hillsdale, N. J., Lawrence Erlbaum
Associates Inc.
Paradis, M. (2000). Generalizable outcomes of bilingual aphasia research. Folia Phoniatrica et
Logopaedica, 52, 1, 54–64.
Paradis, M. (2001). Bilingual and polyglot aphasia. In R. S. Berndt (ed.). Handbook of
neuropsychology, pp. 69-91. Amesterdam: Elsevier Science.
Paradis, M. & Maniyi, M. (1989) Bilingual Aphasia Test (Kurdish version).Hillsdale, N.J.: Lawrence
Erlbaum Associates Inc.
Paradis, M., Paribakht, T., & Nilipour, R. (1987). The Bilingual Aphasia Test (Farsi version). Hillsdale,
N.J. : Laurence Erlbaum Associates Inc.
15
Public awareness among a sample of Turkish population about
Traumatic Brain Injury
Didem Akyıldız, İlknur Maviş, Engin Olgun & Seda Eylikeder
Anadolu University, DİLKOM, Turkey
Introduction
Experts estimate that a new case occurs once every 21 seconds in the US, with a prevelance of
almost 5 million people living in the country. Although the prevalence for brain injuries has
not been estimated yet in Turkey, it is expected to be as high as it was mentioned for the
USA. In a previous study, measuring public awareness on aphasia, when participants were
asked whether they have heard of the name and known the consequences of TBI, the
disorder was found to be “unknown” either by name or by consequences (Maviş, 2004).
Despite the frequency with which brain injury occurs have been not yet evident; it is often
reported to be with children and middle age; that are older than 1 and younger than 45 year
old due to car and motor accidents, sports, games, and/or activities.
Sensation, perception and motor related problems affect the quality of life of the individuals
and children with TBI. Besides cognitive issues, memory loss, lack of care, speech and
language problems and psychosocial behavior issues are encountered as a result of traumatic
brain injury. Popular media is rarely interested in this type of physical, cognitive and psycho-
social consequences depending on traumatic brain injury. Research done in the USA often
report that people in general public are not aware of the physical, cognitive, behavioral and
psychosocial ramifications of sustaing brain damages. Swift and Wilson (2001) summed up
complaints of inaccurate and inadequate knowledge common in reference to the general
public and to health-care professionals. General public has inaccurate ideas or
misconceptions about the kinds of problems resulted from brain injury, the recovery and
therapy durations and the capabilities of the people with TBI. Accordingly, the lack of
information cause lay people to believe in inaccurate concepts about the disorder. The aim of
the present study is to investigate what types of misconceptions individuals have about TBI
in general.
Method
In the present study, it is aimed to test awareness of drivers about problems which could be
encountered after the brain injury; at the same time, to determine whether they are aware of
legal and treatment rights of the patient. A Replicated Survey of Traumatic Brain Injury
including 25 questions was administered to the 60 drivers who went through the
16
Psychotechnical Evaluation (Driver Assessment, Training and Research Center) and who
were attending to driving courses in Eskisehir.
With Psychotechnical Evaluation, person’s abilities related to driving are tested through
special tests developed by the traffic psychologists working in this field. The testing method
is called as Psychotechnical Evaluation. Mental and psychomotor abilities of the drivers who
were banned for driving, or caught drunk while driving or who violated speed limits are
being evaluated by those psychological tests.
The inquires within the survey based on 4 step likert type testing to determine on the general
knowledge and awareness of individuals about brain injury resulted from TBI and the
recovery treatment (Hux et al, 2006; Guilmette and Paglia, 2004). The four items relating to
general knowledge about brain injury, three items relating to coma and unconsciousness,
four items relating to memory deficits and six items relating to brain injury recovery
comprised the modified questionnaire. The questionnaire also included questions about each
respondent’s age, employment, level of education and acquaintance with someone who had
experienced brain injury. According to the results, it will be discussed if the current
knowledge of driver candidates or drivers matches with the misconceptions determined by
Swift & Wilson (2001) about TBI information.
References
Maviş İ. Perspectives on public awareness of stroke and aphasia among Turkish patients in a
neurology unit. Clinical Linguistics & Phonetics; volume 21 number 1 January 2007 pp 55
Hux K, Goeken T, Schram C.D; Misconceptions about brain injury; Brain Injury, May 2006, 20(5):
547-553
Guilmette J.T, Paglia F.M, The public’s misconceptions about traumatic brain injury; Archives of
Clinical Neuropsychology 19 (2004) 183-189
Swift T.L. , Wilson S.L., 2001. Misconceptions about brain injury among the general public and non-
expert health professionals: An exploratory study. Brain Injury 15, pp. 149–165.
17
Spouse Attitudes in Turkey Toward the Person With Aphasia
Özlem Cangökçe Yaşar, Didem Akyıldız, Ayça Bilge, Fulya Çoşkun
Anadolu University, DİLKOM, Turkey
Introduction
Many families of aphasic patients have never been ready to cope with their new situation,
and the resulting stress may negatively influence, particularly, spouse attitudes toward the
person with aphasia (Zraick and Boone, 1991). Most research investigating partners’
thoughts for the aphasic one started very early in 1960s. For example, Buxbaum (1967)
examined the role of care in shaping wives’ views of their aphasic spouse. She found that
wives with high nurturance ratings viewed their husbands as less impaired, expressed
greater marital satisfaction, shared in activities more often, and expressed less negative
feelings toward their spouses than wives with low nurturance ratings. Whitehall-Smith
(1977) found that when spouses were encouraged to acknowledge their grief and anger, their
attitudes were more positive and their physical condition improved. However, Henry (1984)
found that spouses reported significantly more negative emotional behaviors in the aphasic
patient.
Traditionally, speech-language pathologists have used clinical interviews to monitor spouse
attitudes toward their patients with aphasia. Psychologists and other mental health
practitioners frequently used a classic attitudinal assessment technique called Q-
methodology (Stephenson, 1953) which rely on words to describe the person’s attitudes.
Accordingly, the purpose of this study was to use Q-methodology in Turkish to determine the
attitudes of spouses toward the person with aphasia and to compare those of spouses of
normal adults.
Method
To examine the attitudes of individuals toward their aphasic spouses, Zraick and Boone’s
(1991) study was replicated. A checklist including 70 descriptive adjectives were used. Thirty
spouses of aphasic patients who met the study criteria were selected from the Osmangazi
University Medical Faculty and Education, Research and Training Center for Speech and
Language Pathology (DİLKOM) and at local hospitals in Eskişehir. After the spouse was
given the 70 index cards one at a time, s/he was asked to sort out adjectives cards in 5
groups; namely, every time-usually-indecisive-sometimes-never. According to the results
provided from research, it will be discussed to determine the attitudes of spouses toward the
person with aphasia and to compare those of 30 spouses of normal adults matched both in
18
age and education with the other group. Significant differences were expected in attitudes
among those spouse groups. The attitudes most frequently expressed by spouses of aphasic
patients were compared to the ones expressed by spouses of normal individuals. Speech and
language pathologists were encouraged to serve to increase patients’ awareness of their
behaviors and to foster a more realistic expectation of patient behavior on the part of the
spouse.
References
Zraick R.I., Boone D.R. (1991). Spouse attitudes toward the person with aphasia. Journal of Speech
and Hearing Research, volume 34, 123-128
Buxbaum J. (1980). Effect of nurturance on wives’ appraisals of their marital satisfaction and the
degree of their husband’ aphasia. Journal of Consulting Psychology, 31,240-243
Whitehall-Smith C. (1977). Using transactional analysis with women whose husbands have had
strokes. Transactional Analysis Journal, 7,55-57.
Henry G.K. (1984). Emotional behavior following stroke: patient versus spouse reports. Clinical
Gerontologist, 2,70-72.
Stephenson, W. (1953). The study of behavior. Chicago: University of Chicago Press.
19
The comprehension of time reference in agrammatic aphasia:
A comparison of languages with Tense and languages with
Aspectual adverbs
Harwintha Y. Anjarningsih, Elif Bamyaci, Chien-Ju Hsu, Roelien Bastiaanse
Center for Language and Cognition Groningen
Introduction
In Indo-European languages, verb inflections specify, among other things, the time-frame in
which the event takes place (past, present, future) and both production and comprehension
of these verb inflections are difficult for agrammatic speakers. These problems are not
restricted to finite verbs as participles are impaired as well. Past verb forms are affected
more than present (for Dutch production: Bastiaanse, 2008; Dutch comprehension: Jonkers
and de Bruin, in press) and future (for Turkish Yarbay-Duman and Bastiaanse, 2009) verb
forms.
The present cross-linguistic study seeks to investigate and compare comprehension of time
reference by three groups of speakers with agrammatic aphasia: a Turkish, a Chinese, and an
Indonesian group. Of interest is that these three languages have different grammatical
systems in expressing time reference. Turkish, a language that uses verb inflections to refer
to time (Tense and Aspect), and Chinese and Indonesian, two languages that use aspectual
adverbs to refer to past, present, and future. In Turkish, time reference, both Tense and
Aspect, is denoted by inflectional suffixes on finite verbs (see example 1).
(1) adam mektup okuyor okudu
okuyacak.
the man a letter [is reading] read [will read]
In Chinese, time reference is denoted by aspectual adverbs. The verbs in Chinese are not
inflected (see example 2.)
(2) zhe ge ren zai du du le yao du yi fong sin
the man [is reading] read [will read] a letter
The third language, Indonesian, shows time reference by lexical adverbs (3) or free-standing
aspectual adverbs (4). Lexical adverbs are sentential adverbs that in the current study are put
at the beginning of sentences. Aspectual adverbs are a part of the Verb Phrase and always
20
come before the verbs that are only inflected for transitivity. Verbs in Indonesian are not
inflected for tense, agreement, and mood.
(3) baru saja sekarang sebentar lagi dia membaca sebuah surat
[just now] now [in a moment] he/she read a letter
(4) laki-laki itu sudah sedang akan membaca sebuah surat
man the [past] [present] [future] read a letter
Aim
The current study compares the comprehension of past, present, and future markers. The
questions were:
(1) is comprehension of grammatical time-reference is impaired in agrammatic aphasia?
(2) is time-reference through aspectual adverbs (in Chinese and Indonesian) equally
impaired as time-reference through verb inflection (in Turkish)?
(3) there is a difference between reference to past, present, and future as found earlier for
production?
(4) is there a difference between comprehension time reference through lexical adverbs and
aspectual adverbs (in Indonesian)?
Methods
The comprehension task of the Test for Assessment of Reference of Time (TART: Bastiaanse,
Jonkers & Thompson, unpubl.) was used. This test is presently used for 15 languages,
including the three languages of this study. Photographs of 20 transitive verbs representing
(1) actual performance of the action (present), the completed action (past), and the intention
to perform the action (future) were used. The patients were presented with two pictures of
the same verb in two time frames and had to match one of them to a spoken sentence.
Pictures with targets referring to the past are never presented on the same page with pictures
referring to the future because both pictures show “no-action”, which and has been proven to
be difficult to differentiate for normal, healthy subjects. For example, the patients were
presented with two pictures, the first of which depicts “The man is eating an apple” and the
second “The man will eat an apple”. The experimenter asked the patients to point to “The
man is eating an apple” and they had to indicate their choice. For Chinese and Turkish, 9
agrammatic patients and 10 healthy control subjects were examined with the test. For
Indonesian, data collection is still in progress, therefore we will here only discuss the Chinese
and Turkish data.
21
Results
The control subjects scored at ceiling in both languages and their data will be ignored in
further analysis. The results of the agrammatic patients are shown in Figure 1.
Figure 1: Results of the Turkish and Chinese agrammatic patients on the test
for comprehension of grammatical time reference (number of correct answers,
max. = 20).
T-tests were used to analyse the data. Overall performance of the Turkish and Chinese
agrammatic patients on the test was similar (t(16)=-0.074, p=0.94). In both languages,
performance on ‘present’ was significantly better than on both ‘past’ (Turkish: t(8)=-2.713;
p=0.029; Chinese: t(8)=5.813, p=0.001) and ‘future’ (Turkish: t(8)=3.591, p=0.010;
Chinese: t(8)=2.475, p=0.041). ‘Past’ and ‘future’ were not significantly different in both
languages, although there is a tendency for past to be more difficult than future in Chinese
(Turkish: t(8)=-0.845, p=0.424; Chinese: t(8)=-2.224; p=0.06).
Discussion
The results show that (1) comprehension of grammatical morphemes for time-reference in
agrammatic aphasia is impaired; (2) independent on whether this is done through inflection
or aspectual adverbs; (3) comprehension of reference to present is better preserved than to
past and future. To answer the 4th research question, the data from Indonesian will be used.
These findings pair with the findings for agrammatic production, that showed that reference
to the present is relatively well-preserved. This suggests that there is a central problem in
agrammatic aphasia to encode semantic information on time-reference into grammatical
22
morphemes and the other way around, particular information that refers to the past and to
the future, at least in comprehension.
If this is indeed what is happening in processing of time reference by agrammatic patients,
the same patterns of findings are expected to be found in Indonesian.
References
Bastiaanse, R. (2008) Production of verbs in base position by Dutch agrammatic speakers: Inflection
versus finiteness. Journal of Neurolinguistics, 21, 104-119.
Jonkers, R. & Bruin, A. de (in press) Tense processing in Broca’s and Wernicke’s aphasia.
Aphasiology.
Yarbay Duman, T. & Bastiaanse, R. (2009). Time Reference Through Verb Inflection in Turkish
Agrammatic Aphasia. Brain and Language, 108, 30-39.
23
Time reference in Turkish agrammatic speakers
Elif Bamyaci1 & Roelien Bastiaanse2
1European Master’s in Clinical Linguistics (EMCL)
2Center for Language and Cognition Groningen (CLCG)
Introduction
Agrammatic speech is characterized by problems with bound and free-standing grammatical
morphemes. Verb inflections seem to be particularly vulnerable, although not each verb
inflection morpheme is impaired to an equal extent. It has been shown that Tense features
are particularly prone to errors (Friedmann & Grodzinsky, 1997; Wenzlaff & Clahsen, 2004;
2005). Within the Tense domain reference to the past seems to be more impaired than
reference to the present (Bastiaanse, 2008 for Dutch; Stavrakaki & Kouvava, 2003 for Greek;
Simonsen & Lind, 2002 for Norwegian) and than future (Yarbay Duman & Bastiaanse, 2009
for Turkish). These selective problems with past are not restricted to Tensed verbs.
Bastiaanse (2009) showed that for Dutch non-tensed verb forms referring to the past (by
participle constructions) are also more impaired than non-tensed infinitives (in present
continuous constructions).
The present study focuses on Tense in Turkish agrammatic speech. Turkish is an
agglutinative language with a rich verb paradigm. Present continuous, past and future can all
be expressed through Tense, so no auxiliaries are used. Interestingly, Turkish has two forms
of past Tense, so-called ‘seen past’ and ‘heard past’. Seen-past Tense is used when the
speaker himself witnessed the action; heard-past Tense is used when the speaker expresses
information from a source other than himself.
The aims of this study are two-fold: (1) to evaluate whether in Turkish, with the rich verb
inflection paradigm present Tense is better preserved than past Tense, like in Dutch, Greek
and Norwegian; (2) to see whether there is a difference between heard and seen past Tense.
Methods
The Turkish version of the Test for Assessment of Reference of Time (TART; Bastiaanse,
Jonkers & Thompson, unpublished) was used to test 7 agrammatic (4 male; age range 39-74)
and 7 non-brain-damaged Turkish speakers, matched for age and education. The
agrammatic speakers were aphasic due to a single left-hemisphere stroke. A ‘sentence
completion with prompting’ paradigm was used to elicit the intended verb form. Each item
consisted of a pair of photographs of two actions (either both completed or both ongoing)
24
with the verb printed above. The prompting sentence contained a temporal adverb and the
required verb form. An example of a picture pair of the Turkish (with the English text) is:
For this picture, you could say “Now the man is reading a letter”; for this picture, you could
say “Now the man….” and the patient was supposed to finish the sentence with “… is writing
a letter”.
There were three conditions: present continuous Tense (from now on ‘present Tense’), seen
past Tense and heard past Tense. Each condition had 20 items. For the items used, heard
past and seen past were equally plausible, according to native speakers of Turkish.
Results
The non-brain-damaged subjects performed perfectly in all three conditions. In Figure 1 the
results for the agrammatic speakers are graphically represented. For statistical comparison,
Fisher’s exact tests were used. Present Tense is significantly better than both past Tenses
(p=0.0001 in both comparisons). Seen past tense is more difficult than heard past Tense
(p=0.0012).
Figure 1: Performance of the Turkish agrammatic speakers (max.=20)
25
Discussion
These data confirm the findings in earlier group and case studies to Dutch, Greek, and
Norwegian, and Turkish: reference to the past through verb inflection is selectively impaired
in agrammatic aphasia, when tested with an oral production task. Interestingly, within
reference to the past, those forms that refer to events that have been witnessed by the
speaker are even more impaired than those that the speaker only heard of or read about. A
theory on a purely syntactic or morphological basis cannot account for these data: expressing
the notion ‘time’ in verb morphology, with present being least impaired. The truth value of a
proposition describing the event in the here and now is selfevident. This might the reason
why marking a verb for such an event is relatively easy. For describing a past event,
establishing the truth value is not self-evident, which might make verb inflection for the past
more difficult. ‘Heard past’ needs to be checked with someone else’s or general knowledge
that is usually taken for granted; there is no personal involvement. However, in case of ‘seen
past’ the speaker has to think about the truth value and check it against his memory. This
personal involvement apparently makes coding for such an event more difficult for Turkish
agrammatic speakers.
References
Bastiaanse, R. (2008) Production of verbs in base position by Dutch agrammatic speakers: Inflection
versus finiteness. Journal of Neurolinguistics 21, 104–119.
Bastiaanse, R., Bouma G. & Post W. (2009) Linguistic complexity and frequency in agrammatic
speech production, Brain & Language 109 (2009) 18–28.
Friedmann, N. & Grodzinsky, Y. (1997). Tense and agreement in Agrammatic production: Pruning the
syntactic tree. Brain and Language, 56, 397–425.
Simonsen, H.G. & M. Lind 2002: Past tense expression in a Norwegian man with Broca’s aphasia.
Windsor, F., M.L. Kelly & N. Hewlett (eds.): Investigations in clinical phonetics and linguistics.
Mahwah, N.J. & London: Lawrence Erlbaum Associates, 45-56.
Stavrakakia, S., Kouvava, S., (2003). Functional categories in agrammatism: Evidence from Greek.
Brain and Language 86 (2003) 129–141.
Wenzlaff, M., & Clahsen, H. (2004). Tense and agreement in German agrammatism. Brain and
Language, 89, 57–68.
Wenzlaff, M., & Clahsen, H. (2005). Finiteness and Verb-Second in German Agrammatism. Brain and
Language, 92, 33–44.
Yarbay Duman, T. & Bastiaanse, R. (2009). Time reference through verb inflection in Turkish
agrammatic aphasia. Brain and Language, 108, 30-39.
26
Language and calculation: Brain activations in healthy
participants and in patients with aphasia but preserved
calculation
Yael Benn1, Ying Zheng, Iain D. Wilkinson, Roi Cohen Kadosh, Michael Siegal,
Rosemary Varley
1Yael Benn, Department of Psychology, The University of Sheffield, Western Bank Sheffield, S10 2TN UK; Phone: 44 114 22 26501; fax: 44 114 27 66515; e-mail: [email protected]
Research question
A fundamental and enduring issue in research in cognitive neuroscience concerns the extent
to which language influences cognition. In this respect, one major question involves the
relationship between language and exact mathematical calculation. Although functional
imaging studies of healthy participants have indicated that calculation recruits left
hemisphere perisylvian language mechanisms (Dehaene et al., 1999), neuropsychological
investigations have revealed that calculation can be retained despite severe aphasia and
extensive damage to left perisylvian networks (Varley et al., 2005). This pattern of results
suggests that in healthy adults the resources of the language faculty may be recruited as a
‘co-oted’ system that scaffolds and facilitates higher order cognition, but that the potential
for retained calculation despite severe language impairment indicates that these resources
are not necessary for calculation. In an fMRI study, we explored the neural basis of
calculation in healthy controls and individuals with severe aphasia but preserved calculation
ability. The investigation examined calculation across two sets of stimuli: problems
expressed in a familiar number format and problems in a novel shape format. The purpose of
the three experiments reported here was to identify a core calculation system across problem
formats and in both healthy participants and those with severe aphasia.
Design
Experiment 1 examined performance and behavioural patterns of young adults on addition
and subtraction problems expressed in two formats: (1) a conventional digit symbol set that
can be easily encoded into language; (2) a novel shape representation of the problem which
could not be easily encoded into language.
Experiment 2 was an fMRI study of activation across these conditions. A control task
involving recognition of shapes and numbers was developed to permit subtraction of visuo-
perceptual and response selection processes from the calculation conditions. Using a 3T
Scanner (T2 weighted images, TR= 3000ms), our aim was to explore the neural basis of
calculation in healthy, right-handed participants who were age-matched to the aphasic
27
patient sample. We sought to determine the degree to which the different types of calculation
are sub-served by a network that is distinct from traditional language areas.
Experiment 3 examined the neural correlates of calculation in two patients with severe
aphasia but residual ability to calculate using a procedure identical to that of Experiment 2.
Results
In Experiment 1, addition and subtraction evoked different behavioural patterns in both
calculation with shapes and numbers. Furthermore, extensive math training had no effect on
calculation time in either shapes or numbers, but it had a significant effect on error rate,
participants with extensive math training made significantly fewer errors when calculating
with numbers.
In the case of calculation in the healthy sample, imaging results from Experiment 2 indicated
a bilateral superior parietal (BA7) activation distinct from traditional language areas that
was common across addition and subtraction operations in both symbol sets. Differences in
brain activity emerged between the number addition and number subtraction tasks.
Although addition occupied predominantly frontal and temporal areas, particularly in the
left hemisphere, subtraction showed a more bilateral activation pattern extending to the
bilateral parietal areas.
In individuals with severe aphasia who retained calculation ability, fMRI results from
Experiment 3 also revealed bilateral parietal area activation associated with the calculation
of both numbers and novel shapes. There was no evidence that calculation was associated
with recruitment of strategic resources based on frontal or prefrontal cortex activation.
Discussion
Functional imaging studies of patients with neurological lesions represent an important
method for exploring and interpreting the results of imaging studies of healthy participants.
Through use of patient samples, it is possible to explore whether the activations observed in
the healthy case are necessary for a particular cognitive performance, or represent optional
‘scaffolding’ functions based on a ‘co-opted’ system. The resources of the language faculty are
a prime candidate for supporting other intrinsically non-linguistic performances such as
calculation. Nameable elements of a problem can be rapidly encoded into words and
intermediate stages of a calculation or a final product can be maintained in verbal working
memory prior to responding. Similarly, in acquiring a capacity for calculation, verbally
presented information such as multiplication tables can be learned by rote in a linguistic
format. This stored linguistic information can then be activated and use to reduce demand
for online computation.
In healthy participants, increased activity increased activity in perisylvian areas, such as
Broca’s area (BA 44/45) was observed in number addition compared to number subtraction,
28
indicating that language involvement differs for calculation operations – an involvement that
is consistent with the role of verbally rehearsed facts that occurs more often with addition.
However, shared areas of activity in bilateral parietal cortex between the shape and the
number calculation tasks point to a core non-linguistic area that is necessary for calculation.
In the aphasic case, calculation can be preserved if lesions do not extend to these parietal
areas. Activations observed in the aphasic calculators showed overlap with those of healthy
calculators, and do not suggest that preserved calculation in these cases is due to wide-scale
functional reorganisation. Instead parietal and mental visual manipulation areas that are
intact and which form the core calculation system are sufficient to support simple
calculation. This finding further indicates that the perisylvian activity associated with
calculation represents a scaffolding mechanism and is not a component of the core
mechanism. Although calculation difficulties can co-occur with aphasia, it appears that this
association of deficits is not due to necessary involvement of language in exact calculation,
but rather may be the consequence of the lesions that cause aphasia extending to the left
inferior parietal area or interrupting fibres running between the left and right calculation
zones.
References
Dehaene,S., Spelke, E., Pinel, P., Stanescu, R., & Tsivkin, S. (1999). Sources of mathematical thinking:
Behavioral and brain-imaging evidence. Science, 284(5416), 970-974
Varley R. A., Klessinger N. J. C., Romanowski C. A. J., & Siegal M. (2005). Agrammatic but Numerate.
PNAS 102 (9): 3519-3524
29
Inefficient Fluent and Proficient Non-fluent:
A dissociation between micro- and macrostructure of aphasic
discourse
Mira Bergelson (1), Olga Dragoy (2)
(1) Moscow State University (Russia), (2) University of Groningen (the
Netherlands), Center for Speech Pathology and Neurorehabilitation (Russia)
It has been repeatedly shown that expressive speech is far more impaired in particular
aphasia types, which gives grounds for distinguishing between non-fluent and fluent aphasic
speakers (Grodzinsky, 1990). But difficulties at the word and sentence level do not exclude a
sensible story and a communicative success. The reverse may be true as well: spared
grammar and lexis do not assure coherent speech to the point. The question of how strengths
and weaknesses at the local structure level relate to the success at the discourse level
presents a great challenge for aphasic speech research (Armstrong, 2000). The present study
focuses therefore on discourse analysis in fluent and non-fluent aphasic speakers and
investigates the relationship between micro- and macrostructure of their speech.
The first goal was to analyze aphasic discourse from two different perspectives. The focus on
microstructure allowed to evaluate the degree of lexical and grammatical damage. The
macrostructural perspective, however, dealt with global discourse coherence and focused on
how aphasic speakers meet the requirements of a specific discourse genre, namely a story.
The relationship between breakdown or success at those two levels (micro- and
macrostructural) was further explored. The second goal of the study was to compare
discourse elicited from non-fluent and fluent aphasic speakers. These two aphasia types
demonstrate significant dissociation on the microstructural level (Caplan, 1987). It is
unclear, however, how fluent and non-fluent speech samples differ from the point of view of
their global structure and its interaction with local characteristics.
The discourse of four aphasic speakers was analyzed. Two participants (both females, mean
age 41 y.o, 21 months post onset) were non-fluent speakers and two (a male and a female,
mean age 43 y.o, 13 months post onset) were fluent. Two photo pictures were developed for
discourse elicitation, one for the training session, another for the experimental one. Both
pictures were complex rather than compound (Luria, 1966) and represented a complication
of an action with its obvious after-effects. Such pictures are assumed to cause viewers to infer
what happened before, and what may happen afterwards, thus, producing a story. Besides,
an instruction that explicitly requested temporal sequencing was used to elicit a narrative
rather than a descriptive discourse. The average duration of the experimental session was 3.5
minutes in the fluent group and 2 minutes in the non-fluent one. Responses were audio
30
recorded and orthographically transcribed. The initial transcription was checked and
reconciled through agreement between two transcriptionists.
Microlinguistic abilities of aphasic speakers were addressed through the following measures:
number of utterances, mean number of clauses per utterance, proportion of agrammatic
clauses, mean length of utterance, proportion or nouns, pronouns and predicates. As
compared to the fluent speakers, the non-fluent group produced fewer utterances (8.5 vs.
20.5 on average), the complexity and length of their utterances were more reduced (1.3 vs.
2.1 clauses per utterance, 7 vs. 20 syllables per utterance), and 53% of their clauses were
agrammatic (vs. 0% in the fluent speakers). As for the proportion of word classes, there was
no consistent pattern found in either group for nouns and verbs, but non-fluent speakers
used fewer pronouns than fluent speakers (3.5% vs. 13.5% of all words produced). As
predicted, the discourse microstructure was more impaired in non-fluent speakers: their
utterances were fewer, shorter, less accurate, and contained few pronouns that are the means
of local coherence.
Macrolinguistic measures included story component scheme and its variables, such as
Storyworld and Non-StoryWorld clauses, the former being divided into Main Line Story
Event clauses (semantically non-iterative, non-habitual and temporarily bounded), and
Durative-Descriptive clauses (states of affairs which persist over some interval of time in the
discourse rather than occurring at one discrete instant in the discourse world). Non-Story
World clauses refer to Evaluations (linguistic means that make it clear to the listeners which
circumstances and events are crucial for the point being made (Polanyi, 1989)) and verbal
interaction between the storyteller and the interviewer. Evaluations (including Coda) provide
more insight as regards the goals of this study, while verbal interaction clauses reflect
(inter)personal dynamics.
The component scheme analysis showed that non-fluent speakers’ discourse included, on
average, 72% Main Line clauses and 14% Descriptive clauses, while fluent speakers used a
comparable number of Storyworld clauses of the two types – 41% and 39% correspondingly.
As for the Non-StoryWorld evaluations, those clauses were as rare as Descriptive clauses in
the non-fluent discourse (14%), but the proportion of evaluations in the fluent discourse
equaled to the proportion of Storyworld clauses (42%). Thus, non-fluent speakers mainly
used a narrative strategy, and non-fluent speakers alternated their story telling with
descriptive and evaluative digressions.
The results support the double dissociation between micro- and macrostructure of the fluent
and non-fluent discourse elicited in the story telling task. Microlinguistic breakdown was
shown to be more prominent in the non-fluent group, but despite (or even due to) the
limited verbal capacities those speakers produced a coherent story to the point. The
procedure and materials were specifically developed to elicit a narrative discourse, and non-
fluent aphasic speakers successfully performed the task. On the contrary, the microlinguistic
31
level was relatively spared in the fluent group, but they constantly shifted from story telling
to picture description and evaluations, which made their discourse incoherent. Taking into
account the underlying deficit in the two aphasia types, different strategies employed by
fluent and non-fluent speakers might be suggested. Severe deficit at the sentence level forces
non-fluent speakers to focus on the core components of the story and produce therefore a
compact narrative discourse. On the other hand, the increased decay rate might extend
beyond a series of acoustic stimuli to a series of events in fluent aphasic speakers, which
could explain their easy shift from the story telling to describing the visually presented
material or emotional digressions.
References
Armstrong, E. (2000). Aphasic discourse analysis: The story so far. Aphasiology, 14, 875-892.
Caplan, D. (1987). Neurolinguistics and Linguistic Aphasiology: An Intoduction. Cambridge:
Cambridge University Press.
Grodzinsky, Y. (1990). Theoretical Perspectives on Language Deficits. Cambridge, Mass: MIT Press.
Luria, A. R. (1966). Higher Cortical Functions in Man. New York: Basic Books.
Polanyi, L. (1989). Telling the American Story. A Structral and Cultural Analysis of Conversational
Storytelling. Cambridge, Mass: MIT Press.
32
The production of novel and lexicalized compounds in aphasia
Susanne R. Borgwaldt1 & Arpita Bose2
1 University of Braunschweig, Braunschweig, Germany
2 University of Windsor, Windsor, Canada
Introduction
Compound words, concatenations of two lexical items with a more or less predictable
idiosyncratic meaning, can enrich languages’ vocabulary with lexicalized words such as
rainbow or butterfly, as well with novel compounds, ad-hoc coinages, such as soccer mom,
or cricket mom. Both compound types are effortlessly understood by unimpaired adult
speakers, but might cause specific problems for impaired or developing populations.
To date, research in aphasia has mainly focused on lexicalized compounds, supporting the
idea of awareness of underlying morphological structures, compound decomposition into
components and simultaneous componential access, based on component omission and
substitution patterns. For lexicalized compounds, processing disadvantages have been
reported in aphasia; performance on novel compound processing, however, has generally
been assumed to be relatively unimpaired, based on aphasics’ ability to coin morphologically
well-formed compound neologisms (Badecker, 2001, Blanken 2000, Delazer & Semenza,
1998).
Using hybrid object stimuli in a picture naming task that traditionally had been employed to
assess novel compound processing in first language acquisition, Borgwaldt and Bose (2008)
presented data from three fluent aphasics that displayed severely impaired novel compound
production with a strong preference for insertions, such as “and” between the compound
components.
Extending the scope of experimental investigations into compound processing in aphasia to
novel compounds might help further refine the position of compounds in languages’ twilight
zone between grammar and lexicon and possibly discover different performance patterns
between aphasia types that might reveal selective impairments. This study investigates the
production of lexicalized and novel compounds in fluent and non-fluent aphasic speakers.
1 Corresponding author’s email: [email protected]
33
Study
Method
Participants
Participants were five monolingual English-speaking moderately aphasic individuals, who
were between 42 and 78 years old. They were at least 1 year post-onset to their neurological
disorders and had no other significant sensory and/or cognitive deficits that interfered with
their performance in the study.
Tasks and procedure
Two picture naming tasks examined the oral production of lexicalized and novel compounds:
In the lexicalized compound production task the stimuli were 50 colour photographs of
objects corresponding to lexicalized compound nouns (e.g. rainbow, butterfly), presented
along with 70 filler items, i.e., photographs of objects corresponding to monomorphemic
words. Participants had to name the objects.
The stimuli for the novel compounds were 50 digitally manipulated photographs. They
depicted hybrid objects composed of two roughly equally salient components, e.g., a tent in
the shape of an apple (apple tent) or an object that was part snake, part banana (banana
snake). Participants had to name the hybrid objects as novel compounds.
The two groups of novel and lexicalized compounds were matched for syllable length.
Scoring
Naming responses were scored for morphological well-formedness: either as
morphologically correct (e.g., rainbow, apple tent), or as one of two types of errors, (a)
insertions such as “and” or “he is” between compound components (e.g., rain and bow,
apple he is a tent), or (b) other, such as descriptions, component omissions, or unrelated
errors.
Results and Discussion
As shown in Table 1, in the lexicalized compound production task all five aphasic
participants could produce morphologically correct compound structures and did not show a
particular preference for insertions between the compound components.
In the novel compound production task, performance differed between fluent and non-fluent
aphasia types. The three participants with fluent aphasia displayed severely impaired novel
compound production with a strong preference for insertions, such as “and” between the
compound components (data from Borgwaldt and Bose, 2008). This resulted in productions
that disobeyed English compound structures and rather corresponded to coordinate noun
phrases.
34
Table 1. Results for novel and lexicalized compounds production tasks.
The two participants with non-fluent Broca’s aphasia performed much better in the novel
compound production task and showed no particular preference for insertions between the
compound components.
Conclusion
The above results challenge the assumption that novel compound production might be intact
across aphasia types. The fluent aphasics showed a curious contrast between
morphologically well-formed compound neologisms and substitution errors, such as
redberry for strawberry, bunny pear for piggy bank, or wheelmarrow for windmill (errors,
that had been recorded in the lexicalized compound production task), and their failure to
produce novel compounds in an explicit word-formation task.
The striking difference between novel and lexicalized compound production performance
observed in fluent aphasia might ultimately point to dissimilar underlying compound
representations at the componential level for lexicalized and novel compounds, and
subsequent different processing mechanisms that in turn might be selectively impaired
across aphasia types.
Acknowledgments
This research was supported by a German Academic Exchange Service Research Grant to S.
Borgwaldt and A. Bose.
Lexicalized Compounds
(n = 50)
Novel Compounds
(n = 50)
Errors Errors Participants
Aphasia
Type Morph.
Correct Insertions Other
Morph.
Correct Insertions Other
1 23 (46%) 1 (2%) 26 (52%) 1 (2%) 39 (78%) 10 (20%)
2
Anomia
13 (26%) 0 (0%) 37 (74%) 4 (8%) 35 (70%) 11 (22%)
3
fluen
t
Wernicke 4 (8%) 0 (0%) 46 (92%) 2 (4%) 35 (70%) 13 (26%)
4 13 (10%) 0 (0%) 37 (74%) 32 (64%) 0 (0%) 18 (36%)
5
no
n-f
luen
t
Broca
4 (8%) 0 (0%) 46 (92%) 33 (66%) 3 (6%) 14 (28%)
35
References
Badecker, W. (2001). Lexical Composition and the Production of Compounds: Evidence from Errors
in Naming. Language and Cognitive Processes 16, 337 – 366.
Blanken, G. (2000). The Production of Nominal Compounds in Aphasia. Brain and Language 74, 84
– 102.
Borgwaldt, S., & Bose, A. (2008). Novel Compound Production and Comprehension in Aphasia.
Poster presentation at the 46th Annual Meeting of the Academy of Aphasia (AoA), Turku,
Finland, October.
Delazer, M., & Semenza, C. (1998). The Processing of Compound Words: A Study in Aphasia. Brain
and Language 61, 54 – 62.
36
Verb retrieval in anomic aphasia: effects of instrumentality
and transitivity
Olga Dragoy (1,2), Ekaterina Iskra (2), Elena Mannova (2), Olga Kuzmicheva (2),
Anna Masalova (2)
(1) University of Groningen, the Netherlands, (2) Center for Speech Pathology and
Neurorehabilitation, Russia
Verb retrieval has been repeatedly shown to be influenced by semantic aspects of verbs in
individuals with anomic aphasia. In general, semantic complexity (the number of semantic
components) of a verb is hypothesized to enhance its production. The present study focuses
on testing two factors that might have an effect on action naming in Russian aphasic
speakers with anomia: instrumentality and transitivity of a verb.
Specific verb retrieval problems have been shown to occur not only in agrammatic, but also
in anomic speakers (Bastiaanse, 1991). However, verbs are not equally impaired in anomia.
The effect of semantic complexity was demonstrated on the material of Dutch (Jonkers &
Bastiaanse, 2007), English (Breedin et al., 1998), and Greek (Kambanaros & van
Steenbrugge, 2006): semantically complex verbs are retrieved better than simpler verbs. For
example, verbs with more specific meaning (to run) present less challenge than core
predicates (to go), and instrumental verbs (to cut) are easier to produce than non-
instrumental (to tear). The explanation relates the effect to the number of semantic
components that positively influence lemma activation. It is easier to retrieve verbs with
more components incorporated in their conceptual representation.
The influence of semantic complexity on verb retrieval in Russian speakers with anomic
aphasia was tested in the present study. The first aim was to confirm the effect of
instrumentality found in other languages. An instrument being an extra part of the
conceptual representation of a verb is assumed to increase the activation of the verb lemma.
However, action semantics incorporate not only the information about the action as such
and the manner it is performed. Each action implies a particular set of participants involved
in its realization. This is closely related to such a syntactic characteristic of a verb as
transitivity. It has been argued that transitivity plays no role in anomic production, because
anomic speakers do not have syntactic difficulties (Jonkers, 1998). But the presence of an
object is determined by the semantics of a verb, this is why in the present study transitivity is
regarded as a semantic characteristic of an action. We hypothesized that not only the
instrument, but also the object of an action incorporated into the conceptual representation
of a verb may add to the lemma activation. Thus, the second aim of the study was to retest
the effect of transitivity on verb retrieval in anomic speakers.
37
An action naming task was administered to test participants’ ability to retrieve four groups of
verbs: intransitive non-instrumental (e.g. to crawl), intransitive instrumental (to shoot),
transitive non-instrumental (to kiss), and transitive instrumental (to iron). In the
preliminary test 20 healthy Russian speakers were asked to name 64 actions, and 36 verbs
with 90% name agreement were chosen for the experiment, 9 verbs per condition. Verbs
were balanced between groups by frequency, so that verbs of comparable frequencies were
included in each group. 13 aphasic individuals with anomia due to temporal left hemisphere
lesions participated in the study (6 males, 7 females; mean age 41 y.o.; 16 months post onset,
on average). A black-and-white picture was shown to the participant, and he/she was asked
to answer the question “What is the subject doing?” in a word or two.
A mixed-effect modeling with crossed random effects for subjects and items was conducted
to test if the two mentioned factors (instrumentality and transitivity) influence verb retrieval
in anomic speakers. The analysis revealed that transitive verbs were retrieved significantly
better than intransitive verbs (z = –3.1, p = .001). However, the trend to name instrumental
actions than non-instrumental did not achieve statistical significance (z = –1.5, p = .13). The
post-hoc analysis of individual responses revealed that instrumental actions were not named
better than non-instrumental only in four participants. After those participants were
excluded from the analysis the effect of instrumentality became significant (z = –2.68, p =
.01). On the other hand, in the group of the four excluded participants there was no
instrumentality effect found (z = 0.6, p = .54). The further study of case histories showed
that the four excluded participants were remarkable from the point of view of their speech
deficit. Although their primary problems concerned comprehension and word retrieval,
either the elements of agrammatism occurred in their speech or the motor aphasia was
present in anamnesis, which is usually associated with a frontal lobe dysfunction and Broca’s
aphasia.
The data are support for the semantic complexity hypothesis: extra components of
conceptual representation of a verb enhance verb retrieval in aphasic speakers, whose
expressive speech deficit are restricted to anomia and do not involve agrammatism or motor
problems. In particular, the previously found instrumentality effect was proved once again
using a new material. An instrument as a necessary part of an action semantics adds to the
activation of the proper lemma, and instrumental verbs are retrieved better than non-
instrumental. Interestingly, the anomic participants with elements of Broca’s aphasia did not
show the effect of instrumentality, which might indicate the positive role of healthy frontal
lobes for retrieving an instrument concept: whenever their dysfunction occurs, the beneficial
effect of instrumentality disappears. Another new result, which is not in line with the Dutch
results (Jonkers, 1998), is that the presence of an object in the conceptual representation of a
verb enhances verb retrieval too in Russian anomic speakers.
38
References
Bastiaanse, R. 1991. Naming of instrumental verbs in aphasia: An explorative study. Clinical
Linguistics and Phonetics 5, 355-368.
Breedin, S.D., Saffran, E.M., & Schwartz, M.F. 1998. Semantic factors in verb retrieval: an effect of
complexity. Brain and Language, 63, 1-31.
Jonkers, R. 1998. Comprehension and production of verbs in aphasic speakers. Groningen, GRODIL
25.
Jonkers, R., & Bastiaanse, R. 2007. Action naming in anomic aphasic speakers: Effects of
instrumentality and name relation. Brain and Language, 102, 262-272.
Kambanaros, M., & van Steenbrugge, W. 2006. Noun and verb processing in Greek-English bilingual
individuals with anomic aphasia and the effect of instrumentality and verb-noun relation.
Brain and Language, 97, 162-177.
39
Determining the Amount of Social-Interactional Participation
in People with Aphasia through CAL
Özlem Ünal, Seren Düzenlï Öztürk, Selen Ölmez, Aslı Bal & Azize Kılıç
Anadolu Üniversitesi-DİLKOM, Turkey
Introduction
The damage occurring in any part of the brain may affect the life of individual in many ways.
Aphasia occurs when the areas relating to the language on the left hemisphere of the brain
and connecting parts between those areas are damaged no matter what triggers it (paralysis,
vein diseases etc.).
Like every person, the individual with aphasia also has his/her place in the society of
communication world. The language loss changes the quality of social interaction
environment. Having lost his/her communicative routine, the individual with aphasia
acknowledges the situation and keeps attaining the same social environment or completely
alienates himself from his/her social communicative environment.
Method
Pulvermüller and friends (2001) developed a questionnaire called “The Communicative
Activity Log (CAL)” to ascertain patient use of language in everyday life so as to determine
the amount of communication. CAL versions existed for patient self evaluation and for
evaluation by professional clinicians. The original log includes 18 items; however, a few
questions have been added or subtracted having taken some cultural facts into consideration.
The interviewed aphasic people (n=20) have been asked to use the communication mode, in
which they are familiar to express themselves with while replying the questions. Replies can
be either with spoken language or any other communication mode other than that such as by
mimics, writing, picture showing or drawing, etc., none of which affect the results of the
study.
Results
The people with aphasia taking part in social-interactional environment after the language
loss have been evaluated through CAL. The expected results showed that the individuals with
aphasia still attempt for social-interactional participation after language loss. The output
lists demonstrated positive and negative alterations, as well.
40
References
Pulvermüller, F., Neininger, B., Elbert,T.., Mohr, B., Rock stroh,B., Koebbel,P., Taub, E. (2001).
Constraint-Induced Therapy of Chronic Aphasia after Stroke, Journal of The American Heart
Association, 32:1621.
41
Determining verbal fluency of Turkish speakers with literate
and illiterate educational background
Esra Ertan, Özlem Ünal, Şenol Sat, Fırat Sakar
Anadolu University, Turkey
Center for Speech & Language Disorders (DİLKOM)
Verbal fluency tests are sensitive to cognitive impairment; therefore are useful in detecting
dementia, in differentiating among types of dementia and in pointing out frontal lobe
involvement (Kave, 2005). Although not consistent, demographic characteristics such as
age, education level, and gender probably affect fluency performance. Verbal fluency is
measured by tests through which participants are required to list as many words as possible
from a given category in a given time (generally 60 seconds) (Lezak, 1995; Benito-Cuadrado
et al., 2002). Letter fluency and semantic/categorical fluency (Tombaugh et al., 1999;
Abwender et al., 2001) are the two types of testing fluency. This study focused on testing
\'semantic categorical fluency. The semantic fluency task is a method widely used by speech
pathologist to investigate retrieval of words from semantic memory (Pekkala, 2004). It is
especially used to determine the cognitive damage quantity and frontal lobe functioning.
These tasks have also been conducted in the investigation of illiterate and other low level
educational groups because they do not require reading or writing skills. There is yet no
norm study about verbal fluency in Turkish; consequently, this study was designed to obtain
normal verbal fluency pre-norms. The purpose of the study is to determine the frequency of
the semantic category naming which are produced by neurologically intact group and to
compare whether there is a difference between illiterate and literate groups distributed in
gender and levels of age.
The semantic categories are “animals, vegetables and fruits, vehicles, clothes, body parts, and
household goods”. The subjects participated in the study were 90; 45 males and 45 females.
The age groups were determined as three subgroups (20-44, 45-64, 65+). Educational
backgrounds were illiterate and literate (primary school to high school and college). The
participants were given a minute to list as many names as possible in one category. Tape
recording was used to make the transcriptions of outputs easy. The performance of subjects
was analyzed by counting correct responses and errors, such as category violations and
repetitions of words. Subcategories and subordinates were considered. The study implied
some conclusive remarks whether semantic fluency has been affected by literacy and formal
education. The results of this study will be discussed as a part of an extended study.
42
References
Abwender, D.A., Swan, J.G., Bowerman, J.T., and Connolly, S.W. (2001). Qualitative analysis of verbal
fluency output: Review and comparison of several scoring methods. Assessment, 8, 323–336.
Benito-Cuadrado, M. M., Esteba-Castillo, S., Böhm, P., Cejudo-Bolívar, C. and Peña-Casanova, J.
(2002). Semantic verbal fluency of animals: A normative and predictive study in a Spanish
population. Journal of Clinical an Experimental Neuropsychology, 24, 1117-1122.
Kavé, G. (2005). Phonemic fluency, semantic fluency, and difference scores: Normative data for adult
Hebrew speakers. Journal of Clinical and Experimental Neuropsychology, 27, 690-699.
Lezak, M. (1995). Neuropsychological Assessment (3. ed.). New York: Oxford University Press.
Pekkala, S. (2004). Semantic fluency in mild and moderate Alzheimer’s disease. Doctoral dissertation.
University of Helsinki, Department of Phonetics, Finland.
Tombaugh, T.N., Kozak, J., and Rees, L. (1999). Normative data stratified by age and education for
two measures of verbal fluency: FAS and Animal Naming. Archives of Clinical Neuropsychology, 14,
167–177.
43
How to diagnose Apraxia of Speech?
Towards the development of a diagnostic instrument
Judith Feiken1, Roel Jonkers2, Leonore Meilof1,2, Tineke den Exter1,2, Roelie
Sijbinga1,2, Marleen Schönherr1
1. Center for Rehabilitation, University Medical Center Groningen, University of
Groningen (NL); 2. University of Groningen, Department of Linguistics (NL)
Introduction
The underlying deficit in Apraxia of Speech (AoS) evoked a lot of debate (see Den Ouden,
2002 for an overview). The main question is whether the symptoms of patients with AoS are
caused by one underlying deficit, i.e. in motor programming (Maas, Robin, Wright & Ballard,
2008) or by a disturbance in two processes, namely a deficit in the programming and/or
planning of speech, as is suggested by McNeil (2002). One of the main reasons for the
ongoing debate is the lack of proper diagnostic materials for AoS. We developed a test
battery by taking the most extensive definition of AoS as starting point, where it can be a
deficit in the programming or in the planning of speech or in both (cf. McNeil, 2002). This
Diagnostic Instrument for AoS (DIAS), should differentiate AoS from dysarthria and
phonological disorders in aphasia in a clinical setting, since these disorders have many
common symptoms. Currently we are in the process of validating this diagnostic instrument.
Preliminary data are already available. The aim of the current study is to find out whether
the DIAS is able to distinguish AoS from dysarthria and aphasia.
Method
We tested 18 patients who suffered from a CVA (10 male, 8 female). 11 were diagnosed as
having AoS according to clinical judgments of a clinical linguist that was not involved in this
study. 3 patients had conduction aphasia (CA), diagnosed on the basis of the Dutch version
of the Aachen Aphasia Test (AAT; Graetz, De Bleser & Willmes,1992) and 4 were suffering
from atactic dysarthria (AD). Diagnosis of dysarthria was based on the Dutch version of the
Frenchay Dysarthria Test (FDT; Lambert & Rutten, 1996). Mean age of the patients was 60.
A group of 10 non-brain-damaged control subjects that was matched in age, sex and
education to the brain-damaged patients was also tested.
Design of the diagnostic instrument
The most common symptoms of AoS that are described in the literature were considered
while developing our diagnostic instrument. Many of these symptoms are, however, also
44
observed within patients with dysarthria or aphasia. The aim of the DIAS is to evaluate the
presence of ‘exclusive’ symptoms, like the presence of buccofacial apraxia, inconsequent
realizations of phonemes, more errors in consonants than in vowels, more difficulty in
alternating syllables than sequencing syllables, struggle to position the articulators,
initiations problems, intersyllabic pauses, segmentation of consonant combinations and an
effect of articulatory complexity. An overview of the test battery and the symptoms that are
studied with the different subtests (including references) is presented in table 1.
Table 1: Diagnostic Instrument for AoS (DIAS)
Task
Items Symptoms exclusively linked
to buccofacial apraxia
References
1 Task for
buccofacial
movements
10 items improved execution by imitation
struggle to position the
articulators
Wertz et al. 1984, Duffy,1995,
Romani & Galuzzi, 2005.
Task
Items Symptoms exclusively linked
to apraxia of speech (AoS)
2 Articulation of
individual
consonants and
vowels
15
consonants
and 15 vowels
inconsequent realizations of
phonemes
more errors in consonants than in
vowels
Wertz et al. 1984; Varley en
Whiteside, 2001; Den Ouden, 2002.
LaPointe & Johns, 1975; Darley et
al. 1975; Wertz et al. 1984; Romani
& Galuzzi, 2005.
3 Diadochokinesis
task
6 sequencing
and 6
alternating
items of 3
syllables/
words
more difficulty in alternating
syllables than sequencing
syllables
struggle to position the
articulators
Wertz et al.,1984, Deger & Ziegler
2002, Ogar et al. 2006.
Darley et al. ,1975, Wertz et al.,
1984; LaPointe, 1990; Haynes,
1992.
4 Articulation of
words
66 items,
increasing in
difficulty
(clusters,
number of
syllables)
initiation problems
intersyllabic pauses
segmentation of consonant
combinations
effect of articulatory complexity
La Pointe, 1990; Haynes, 1992.
Darley et al. ,1975, Wertz et al.
1984; Varley & Whiteside, 2001;
McNeil, 2002 ; Edmonds &
Marquandt, 2004.
McNeil, 2002.
Romani & Galuzzi, 2005; Ogar et al,
2006, Staiger & Ziegler, 2008.
The items in the tests were always presented orally and written. By this, we tried to avoid
that the reaction of the patient was influenced by aphasic problems like deficits in phoneme
encoding, auditory comprehension, grapheme identification, and word-finding problems.
45
Procedure
All subtests were administered in one session. Reactions were audio and video recorded. All
reactions were scored. In the current study we focused on 4 specific outcomes, namely the
scores on (1) the buccofacial apraxia test and (2) the diadochokinesis test, (3) the number of
initiation problems, and (4) the number of inconsequent realizations of individual
phonemes. These 4 outcomes were thought to be useful to differentially diagnose AoS from
atactic dysarthria or conduction aphasia.
The differences between the scores of the different types of language-disturbed subjects were
compared using non-parametric tests (Mann-Whitney U).
Results
The control subjects scored at ceiling on all tests. In table 2 the comparison between the
scores of the different language-disturbed groups are depicted. Statistical outcomes are
included.
Table 2: Results of the between group analyses (Mann-Whitney test); AD= Atactic dysarthria; CA= conduction aphasia; AoS= Apraxia of Speech
Tests ` Groups Z-score Significance
Buccofacial apraxia AoS-AD -2,649 p < 0,01 Diadochokinesis (DDK) AoS-AD -2,347 p < 0,05 Initiation problems AoS-AD -2,422 p < 0,05 Number of inconsequent realizations of individual phonemes
AoS-AD -0,993 p=0,320
Buccofacial apraxia AoS- CA -1,975 p < 0,05 Diadochokinesis (DDK) AoS - CA -0,781 p=0,435 Initiation problems AoS - CA -1,891 p=0,059 Number of inconsequent realizations of individual phonemes
AoS - CA -2,438 p < 0,05
Between group analyses showed that the patients with AoS scored significantly worse on the
buccofacial test, the diadochokinesis test and the initiation of articulation as compared to the
atactic dysarthric patients (AD). The analyses also showed that the patients with AoS made
significantly more errors on the buccofacial test and showed a higher number of
inconsequent realizations of individual phonemes than the conduction aphasic subjects (CA).
Discussion
This study revealed that the DIAS is able to distinguish between patients with AoS, atactic
dysarthria and aphasia. The fact that the DIAS even differentiates AoS from the atactic
dysarthria (AD) is of particular interest as this form of dysarthria is seen as deficit in the
articulatory motor programming and therefore in general shows a lot of similarities with
46
AoS. Patients with AD show for instance also inconsequent distortions, syllable
segmentations and speech delay (Darley et al., 1975; Kent, Kent, Duffy, Thomas, Weismer &
Stunteback, 2000, Spencer & Rogers, 2005). The differentiating aspects we found in the
current study are the presence of buccofacial apraxia, more difficulty in alternating syllables
than sequencing syllables, and initiation problems in a repetition task in the AoS group
against the AD group. The results of this study also shows that the DIAS differentiates
conduction aphasic speakers (CA) from patients with AoS. This is remarkable as well, as this
type of aphasia also shows similarities with AoS. Like AoS patients, patients with conduction
aphasia show symptoms in all speech production tasks, make substitutions and show an
effect of word length (McNeil, 2002; Romani & Galluzzi, 2005). The differentiating aspects
in the current study were the presence of buccofacial apraxia and the number of
inconsequent realizations of individual phonemes.
As part of the validation of the test, currently 50 language-disturbed subjects and 50 non-
language disturbed control subjects are tested. We hope to find more differentiating aspects,
and to provide a basis for a severity scale for apraxia of speech and implications for therapy.
References
Aichert, I., Ziegler, W. (2004). Syllable frequency and syllable structure in apraxia of speech. Brain
and Language, 88, 148-159.
Ballard, K., Granier, J., Robin, D. (2000) Understanding the nature of apraxia of speech: Theory,
analysis and treatment. Aphasiology, 14, 10, 969-995.
Darley, F., Aronson, A., Brown, J. (1975). Motor Speech Disorders. W.B. Saunders Company.
Philadelphia, London, Toronto.
Deger, K., Ziegler, W. (2002). Speech motor programming in apraxia of speech. Journal of phonetics,
30, 321 – 335.
Duffy, J. (1995). Motor Speech Disorders. Substrates, differential diagnosis, and management.
London: Mosby.
Edmonds, L., Marquardt, T. (2004) Syllable use in apraxia of speech: Preliminary findings.
Aphasiology, 18, 12, 1121-1134.
Graetz, P., De Bleser, R, Willmes, K. (1992). Akense Afasie Test (AAT). Lisse: Swets & Zeitlinger.
Haynes, W., Pinzola, R., Emerick, L. (1992). Diagnosis and evaluation in speech pathology.
Englewood-Cliffs:Prentice-Hall.
Kent, R., Kent, J., Duffy, J., Thomas, J., Weismer, G., Stuntebeck, S. (2000). Ataxic dysarthria.
Journal of Speech, Language and Hearing research, 43, 1275-1289.
Lambert, J. J., Rutten, Ch. (1996). FDO: Frenchay Dysartrie Onderzoek (handleiding). Lisse: Swets
&. Zeitlinger.
LaPointe, L.L., Johns, D.F. (1975). Some phonemic characteristics in apraxia of speech. Journal of
Communication disorders, 8, 259-269
LaPointe, L. (1990). Neurogenic disorders of speech. In: Shames, G., Wiig, E. (eds.) Human
communication disorders. Columbus, Merrill Publ. Comp.
47
Maas, E, Robin, D.A., Wright D.L., Ballard K.J. (2008). Motor programming in apraxia of speech.
Brain and Language 106, 107-18.
McNeil, M. (2002). Clinical Management of Sensorimotor Speech Disorders. Thieme Medical
Publishers, New York.
Ogar, J., Willock, S., Baldo, J., Wilkins, D., Ludy, C., Dronkers, N. (2006). Clinical and anatomical
correlates of apraxia of speech. Brain and Language 97, 343-350.
Ouden, D.B. den (2002). Phonology in Aphasia. Syllables and segments in level-specific deficits.
Groningen Dissertations in Linguistics. Enschede: Print Partners Ipskamp.
Robin, D.A., Jacks, A., Hageman C., Clark, H.M., Woodworth, G. (2008). Visuomotor tracking
abilities of speakers with apraxia of speech or conduction aphasia. Brain and Language 106, 98-106.
Romani, C., Galluzzi, C. (2005). Effects of syllabic complexity in predicting accuracy of repetition and
direction of errors in patients with articulatory and phonological difficulties. Cognitive
neuropsychology, 22-7, 817-850.
Spencer, K., Rogers, M.A. (2005). Speech motor programming in hypokinetic and ataxic dysarthria.
Brain and Language, 94, 347-366.
Staiger, A., Ziegler, W. (2008). Syllable frequency and syllable structure in the spontaneous speech
production of patients with apraxia of speech. Aphasiology 22, 1201-1215.
Varley, R.A., Whiteside, S,P. (2001). What is the underlying impairment in acquired apraxia of
speech? Aphasiology, 15, 39-84.
Wertz, R., LaPointe, L., Rosenbek, J. (1984). Apraxia of speech in adults. The disorder and its
management. Orlando: Grune & Stratton.
48
Wiebke Grauel, Linda Cupples, Caroline Law
Centre for Language Sciences, Macquarie University
Background
Production of passives has been found to be difficult for individuals with aphasia, regardless
of the nature of the underlying deficit (e.g. grammatical deficit in non-fluent aphasia versus
lexical semantic deficit in fluent forms). Passivisation has traditionally been assessed with
verbs that assign the agent to subject position and the theme to object position (e.g., agent-
theme verbs, such as 'climb'), resulting in a canonical active voice structure. Different types
of verbs (e.g., theme-experiencer verbs, such as 'frighten'), however, assign the theme to
subject position in the active voice, while it is assigned its preferred object position in the
passive voice. Unlike agent-theme verbs, theme-experiencer verbs have been found to be
more commonly produced in the passive voice by non-brain-damaged (NBD) individuals.
This study investigated whether people with aphasia would also find it easier to produce
grammatical passives with theme-experiencer verbs than with agent-theme verbs.
Method
Data were collected from four individuals with aphasia, two of whom presented with
moderate (NF1) and moderate-severe (NF2) non-fluent aphasia, and two with mild (F1) and
moderate (F2) fluent aphasia. In addition, forty-five NBD individuals participated in the
study. Production of passives was assessed by means of a highly constrained sentence
anagram task as well as a picture description task comprising two conditions with different
levels of constraint.
Results
In the highly constrained anagram task, all participants produced passives with both types of
verb. By contrast, in the moderately constrained picture description context, both NF1 and
F1 as well as NBD participants produced grammatical passives with both verb types, while
NF2 and F2’s passives were restricted to theme-experiencer verbs. In the least constrained
context passives were readily produced only with theme-experiencer verbs, both by NBD
speakers and individuals with aphasia.
Conclusion
Results suggest that difficulty producing passives in non-fluent and fluent forms of aphasia
may not result purely from grammatical or lexical semantic deficits but is influenced by the
severity of the aphasia, as well as the thematic structure of the verb, resulting in a higher
49
likelihood for grammatical passives to be produced with theme-experiencer verbs than with
agent-theme verbs.
50
Language Characteristics of a Child with Landau-Kleffner
Syndrome: A 4-Year Outcome Study
Selçuk Güven, Dilek Eroglu, Dilber Kaçar &, Ayşegül Zencir
Anadolu University-DİLKOM (Center for Speech and Language Pathology), Turkey
Introduction
Acquired epileptic aphasia (AEA, or Landau-Kleffner syndrome) was first reported by
Landau and Kleffner (1957) as a regressive syndrome encountered in children during early
developmental milestones. The prevalence is unclear. Onset may be sudden or gradual and it
occurs usually between ages of 3-7 years in the critical period for language development
(Baynes, et al., 1998). They show paroxysmal bilateral epileptic EEG discharges in the
temporal lobes. Deonna (2004) argue that the language loss is a direct consequence of the
focal epileptic seizures. Children with AEA show normal peripheral hearing and normal
brain imaging. In AEA the first manifestation of the language difficulties is “word deafness”
or “auditory verbal agnosia” (Pearl, et al., 2001). Previous research on speech and language
characteristics of these children noted that most of these children have difficulty in
discriminating speech sounds due to impaired receptive language. After the child loses the
ability to understand language, his speech begins to deteriorate. Some studies demonstrate
that, in early forms of this syndrome, the phonology and syntax use of the children may be
intact whereas production of grammatical morphology is more impaired. In addition to
communication problems, these children may also have attention deficits, behavioral
problems and cognitive impairments (Deonna, 2004). Some of these abilities may recover,
but many children with AEA have significant language impairments that persist into
adulthood. Studies in Turkish children with AEA have been limited and most of these
focused on medical symptoms of the syndrome.
Method
In present study, we describe speech and language characteristics of a 7-year-old male whose
speech and language got impaired after the onset of AEA.T he child was referred to D İLKOM
with a complaint of “speech and language impairment” associated with epileptic discharges
at 3 of age. Onset was by age 2;6 year old with the first seizure. He was diagnosed in 6
months as AEA, especially after his speech and language development began to regress.
According to the parental report, the child demonstrated normal speech-language
development until the first seizure. He had more than 50 single words and word
combinations. With the seizure, the child lost his word combinations, his receptive language
got impaired and after the last seizure he had auditory verbal agnosia and he started
51
producing nonspeech sounds. This child received 3 year long speech and language services at
D İLKOM. Language assessments included the Turkish version of the Peabody Picture
Vocabulary Test (PPVT), Turkish version of the Test of Early Language Impairment (TELD-
3-T), Ankara Developmental Screening Inventory (AGTE) , and interviews of mother-child
and child clinician interactions. We aimed in this study to compare pre and post language
assessment findings of the case in 4 year-therapy duration.
Results
We investigated that the linguistic abilities of the child was compatible with the literature
findings. We observed that he has still difficulties in understanding spoken words,
discriminating speech sounds, production of speech sounds and he has also morphological
errors. In addition to communication problems, his behavioral problems and attention
deficits continued until this age.
References
Landau,W. M., & Kleffner F. R. 1957. Syndrome of acquired aphasia with convulsive disorder in
children. Neurology, 7, 523-530.
Pearl, L.P., Carrazana, E. J., & Holmes, L. G. 2001. The Landau-Kleffner syndrome. Epilepsy Currents,
2, 39-45.
Baynes K., Kegl, A.J., Brentari, D., Kussmaul, C., & Poizner, H. 1998. Chronic Auditory Agnosia
Following Landau-Kleffner Syndrome: A 23 Year Outcome Study. Brain and Language, 63, 381-425.
Robinson, O. R., Baird, G. , Robinson, G., & Simonoff, E. 2001.Landau–Kleffner syndrome:course and
correlates with outcome. Developmental Medicine & Child Neurology , 43, 243–247.
Deonna T. 2004. Acquired epileptic aphasia (AEA) or Landau-Kleffner syndrome: From Childhood to
adulthood.In D.VM. Bishop & L.B.Leonard (Eds), Speech and language impairments in children (pp.
261-272). New York:Psychology Press.
52
Effects of semantic transparency in reading German prefixed
verbs: Evidence from acquired dyslexia
Judith Heide, Antje Lorenz, Ria De Bleser & Frank Burchert
University of Potsdam, Dept. of Linguistics
Background
Dual route models of morphological processing assume that complex words can be processed
both holistically and via their constituent morphemes. Amongst other factors, a full form’s
semantic transparency seems to determine which route becomes active.
Semantic transparency describes the degree to which a full form’s meaning can be inferred
from the meaning of its constituent morphemes. Schreuder & Baayen (1995) assume that
fully transparent forms are not represented in the mental lexicon as their meaning can be
accessed through their morphemes. In contrast, (partially) opaque words receive their own
concept node to specify semantic properties.
Neurolinguistic evidence for this assumption comes from Hamilton & Coslett (2008) who
describe two English speaking patients with phonological dyslexia. Both patients performed
better on reading opaque than transparent derived suffixed words. For derived prefixed
words, the patients performed equally well on opaque and transparent words, although
prefix errors were more frequent for transparent than for opaque words. Hamilton & Coslett
(2008:357) argue that their patients are impaired in “engaging decompositional mechanism”
and thus show a greater deficit with semantically transparent words.
The aim of our study was to test whether this error pattern also holds for the processing of
German derived verbs which are at the same time prefixed and suffixed. Specifically, we
asked whether patients with deep or phonological dyslexia would read (partially) opaque
words better than fully transparent items and whether affix errors occur more frequently for
transparent full forms.
Methods
Nine patients with phonological dyslexia participated in a multiple single case study. They
were asked to read aloud 171 German prefixed verbs. All verbs were derived by prefixation
and consisted of the prefix ver-, a root, and the infinitival suffix -(e)n. The root was either a
noun, a verb or ambiguous for noun/verb status. According to the results of a rating with
n=60 German native speakers, each item was classified as (partially) opaque (e.g. verpulvern
~ to splash sth., derived from Pulver ~ powder; n=97) or fully transparent (e.g. verdursten ~
to die of thirst, derived from Durst ~ thirst; n=74). Subsets were matched on full form and
root frequency, word length and number of orthographic neighbours.
53
Reading correctness (0/1) was scored for the full form as well as for the prefix, the root and
the suffix. We used Pearson’s chi-square test to compare the performance on opaque vs.
transparent items.
Results
Six of the nine patients tested did not show any effect related to the semantic transparency of
the full form. Three patients (BE, NI, SC) showed better performance for transparent than
for opaque items. Significant differences and trends that were close to significance were
detectable when comparing error rates on full forms (BE, SC), prefixes (NI, SC) and roots
(BE, NI; see also Figure 1), but not for the suffix. Most errors were visually related to the
target word. Morphological errors included the omission and substitution of prefix and
suffix.
Figure 1: Reading correctness for transparent (light grey) and opaque (dark grey) full forms, and their
prefix, root, and suffix. ***p<.001; **p<.05; *trend p<.08 (χ2 analyses)
Discussion
Three of nine patients with phonological dyslexia showed effects of semantic transparency
when reading German verbs derived by prefixation. This shows that a full form’s
transparency can (but not necessarily has to) influence the lexical processing of a
morphologically complex word. Differences in the performance on (partially) opaque vs. fully
transparent items is in line with dual route models of morphological processing.
In contrast to the patients described by Hamilton & Coslett (2008), our patients showed
better performance for transparent than for opaque items. Even more surprising, clear
morphological errors (i.e. omissions or substitutions of the prefix) occurred more frequently
for opaque than for transparent full forms. Still, the results mirror unimpaired processing of
*
**
**
*** **
*
54
German prefixed verbs. In a lexical decision task (Heide et al., unpublished data), response
latencies were significantly higher for (partially) opaque than for fully transparent words.
Thus, the processing of opaque ver-prefixed verbs seems to be more difficult than the
processing of matched transparent full forms, leading to more reading errors in dyslexia and
to higher response latencies in unimpaired processing.
How can these results be reconciled with the assumption that opaque words are represented
holistically in the mental lexicon while transparent full forms rely on a decomposed
representation (Schreuder & Baayen 1995)? It is possible that although semantic
transparency affects the mental representation, all ver-prefixed verbs undergo
morphological decomposition during visual word recognition. This is plausible because ver-
is easily identifiable as a prefix. Due to left-to-right processing, the prefix is processed before
the root and also before the reader can encounter the semantic transparency of the full form.
For semantically transparent full forms, the automatic decomposition is successful as the
meaning can be computed through the constituent morphemes. In the case of opaque words,
morphological decomposition takes place – which explains the existence of morphological
reading errors – but it fails. Therefore, in a second step, the lexicon has to be searched for a
holistic representation which increases processing time in unimpaired word recognition.
References
Hamilton, A.C. & Coslett, H.B. (2008). Role of inflectional regularity and semantic transparency in
reading morphologically complex words: Evidence from acquired dyslexia. Neurocase, 14 (4),
347-368.
Schreuder, R. & Baayen, R.H. (1995). Modeling Morphological Processing. In: Feldman, L.B. (ed.).
Morphological Aspects of Language Processing. London: LEA.
55
The influence of phonetic features on aphasic speech
perception
Dörte Hessler, Roel Jonkers & Roelien Bastiaanse
University of Groningen, Center for Language and Cognition Groningen
Introduction
Speech perception is multimodal. Not only auditory but also visual information (seen
speech) is processed in perception (Rosenblum, 2008). It has been proven that seeing the
speaker facilitates comprehension in a noisy environment or with demanding contents.
Further evidence for the contribution of seen speech was added by the McGurk effect
(McGurk and MacDonald, 1976). This effect describes the fusion between seen and heard
speech. Subjects watched dubbed videos with auditory and visual information that did not
match and were asked to report what they perceived. Instead of answering with the auditory
(/ba/) or the visual (/ga/) component of the video they mostly reported a fusion of both
(/da/).
Furthermore, studies have shown that aphasic listeners benefit from seeing the speaker’s
face (Shindo, 1991). It is however unclear which phonetic features (“place of articulation”,
“manner of articulation” and/or “voicing”) can actually make use of the additional
information from seen speech.
Blumstein et al. (1977) showed that English-speaking aphasic listeners had more problems
detecting differences in only one phonetic feature than differences in two or all three. Klitsch
(2008) used the Dutch version of the PALPA (Bastiaanse et al., 1995) to investigate whether
there were differences in detecting distinctions in the features “place of articulation”,
”manner of articulation” and ”voicing”. She came to the cautious conclusion that “place of
articulation” was affected most, but also noted that the feature “voicing” could not be
compared reliably to the other features, as the material included “voicing” contrasts only in
initial, the other contrasts however only in final position or metathesis.
Aim
The current study investigates the influence of (additional) lip-reading on the aphasic
subject’s perception of speech. It furthermore aims at determining whether Dutch aphasic
subjects also can detect wider distinctions (three phonetic features) more easily than more
narrow ones. Finally we examine which phonetic features are most vulnerable in aphasia if
manipulated in the same position.
56
Procedure
So far 4 Dutch aphasic subjects with problems in comprehension as well as 14 non-brain-
damaged control subjects participated in the current study. All aphasic subjects were at least
4 month post onset (4-146, average 43.5) and between 47 and 64 years (average 54) old. 1
was female.
A nonword discrimination task (same-different judgment) was carried out to investigate how
different phonetic features were affected by aphasia. In this task there were three conditions:
“auditory only” (subjects could only hear the speaker), “visual only” (the speaker could only
be seen) and “audiovisual” (subjects heard and saw the speaker).
Material consisted of CVC(C) syllables that formed strings phonologically possible but non-
existing in Dutch. The stimuli were recorded in a quiet environment and spoken by a male
Dutch native speaker.
All differences occurred in the first segment only because Dutch has final devoicing
precluding a distinction in voicing in the final segment. The material was controlled for the
amount of features differing within a pair as well as the actual feature differentiating it (see
Figure 1).
Results
Non-brain-damaged control subjects scored at ceiling for the “auditory only” and
“audiovisual” conditions. In the “visual only” condition they performed worse, failing mainly
in contrasts involving only “voicing” or “manner of articulation” or the combination of both.
57
Aphasic subjects scored significantly lower than control subjects on all conditions (see table
1). Their performance also differed between the three conditions (Friedman Anova: X2 = 8,
df = 2, p < .05). Performance was best in the “audiovisual” condition and worst in the “visual
only” condition.
Further analyses revealed that the number of features differing within the pair indeed plays a
role for aphasic subjects in the “auditory only” (Friedman Anova: X2 = 6.857, df = 2, p < .05)
and “audiovisual” (Friedman Anova: X2 = 7.429, df = 2, p < .05) conditions. In both
conditions differences in only one feature are least often detected (67% for “auditory only”
and 73% “audiovisual”) compared to similar results for differences in two and three features
(“auditory only”: 89% and 87%; “audiovisual”: 89% and 93%). Furthermore the influence of
type of feature (“place” vs. “manner” vs. “voicing”) was investigated, leading to significant
results for the “audiovisual” condition (Friedman Anova: X2 = 6.533, df = 2, p < .05) and
marginally significant results for the “auditory only” condition (Friedman Anova: X2 =
5.692, df = 2, p = .058). For both conditions it appears that “voicing” was the most difficult
distinction, followed by “place of articulation” and “manner of articulation” (see Figure 2).
Discussion
The results obtained so far indicate that aphasic subjects have difficulties in discriminating
pairs of nonwords. These difficulties are more profound for smaller differences. As
58
previously reported by Blumstein et al. (1977) for English, we also find for Dutch speakers
that differences are less likely to be detected if the items within the pair differ by only one
phonetic feature. This holds for the “auditory only” as well as for the “audiovisual” condition.
Klitsch (2008) reported that “place of articulation” seems to be the most vulnerable feature;
we found opposite results in the current study, showing that for the subjects included so far
differences in “voicing” were least likely to be detected. This difference in results can be
explained by the different materials used. In the PALPA (Bastiaanse et al., 1995)
discrimination tasks, used by Klitsch (2008), the position of the difference is not balanced
between “place of articulation”, “manner of articulation” and “voicing”. In the current study,
however, all differences were manipulated in initial position providing a better basis for a
comparison between the features.
The current study clearly shows that if contrasts between items result from more phonetic
features, they are more easily detected. Furthermore, the type of feature differentiating items
is of importance, indicating that differences in “voicing” are most difficult to perceive for
Dutch aphasic listeners.
References
Bastiaanse, R., Bosje, M., & Visch-Brink, E. (1995). PALPA: Nederlandse Versie. Hove: Lawrence
Erlbaum.
Blumstein, S. E., Baker, E., & Goodglass, H. (1977). Phonological factors in auditory comprehension in
aphasia. Neuropsychologia, 15(1), 19-30.
Klitsch, J. (2008). Open your eyes and listen carefully. Auditory and audiovisual speech perception
and the McGurk effect in Dutch speakers with and without aphasia. Groningen Dissertations
in Linguistics (GRODIL), 67.
McGurk, H., & MacDonald, J. (1976). Hearing lips and seeing voices. Nature, 264(5588), 746-748.
Rosenblum, L. D. (2008). Speech Perception as a Multimodal Phenomenon. Current Directions in
Psychological Science, 17(6), 405-409.
Shindo, M., Kaga, K., & Tanaka, Y. (1991). Speech discrimination and lipreading in patients with word
deafness or auditory agnosia. Brain and Language 40(2), 153-161.
59
Verb retrieval and Aspect in Chinese agrammatic aphasia
Wei-Ming Ho1 & Roelien Bastiaanse2
1Klinische Linguistik Einheit, Universität Bielefeld, Germany
2Center for Language and Cognition Groningen (CLCG), University of Groningen, The
Netherlands
Bastiaanse & Jonkers (1998) showed that in Dutch spontaneous speech of agrammatic
patients, although the number of lexical verbs on 300 words is normal, the diversity of
lexical verbs is reduced and the proportion of finite verbs is lower than normal. It was argued
that this was caused by problems combining verb retrieval and verb (cluster) inflection. In
Dutch, the verb is inflected for Tense and Agreement.
The question this study raised was whether a similar relationship could be established in a
language that does not use verb inflection but expresses time reference through free-
standing aspect markers. Mandarin-Chinese is such a language. It has been suggested that
aspectual markers, though quantitatively less frequent, play a similar role as verb inflections
(Norman, 1988; Packard, 1990, 1993).
For this study, the use of three aspect markers has been analysed; these were the aspect
markers that were used by the subjects participating in the study.
(1) Zai is a progressive marker used to indicate the progressive aspect of an action. Zai is
special in terms of its pre-verbal position in a sentence, while the rest of the aspect
markers occur postverbally.
Zhangsan zai xue yinyu
Zhangsan ZAI learn English
“Zhangsan is learning English”
(2) Le is the perfective marker is to express the completion of an action. When le comes
immediately after a verb, it carries the meaning of a completed action connected to
the verb preceding it; when it appears at the end of a sentence, le acts as a sentence-
final particle for emphasis, not really as an aspect marker (Law & Cheng, 2002).
Zhangsan xue-le sannian yingyu
Zhangsan learn-LE three year English
“Zhangsan has learned English for three years”
Wo bu mai shu le
I no buy book LE
“I don’t buy books (anymore)”
60
We were mainly interested in the verb final le, since this form is comparable to verb
inflection, but both were counted.
(3) Zhe is used to indicate the continuous manner of an action or a state, implying
that the action is still going on during the reference time of the speech.
Zuotian chuangzi hei kei-zhe
Yesterday window still open-ZHE
“Last night the window was still open”
The research questions was:
Methods
For the present study, speech samples from constrained conversation and picture story-
telling were collected from 6 agrammatic Broca patients (as diagnosed by the Chinese
version of the BDAE) and 5 non-brain-damaged speakers and 300 words were transcribed
and analyzed for each individual. To measure the diversity of the lexical verbs, the number of
different lexical verbs were counted. Instead of the proportion of finite verbs, the proportion
of aspect markers (number of aspect markers divided by the number of clauses) was
calculated, once with the sentence final le and once without.
Results
The results are given in Table 1 and graphically represented in Figure 1.
Table 1: The percentages of le postverbally as an aspect marker and sentence-final as an
emphasis marker; the range for the percentages of the aspect markers; and the range for
the number of different verbs on 300 words for both subject groups. (Percentages are on
number of clauses.)
non-brain-damaged-speakers agrammatic speakers
% verb-le 18 5
% sentence-le 3 16
range % aspect
markers
35-90 4-33
range verb diversity 27-34 17-26
61
Figure 1: Graphical representation of the mean percentages (on
number of clauses) of aspect markers with (left bar) and without
(middle bar) sentence final le and the mean number of different verbs
on 300 words (right bar)
As can be seen in Table 1, the use of the two different kinds of le is different among the
groups: the NBD-speakers use it primarily as an aspect marker in post-verbal position,
whereas agrammatic speakers use le very frequently, but mainly as a marker for emphasis
(Fishers’ exact: p=0.0001). It was decided to exclude sentence final le from further analysis.
The ranges of the scores are compared, because the groups are too small for statistical
comparisons. The number of lexical verbs on 300 words is normal (non-brain damaged
speakers 33-41; agrammatic speaker 29-40), but the diversity of lexical verbs is reduced: all
agrammatic speakers score outside the normal range. This pattern, a normal number of
lexical verbs on 300 words but reduced diversity, is similar to the Dutch results. None of the
agrammatic speakers scores inside the normal range on the proportion of aspectual markers,
which is comparable to the reduced proportion of finite verbs found in Dutch agrammatic
speech.
Discussion
Aspect markers in Chinese and Tense in Dutch both have the function to mark the timeframe
of the proposition. A verb is used to ‘name’ this proposition. Bastiaanse & Jonkers (1998)
assumed that the combination of retrieving the verb and inflecting it for Tense and
Agreement was the core problem of verb use for agrammatic speakers. However, considering
the Chinese data, this interpretation is too language specific, since aspectual markers in
Chinese are not inflectional morphemes. Considering that the underlying deficit is the same
in both languages, we, therefore, propose that it is the combination of retrieving a verb to
62
name the proposition and marking the timeframe in which the proposition takes place that is
the core problem. The result is reduced verb diversity in both languages, a low number of
aspect markers in Chinese and a low proportion of finite verbs in Dutch.
References
Bastiaanse, R., & Jonkers, R. (1998). Verb retrieval in action naming and spontaneous speech in
agrammatic and anomic aphasia. Aphasiology, 12, 951-969.
Law, S., & Cheng, M. (2002). Producition of grammatical morphemes in Cantonese aphasia,
Aphasiology, 16, 693-714
Norman, J. (1988). Chinese. Cambridge: Cambridge Language Surveys.
Packard, J.L. (1990). Agrammatism in Chinese: A case study. In L. Menn & L.K. Obler (eds.),
Agrammatic Aphasia: Cross-language Narrative Sourcebook (pp.1191-1223). Amsterdam,
Philadelphia: John Benjamins.
Packard, J.L. (1993). A Linguistic Investigation of Aphasic Chinese Speech. Dordrecht: Kluwer
Academic Publishers.
63
Capturing nonverbal information in narrations of aphasic
patients with reduced verbal output: Does limb apraxia make
a difference?
Katharina Hogrefe, Nicole Weidinger, Wolfram Ziegler, & Georg Goldenberg
Clinical Neuropsychology Research Group (EKN),
Clinic for Neuropsychology, Hospital Bogenhausen, Municipal Clinic München
GmbH
Introduction
In treatment of language disorders, several approaches make use of gestures as a
compensatory strategy or for the facilitation of word retrieval. However, the prerequisites for
successful treatment remain unclear. Especially the relationship between limb apraxia and
the ability to produce gestures in spontaneous conversation is not yet well understood.
Research in this field has led to a number of contradictory outcomes. While results of some
studies (e.g. Borod 1989) suggested an influence of limb apraxia on spontaneous gesturing in
conversation, more recent data provide evidence for dissociations (e.g. Rose & Douglas
2003). However, patients’ aphasia type and degree of severity of the language impairment
varied considerably across the above mentioned studies. Conflicting results could be due to
variation of the need for nonverbal means of communication to support or replace oral
speech.
In the present study, we investigated the impact of limb apraxia on spontaneous gesturing in
patients suffering from aphasia with highly limited verbal output. Concerning limb apraxia
assessment, we were mainly interested in the scores of the pantomime task, since the ability
to produce meaningful gestures to command may count as an adequate indicator for
spontaneous gesturing skills.
In a previous study, aphasic patients were videotaped while retelling six short video-clips. In
the present study, we used these narrations as stimulus materials in a forced-choice
identification paradigm. Healthy persons were asked to match patients’ narrations to the
original films, solely on the basis of the visual nonverbal information. The aim of the study
was to determine the influence of limb apraxia on the ability to convey relevant information
of a given story nonverbally.
Method
Participants: 18 healthy subjects took part in the study. All subjects were native speakers of
64
German.
Material: In a previous study, 24 aphasic patients had narrated six short video-clips. Their
narrations had been videotaped and were now presented to the experimental subjects of this
study.
The 24 patients had left hemisphere brain damage and aphasia with highly limited
verbal output. Only patients who scored 2 points or less for communication behavior
on the “verbal communication” scale of the Aachen Aphasia Test (AAT; Huber et al.
1983) were included in the study.
All patients underwent limb apraxia assessment. Eleven patients showed normal
performance in the Pantomime-to-Command-Test, the other thirteen displayed
disturbances in this task.
For each healthy subject an individual PowerPoint-Presentation was prepared. This
presentation contained the instruction of the experiment and a training part as well as the
experimental trial. The experimental trial contained 32 patient narrations.
Procedure:
Data collection took place in two experimental runs. In the first run, narrations of 16 patients
were presented. In the second run, narrations of another eight patients as well as narrations
of eight of the patients of the first run were used.
Each healthy person was presented with the individual PowerPoint-Presentation. After a
training part, the subject saw the narrations of the patients without sound. After each
narration, the participant had to indicate on a questionnaire which film he thought he had
seen and how certain he was about this decision. Further, the subject was asked to indicate
on a given list which features of the film he had detected. There was a fifteen-minute-pause
after the first 16 of the 32 stimulus films.
Data analysis:
For each of the 24 patients, three scores of gestural information were determined:
1. Identification Rate: the percentage of correct matches (out of 18 judgements).
2. Feature Ratio: Healthy persons were asked to mark on a list, which features of the story
they thought to have recognized. As the feature number differed between the films, we first
calculated a quotient per correctly recognized film. On the basis of these ratios the mean for
each patient was calculated.
3. Certainty Index: The healthy subjects indicated on a six point scale how certain they were
about the identification of each film (1 = I guessed, 6 = I was completely sure). On the basis
of these ratings, we calculated a mean score for each patient.
Finally, the Gesture Scores were compared to the scores obtained in the pantomime testing.
Results
The three Gesture Scores were strongly correlated with the scores for Pantomime of Tool
65
Use.(Pearson-Test: Pantomime-Score with Identification Rate: r = 0,786, p < .001, with
Feature Ratio: r = 0,605, p < .01, with Certainty Index: r = .761, p < .001). The better the
patients performed in the Pantomime-to-Command-Test, the more videos and features were
recognized and the more certain were the healthy subjects about their decisions.
The figure illustrates the relationship between the obtained Identification Rates and the
Pantomime-Scores. The patients without limb apraxia obtained Identification Rates between
61 and 94% whereas the range for the apractic patients was much wider, namely from 22 to
94%.
Discussion
The results of our study demonstrate a strong relationship between the Gesture Scores and
the scores of the Pantomime-to-Command-Test. These outcomes suggest that disturbed
perfomance in typical limb apraxia assessment leads to difficulties in conveying messages
nonverbally. However, our results give the impression that the severeness of the apractic
disorder is a crucial factor for the prediction of effective nonverbal communication in those
patients.
References
Borod JC. The Relationship between Limb Apraxia and the Spontaneous Use of Communicative
Gesture in Aphasia. Brain and Cognition 1989, 10:121-131.
Huber W, Poeck K, Weniger D & Willmes K. Aachener Aphasie Test. Göttingen: Hogrefe, 1983.
Rose M & Douglas J. Limb apraxia, pantomime, and lexical gesture in aphasic speakers: Preliminary
findings. Aphasiology 2003, 17: 453-464.
66
Oral Diadochokinesis in Apraxia of Speech
Joost Hurkmans[1], Roel Jonkers[2], Judith Feiken[3], Annemarijke Boonstra[1],
Madeleen de Bruijn[1], Hans Arendzen[4], Heleen Reinders-Messelink[1]
Background
The precise nature of apraxia of speech (AOS) is elusive. The consensus is that AOS
represents a speech motor programming deficit. However, there are some models of AOS
that distinguish between motor planning and motor programming (e.g. Van der Merwe,
1997). AOS is characterized by multiple symptoms one of which is dysfluent speech
production. Different therapy programs have been developed to train speech fluency (Wertz
et al, 1984; Wambaugh & Martinez, 2000; Albert et al, 1973). Since 2005, Speech-Music
therapy for Aphasia (SMTA, De Bruijn et al, 2005) has been used to train fluency of speech at
the phonological and articulatory level. To evaluate the efficacy of this program we aimed to
design a sensitive test to measure changes in motor speech functioning Therefore a
diadochokinesis test (DDK) was developed. Diadochokinesis requires (rapid) syllable
repetitions of sequencing and alternating articulatory movements (Ackermann et al 1995).
Repetitions of monosyllables are considered to be a sensitive test to measure motor
performance of speech (Ziegler, 2002). The current study explores the coherence between
consistency, accuracy and fluency in a DDK test in subjects with AOS. Furthermore the
influence syllable structure and type of alternation have on the performance will be
evaluated.
Methods
Participants: 10 subjects with AOS and 10 healthy control speakers were tested. Subjects
with AOS vary in the time post onset (1 – 28 months post onset) and severity of speech
impairment. The diagnoses were established by a speech therapist on the basis of the
following characteristics: dysfluent speech, frequent starts, phonemic errors, groping for
articulation. Individual data of the subjects with AOS are presented in table 1. The control
speakers matched the subjects with AOS in age, gender and education.
Materials and procedures: The test consists of 4 blocks containing 17 items all of which were
syllables. The blocks represent the following syllable structures: CV, CVC, CVCC and CCVC.
Each block starts with sequential diadochokinesis namely repetition of the same syllable,
(e.g. /pa/, /pa/, /pa/) and then systematically alternates in the distinctive features place
[1] Rehabilitation Centre “Revalidatie Friesland”, The Netherlands [2] University of Groningen, Department of Linguistics, The Netherlands [3] Center for Rehabilitation, University Medical Center Groningen, University of Groningen, The Netherlands [4] Leiden University Medical Center, Department of Rehabilitation Medicine, The Netherlands
67
(e.g. /pa/, /ta/, /ka/), manner (e.g. /da/, /na/, /la/) and vowel change (e.g. /pa/, /po/,
/pu/). The participants were required to repeat the syllables 5 times upon visual and/or oral
presentation by an examiner. No particular instructions regarding speech rate were given.
The responses were recorded on videotape.
Table 1:Patient sample
Patient Diagnosis Months since onset Age Gender m/f
1. M.E. AOS 5 36 f
2. D.G. AOS 1 69 m
3. M.J. AOS 5 57 f
4. D.M. AOS 4 70 m
5. D.O. AOS 1 67 m
6. M.W. AOS 2 64 f
7. M.S. AOS 1 46 f
8. D.W. AOS 1 44 m
9. P.R. AOS 28 52 m
10. J.J. AOS 9 34 m
Analysis: The recordings were analysed in terms of:
A) consistency (i.e. first imitation is baseline, repetitions are compared with the
baseline)
Scores: 0: repetition equal to baseline, 1: repetition different from baseline, 2:
repetition different from response one and two, 3: repetition different from all other
repetitions
B) accuracy (i.e. number of articulatory errors like deletions, substitutions,
insertions)
Scores: 0: response is equal to target item, 1: one or two articulatory errors like
deletion, substitution, insertion and distortion, 2: three to five articulatory errors, 3:
more than five articulatory errors.
C) speech fluency (i.e. presence of intersyllabic pause, visual groping, selfcorrections,
prosodic errors, phonemic repetitions)
Scores: 0: response is fluent, 1: response is more fluent than dysfluent, 2: response is
as much fluent as dysfluent, 3: response is more dysfluent than fluent, 4: response is
severely dysfluent
To test the coherence between these 3 measures we determined correlations between these
scores using the Pearson’s test (two-tailed).
68
The influence of the syllable structure and distinctive features of articulation (place, manner
and vowel change) were analysed with a repeated measures ANOVA. Since the number of
items of each syllable block was unequal, percentages were used for the analysis of the
syllable structure.
Results
The scores of the healthy control speakers were at ceiling. Their scores are not presented and
will further be ignored. The scores on the different measures of the subjects suffering from
AOS are presented in table 2. Between the measures consistency, accuracy and fluency
significant correlations were found, all at the level of p<.001.
Table 2: Scores of patient group (higher scores resemble worse performance)
A significant effect of syllable structure was found (F(3,27)=14,89, p<.001). Post-hoc
analyses revealed that subjects with AOS made significantly less errors on CV structures as
compared to CVC structures (t (9)=4.29, p=<.01), CVCC structures (t(9)=4.29, p=<.01) and
CCVC structures (t (9)=4.89, p=<.01). They also made less errors on CVC structures as
compared to CVCC structures (t(9)=2.57, p=0.03) and CCVC structures (t(9)=3.25, p=.01).
syllable structures containing consonant clusters did not differ significantly from each other:
CVCC-CCVC (t(9)=1.13, p=.29). The distinctive features of articulation such as place, manner
and vowel change did not have a significant effect on the scores: F(2.18)=0.102, p>.05
Patient Consistency Accuracy Fluency CV CVC CVCC CCVC Place Manner Vowel-
change
1. M.E. 42 125 46 20 43 74 72 56 62 57
2. D.G. 36 215 48 47 86 88 88 59 88 88
3. M.J. 48 233 58 53 110 88 88 77 81 88
4. D.M. 39 178 47 43 64 69 88 83 61 46
5. D.O. 24 59 17 11 27 21 41 18 16 23
6. M.W. 22 121 11 21 38 30 65 35 44 40
7. M.S. 38 175 39 14 68 82 88 62 52 65
8. D.W. 6 12 6 0 13 3 8 11 1 3
9. P.R. 9 23 29 5 14 31 10 27 12 9
10. J.J. 8 9 19 1 11 15 9 13 4 12
Maximum
possible 51 255 68 88 110 88 88 88 88 88
69
Conclusion and discussion
Significant correlations were obtained between consistency, accuracy and fluency for the
subjects with AOS. Therefore these measures can be regarded as coherent. The analysis of
syllable structure show that length (number of phonemes) influences the performance of
subjects with AOS in a DDK test. Subjects with AOS experience more difficulties in being
consistent, accurate and fluent when the number of phonemes increases within a syllable
structure. The location of a cluster within the syllable structure (i.e. initial or final) does
however not effect performance. Also the type of difference (i.e. which distinctive feature was
changed) has no influence on the scores.
Currently we run pilot studies using the treatment program SMTA and the DDK-test to
measure treatment-induced changes. Outcomes of these future studies will be compared to
the scores of the current study in order to evaluate the DDK as a tool for measuring effects of
treatment.
References
Ackermann, H., Hertrich, I. & Hehr, T. (1995). Oral Diadochokinesis in Neurological Dysarthrias,
Folia Phoniatrica et Logopaedica, 47, 15-23.
Albert, M., Sparks, R. & Helm, N. (1973). Melodic Intonation Therapy, Archives of Neurology, 29,
130-131.
Bruijn, M., Zielman, T. & Hurkmans, J. (2005). Speech-Music Therapy for Aphasia (SMTA),
Revalidatie Friesland, The Netherlands.
Van der Merwe, A. (1997). A theoretical framework for the characterisation of pathological speech
sensorimotor control. In M. R. McNeil (Eds.), Clinical management of sensorimotor speech
disorders, 1-25, New York: Thieme.
Wambaugh, J. & Martinez, A. (2000). Effects of rate and rhythm control treatment on consonant
production accuracy in apraxia of speech, Aphasiology, 14, 851-871
Wertz, R., LaPointe, L. & Rosenbek, J. (1984). Apraxia of speech in adults. The disorder and its
management. Grune & Straton.
Ziegler, W. (2002). Task-Related Factors in Oral Motor Control: Speech and Oral Diadochokinesis in
Dysarthria and Apraxia of Speech, Brain and Language, 80, 556-575.
70
Paradoxical successive recovery of Bilingual Aphasia
(Turkish – Persian) after stroke
K.Johari2*, H. Ashayeri3**
*Tehran University of Medical Sciences, Faculty of Rehabilitation Sciences.
** Iran University of Medical Sciences, Faculty of Rehabilitation Sciences.
Objective
To report the successive recovery pattern of a bilingual Aphasia
Background
Aphasia in bilingual individuals provides an interesting context in which to investigate the
cerebral representation of language. Different language recovery patterns are reported in
bilingual aphasia. We present a case of paradoxical successive recovery in (Turkish –
Persian) bilingual patient.
Design of study
A case report of an educated Turkish – Persian bilingual. (B.E Patient, 61 – year old, right
handed male who is has become nonfluent in his native Turkish and also Persian.
Result
B. E experienced stroke that involved: basal ganglia, left thalamus and left temproparietal
cortex. Neurological findings indicated sensorimotor hemiparesia dexter. The
neurolinguistic assessment showed Broca, motoric Aphasia with severe impairment in
naming.
Second language (Persian) has been recovered within 8 weeks after stroke, whereas the
mother tongue (Turkish) recovered with delay of 8 weeks.
Discussion
Studies on bilingual aphasia have suggested that subcortical lesions impair access to mother
tongue, while sparing second language. Two months after stroke second language recovered
whereas mother tongue recovered three months after the stroke. In our study Seven months
2 - Speech and language pathologist 3 - Prof. Dr. med. Neuropsychiatrist. Nauropsycholosist. Correspondings to: Prof. Dr. med. H.Ashayeri Neuropsychiatrist. Nauropsychologist. Email: Hassanasha@ yahoo.com
71
after stroke, B.Es’ second language is better than his mother tongue, which supports the
hypothesis that basal ganglia are essential nods in the the Neural Network subserving the
mother tongue.
Key words: Bilingual Aphasia, paradoxical successive recovery, Stroke.
72
RATS-2: Cognitive-linguistic therapy and communicative
therapy equally effective in early aphasia: a randomized
controlled trial
M. de Jong-Hagelstein, W.M.E. van de Sandt-Koenderman, N.D. Prins, D.W.J.
Dippel, P.J. Koudstaal, E.G. Visch-Brink
Dept. of Neurology, Erasmus Medical Center, Rotterdam, the Netherlands
Introduction
Well-designed trials on the efficacy of aphasia therapy in general and on specific treatments
are needed 1,2,3,4,5. The frequently used therapy techniques can be ranged under two different
approaches: cognitive-linguistic therapy (CLT) directed to the affected linguistic levels and
communicative therapy directed to optimizing information transfer.
We focused on evaluating CLT. In the first Rotterdam Aphasia Therapy Study (RATS-1)6, we
compared two forms of CLT from 4-12 months post stroke: semantic therapy (BOX7, n-29)
and phonological therapy (FIKS8, n=26). They appeared to be equally effective on a
functional outcome measure: verbal communication (Amsterdam-Nijmegen Everyday
Language Test, ANELT9), whereas linguistic measures of semantic and phonological abilities
showed a treatment-specific effect. This differential improvement on the semantic and
phonological measures in response to semantic or phonological therapy correlated with the
improvement on the ANELT. We concluded that there may be two routes that lead to
improved verbal communication: a semantic and a phonological route.
Data from a meta-analysis showed that the greatest treatment effect is reached in the first
three months; the effect of treatment is reported to be nearly twice as large as the effect of
spontaneous recovery alone1. CLT in particular may positively influence the recovery of
specific neural circuits related to specific linguistic functions such as semantics and
phonology10.
Therefore, in the current RCT we investigated the efficacy of CLT applied in the first six
months. Considering the found equal efficacy of BOX and FIKS, we combined the two
therapy programs in the experimental condition.
Methods
Aim
To measure the efficacy of CLT applied in the first six months on verbal communication and
on semantic and phonological processing in aphasic stroke patients. We hypothesized that
73
CLT would be more effective than communicative therapy and that its effect would be the
greatest in the first three months.
Participants
Patients with aphasia due to stroke, within three weeks post stroke, aged 18-85, with a
semantic and/or phonological disorder.
Procedure
The assessment of patients (Table 1) started as soon as possible. Subsequently, patients were
randomized, stratified by center, and therapy was started at the latest three weeks post
stroke. Patients were retested at three and six months and received therapy until month six
or shorter, namely until they had recovered.
Table 1: The tests used for the assessment of participants
Language tests
Semantic tests Phonological tests Other language tests
ANELT9 Semantic
Association Test
(SAT)13, verbal
version
Nonwords
Repetition (PALPA14) ScreeLing11
Partner
Communication
Questionnaire16
Boston Naming
Test12
Semantic
Association words
with low
imageability
(PALPA14)
Auditory Lexical
Decision (PALPA14)
Spontaneous speech
interview
Aachen Aphasia
Test17 (only at 6
weeks post stroke)
Semantic word
fluency (animals,
professions)
Letter fluency (d, a,
t)
Sabadel15 (story
telling)
Token Test18 (36
items version)
General assessment
EuroQol19 Rankin20 Barthel21
Experimental condition: CLT, consisting of BOX and/or FIKS (paper and computer
versions).
74
Control condition: communicative therapy, which aims at improving communicative ability
using all verbal and nonverbal strategies available to the patient. By definition, exercises are
always personally relevant and embedded in a communicative setting. Examples are:
PACE22; verbal and augmentative strategies, e.g. written choice communication and
communication books; role-playing and conversational coaching.
Therapy was applied with a minimum of two hours a week on average and preferably for five
hours a week, partly individual and partly as homework.
Outcomes
The primary outcome measure was the ANELT, scale A (understandability), scored blinded
by two independent, experienced SLTs. The means of both raters’ scores were used in the
analyses. The secondary outcome measures were the three semantic and the three
phonological tests.
Statistical methods
We tested the difference between the two groups in score and in gain score on the ANELT at
three months and at six months with 95% CI by ANCOVA, adjusted for baseline severity23.
In addition, we compared the proportion of patients in each group who had a moderate to
mild/no communication disorder (score 30-50) after three and six months by Odds Ratio
with 95% CI using logistic regression. Finally, we compared the scores and gain scores of
both groups on the semantic and phonological tests by ANCOVA, again adjusted for
baseline.
Sample size
We calculated that a sample of 70 patients would provide a power of 0.87 to detect a
clinically significant difference of seven points on the ANELT between the two therapy
groups at a 5% 2-sided significance level.
Results
We randomized 85 patients (over 70 to compensate for non-evaluable patients). The
intention-to-treat group consisted of patients who were retested at least once, the on-
treatment group required patients to have completed therapy. Figure 1 shows the numbers
and reasons for loss-to-follow-up.
75
Flow chart
R
41 CLT
38 intention-to-treat
37 on-treatment
44 no-CLT
42 intention-to-treat
38 on-treatment
85 enrolled
→ 2 lost (1 ill,
1 refused)3 lost (1 died, ←
1 ill, 1 SLT refused)
1 lost ←
(died)
→ 4 lost (1 anxiety,
1 frustration,
1 non-compliant,1
domestic problems
Figure 1: Flow chart of enrolled patients
There was no significant difference between the mean ANELT score of patients who received
CLT and patients who received communicative therapy, neither three nor six months post
stroke (Table 2). There was a trend at three months regarding the proportion of patients in
each group who had a moderate to mild/no communication disorder: 71% (CLT) versus 55%
(OR=2.03, CI=0.80 – 5.13). However, this trend was not present anymore at six months
(76% versus 71%, OR=1.29, CI=0.47 – 3.52).
Table 2: Mean ANELT scores of the CLT and the communicative therapy
group (intention-to-treat)
CLT
(n=38)
Communicative
(n=42)
Difference
(95% CI)
Adjusted difference
(95% CI)
3 months post stroke 33.4 31.6 1.8
(-7.4 – 3.8)
1.5
(-2.6 – 5.6)
6 months post stroke 35.2 33.2 1.9
(-7.3 – 3.4)
1.6
(-2.3 – 5.6)
Improvement was significantly larger after CLT than after communicative therapy only on
the fluency tasks: on semantic fluency after three months (p=0.02), on phonological fluency
after six months (p=0.03) (Table 3). On all other secondary measures there was no treatment
effect.
76
Table 3: Mean improvement on fluency tasks of the CLT (n=37) and the
communicative treatment (n=41) group (intention-to-treat)
After 3
months
of CLT
After 6
months
of CLT
After 3 months
of
communicative
treatment
After 6 months
of
communicative
treatment
Difference
(95% CI)
Adjusted
difference
(95% CI)
Semantic
fluency
8.5 5.2 3.3
(0.6; 6.1)
3.2
(0.4 – 6.0)
Letter
fluency
7.7 4.5 3.2
(0.4; 6.1)
3.1
(0.3 – 6.0)
We conducted post-hoc analyses on three equally large severity groups based on baseline
ANELT score. In the most severely impaired group, there was a trend towards a beneficial
effect of CLT over communicative treatment (Figure 2).
Post hoc analyses
0
5
10
15
20
25
30
35
40
45
50
0 months 3 months 6 months
CLT
communicative
n=13
n=14
n=14
n=13
n=11
n=15
Figure 2: ANELT scores of three equally large severity groups
77
The most severely impaired patients improved significantly more after six months of CLT
than after communicative therapy on SAT verbal (p=0.04) and on auditory lexical decision
(p=0.03).
Discussion
RATS-2 could not confirm the hypothesis that, at an early stage, treatment directed to
specific cerebral functions is more effective than a broad communicative treatment. It is
unknown what the effect of therapy was over spontaneous recovery.
The intensity of therapy is reported to be a crucial factor for its effectiveness24. Maybe the
intensity in our study (less than the preferred five hours a week) did not reach the threshold
necessary to find a treatment effect. The question whether the acute stage is the optimal
period for aphasia therapy (and for which method) is not yet resolved.
According to our post-hoc analyses the patients with a severe aphasia appeared to benefit
most from CLT. This is in line with the meta-analysis by Robey1: patients with a severe
aphasia were reported to profit the most from therapy.
This meta-analysis has a prominent position in the aphasia literature with conclusions about
the optimal period of therapy and about the group of patients who benefited most. The value
of both assumptions still has to be confirmed.
References 1Robey, R.R. (1998). A meta-analysis of clinical outcomes in the treatment of aphasia. Journal of
Speech and Hearing Research, 41: 172-187.
2Greener, J., Enderby, P., & Whurr, R. (2000). Speech and language therapy for aphasia following
stroke. Cochrane Database Syst Rev., 2: CD000425.
3Cicerone, K.D., Dahlberg, C.,Kalmar, K., et al. (2000). Evidence-Based Cognitive Rehabilitation:
Recommendations for Clinical Practice. Arch Phys Med Rehabil, 81: 1596-615.
4Cicerone, K.D., Dahlberg, C., Malec, J.F., Langenbahn, D.M., Felicetti, T., Kneipp, S., Ellmo, W.,
Kalmar, K., Giacino, J.T., Harley, J.P., Laatsch, L., Morse, P.A., & Catanese, J. (2005). Evidence-Based
Cognitive Rehabilitation: Updated Review of the Literature From 1998 Through 2002. Arch Phys Med
Rehabil, 86: 1681-92.
5Cappa, S.F., Benke, T., Clarke, S., Rossi, B., Stemmer, B., & van Heugten, C.M. (2005). EFNS
guidelines on cognitive rehabilitation: report of an EFNS task force. European Journal of Neurology,
12: 665-680.
6Doesborgh, S.J.C., van de Sandt-Koenderman, M.W.E., Dippel, D.W.J., van Harskamp, F., Koudstaal,
P.J., Visch-Brink, E.G. (2004). Effects of Semantic Treatment on Verbal Communication and
Linguistic Processing in Aphasia After Stroke: A Randomized Controlled Trial. Stroke, 35: 141-146.
7Visch-Brink, E.G., & Bajema, I.M. (2001). BOX, een semantisch therapieprogramma. Lisse: Swets &
Zeitlinger.
8Van Rijn, M., Booy, L., & Visch-Brink, E.G. (2000). FIKS, een fonologisch therapieprogramma. Lisse:
Swets & Zeitlinger.
78
9Blomert, L., Koster, Ch., & Kean, M-L. (1995). Amsterdam-Nijmegen Test voor Alledaagse
Taalvaardigheid. Lisse: Swets & Zeitlinger.
10Code, C. (2001). Multifactorial Processes in Recovery from Aphasia: Developing the Foundations for
a Multileveled Framework. Brain and Language, 77: 25-44.
11Doesborgh, S.J.C., van de Sandt-Koenderman, W.M.E., Dippel, D.W.J., van Harskamp, F.,
Koudstaal, P.J., & Visch-Brink, E.G. (2003). Linguistic deficits in the acute phase of stroke. Journal of
Neurology 250, 977-982.
12Kaplan, E., Goodglass, H., & Weintraub, S. (2001). Boston Naming Test. Philadelphia-Tokyo:
Lippincott, Williams and Wilkins.
13Visch-Brink, E.G., Stronks, D.L., & Denes, G. (2005). De Semantische Associatie Test. Amsterdam:
Harcourt Assessment B.V.
14Bastiaanse, R., Bosje, M., & Visch-Brink, E.G. (1995). Psycholinguïstische testbatterij voor de
taalverwerking van afasiepatiënten. Hove, UK: Lawrence Erlbaum Associates Ltd. A Dutch adaptation
of Kay J, Lesser R, Coltheart M. (1992) Psycholinguistic Assessment of Language Processing in
Aphasia. Hove, UK: Lawrence Erlbaum Associates Ltd.
15Van Eeckhout, Ph., Sabadel, Signoret, J.-L., & Pillon, B. (1982). Histoires insolites pour faire parler.
Paris: Médecine et Sciences Internationales.
16Blomert, L. (1995). Who’s the “expert”? Amateur and professional judgement of aphasic
communication. Topics in stroke rehabilitation, 2: 64-71.
17Graetz, P., de Bleser, R., & Willmes, K. (1991). Akense Afasie Test. Nederlandstalige versie. Lisse:
Swets & Zeitlinger.
18De Renzi, E., & Faglioni, P. (1978). Normative data and screening power of a shortened version of
the Token Test. Cortex, 14: 41-49.
19The EuroQol Group (1990). EuroQol: a new facility for the measurement of health related quality of
life. Health Policy, 16:199-208.
20Van Swieten, J.C., Koudstaal, P.J., Visser, M.C., Schouten, H.J.A. & van Gijn, J. (1988).
Interobserver agreement for the assessment of handicap in stroke patients. Stroke, 19: 604-607.
21Mahoney, F.I., Barthel, D.W. (1965). Functional evaluation: the Barthel index.
MD State Med J, 14: 61-65.
22Davis, G.A., & Wilcox, M.J. (1985). Adult aphasia rehabilitation: Applied pragmatics. San Diego:
Singular.
23Vickers, A.J., & Altman, D.G. (2001). Analysing controlled trials with baseline and follow up
measurements. British Medical Journal, 323: 1123-4.
24Bhogal, S.K., Teasell, R., & Speechley, M. (2003). Intensity of aphasia treatment, impact on
recovery. Stroke, 34(4): 987-993.
79
Perceptual discrimination of Shona lexical tones and hums by
Left and Right hemisphere damaged patients
McLoddy Kadyamusuma, Ria De Bleser, Jöerg Mayer
Potsdam University, Potsdam University, Stuttgart University
The current study investigated the ability to discriminate Shona lexical tones and hums by
left-hemisphere-damaged aphasic patients (LHD), right-hemisphere-damaged (RHD)
patients and a control group. Shona is a language that is spoken in Zimbabwe, central
Mozambique, northeastern Botswana and southern Zambia by about 10 million people
(Gordon 2005). Typologically the language belongs to the Bantu family. Shona uses pitch
variations that are linguistically significant at the syllable level to change the meaning of
individual words. Prosody therefore performs a phonemic function in Shona which operates
under a two tone system High and Low, (H) and (L) respectively. Unlike in other tone
languages for example in Thai with monosyllabic words for instance (kha:) leg, galangal root
and to be lodged in, Shona does not license monosyllabic words. Although the tone inventory
of Shona is not as sophisticated as the Asiatic languages which have at least 4 tones
(Brunelle, 2009), the two tones in Shona combine to form tone groups for instance, HL
(Falling), LH (Rising), HH (High tone) and LL (Low) in bisyllabic words.
Research that has been done on the perception of lexical tones has shown that LHD patients
are more impaired than RHD patients. Van Lancker (1980) proposed that the more linguistic
pitch contrasts are, the more laterized they are to the left hemisphere while the less linguistic
pitch contrasts are laterized to the right hemisphere. According to Van Lancker’s functional
lateralization hypothesis, hemispheric specialization is associated with different domains of
pitch contrast, with lexical tone being the "most linguistically structured" and affect and
voice quality being the “least linguistically structured.” The present study tested six LHD and
six RHD patients in two discrimination tasks. The stimuli consisted of 120 pairs of bisyllabic
Shona words and 120 pairs of hummed versions of the same words for the speech condition
and non-speech condition, respectively. Segmental information was eliminated (i.e.,
consonant, vowel) from lexical versions of the stimuli while preserving suprasegmental
information (i.e., duration, pitch) to derive the hum version of the stimuli. Stimulus pairs
were of two subgroups, 60 pairs elicited the judgment the pair is the same while the other 60
pairs elicited the response the pair contains different sounds or words (e.g. /gu´ru´/ (HH), -
/guru´/ (LH)) polygamous union and third stomach of a ruminant respectively.
The two tasks measured the participants´ accuracy in making same/different judgments on
pairs of hums and pairs of Shona words. The RHD patients performed significantly better
than the LHD aphasic patients on both tasks. The LHD patients discriminated the lexical
80
tones at only 59% and the hums at 61% accuracy, while the RHD patients discriminated the
hums at 81 % and the lexical tones at 82% accuracy. These results are in line with what has
been found in Asiatic studies that lexical tones are relatively more impaired in LHD patients
in comparison to RHD patients. Despite different tonal systems between the Asiatic and
Bantu languages, a unilateral brain damage also results in tonal impairment in Shona
patients. The current study supports Van Lancker´s functional lateralization hypothesis at
the phonemic and lexical levels.
References
Baum, S.R. and Pell, M. D. 1999. The neural bases of prosody: Insights from lesion studies and
neuroimaging. Aphasiology, 13, 581-608.
Brunelle, M. 2009. Tone perception in Northern and Southern Vietnamese. Journal of Phonetics 37,
79-96.
Burnham, D., Francis, E., Webster, D., Luksaneeyanawin, S., Attapaiboon, C., Lacerda, F., Keller, P.
1996. Perception of lexical tone across languages: Evidence for a linguistic mode of processing Spoken
Language, ICSLP 96. Proceedings Fourth International Conference, Issue, 2514 – 2517.
Boersma, P., & Weenink, D. (2007). Praat: doing phonetics by computer
Gandour, J. (2000). Frontiers of brain mapping of speech prosody. Brain and Language 71, 75-77.
Gandour, J., Petty, S H., & Dardarananda, R. 1988. Perception and production of tone in aphasia.
Brain and language 35, 201-240.
Gandour, J., Wong, D.,Hsieh L., Weinzapfel, B., Van Lancker, D., and Hutchins, G.D. (2000). A
crosslinguistic PET study of tone perception. Journal of Cognitive Neuroscience, 12, 1, 207-222.
Gordon, Raymond G., Jr. (Editor). 2005. Ethnologue: Languages of the World, Fifteenth Edition.
Dallas: SIL International.
Klein, D., Zatorre, R.J., Milner, B., and Zhao, V. (2001). A cross-lingustic PET study of tone perception
in Mandarin Chinese and English speakers. NeuroImage, 13, 646-653.
Moen, I., and Sundet, K. 1996. Production and Perception of word tones (Pitch accents) in patients
with left and right hemisphere damage. Brain and Language, 53, 267-283.
Klein,D., Zatorre, R.J., Milner, B., and Zhao, V. (2001). A cross-linguistic PET study of tone
perception in Mandarin Chinese and English speakers. NeuroImage, 13, 646-653.
Van Lanker, D. 1980. Cerebral lateralization of pitch cues in the linguistic signal. Papers in
Linguistics, 13 (2), 201-277.
Yip, M. (2002). Tone. Cambridge, MA: Cambridge University Press.
Zatorre, R.J. and Gandour, J.T. 2007. Neural specializations for speech and pitch: moving beyond the
dichotomies. Philosophical Transactions of the Royal Society B.
81
A Descriptive study of Aphasia-Apraxia's features'
Correspondence and conformity of a Trilingual Broca's
Aphasic Patient (A Case Study)
Leila Karimi- Dr. Mohammad hadi Falahi- Dr. Mohammad Reza Parhizgar
Fars science and research university- student of M.A
Regard the importance of brain and language studies, clearly their disorders also worth to
survey. This article undertakes to describe the linguistic speech features of a trilingual
Broca's aphasic patient (a case study) who is 32 years old, right-handed, and able to speak
(not fluently) and understand (almost completely) German, Persian and English. Since there
are some evidences for close relationship, correspondence and conformity between his
behavior disorders (Apraxia), and his speech disorders' (aphasia, agraphia, dyslexia)
characteristics, as well as describing these evidences, it is confirmed and concluded that it
must exist something in the body ( as an specific Chromosome) to affect and relate this
sameness and harmony between speech and behavior disorders.
Methodology
This case study is done by interviewing with each language of the three on different days. The
patient's psychologist helped researchers to diagnose his Apraxia's disorders.
Main References:
- Pearce J.M.S : A Note on Aphasia in Bilingual Patients: Pitres' & Ribot's Law,
Europian Neurology, 2005;54:127-131
- Paradis Michel: A Neurolinguistic Theory of Bilingualism. 2004: No.18, John
Benjamin Pub.Co.
- Perani Daniel & Abutalebi Jubin: The Neural Basis of First & Second Language
processing, Elsevier, 2005;15:202-206
- Meinzer M. et al: Recovery From Aphasic as a Function of Language Therapy in an
Early Bilingual Patient: Elsevier, 45(2007) 1247-1256
- David W. Green: The Neurocognition of Recovery Patterns in Bilingual Aphasics
University College London
- Nilipour, R., & Ashayeri, H. (1989). Alternating Antagonism between Two Languages
with Successive Recovery of a Third in a Trilingual Aphasic Patient. Brain and
Language, 36, 23-4
Key words: Broca aphasic, Apraxia, trilingual, language disorders
82
Verb retrieval, argument structure and inflection in aphasia:
Insights from different verb classes in Greek
Vasiliki Koukoulioti, Stavroula Stavrakaki
Aristotle University of Thessaloniki
Previous studies on aphasia revealed two main vulnerable domains: a) verb inflection and b)
verb retrieval. With respect to the first, findings suggest that verb inflection is selectively
impaired in a range of languages, with Tense and/or Aspect being more impaired than
Agreement (e.g. for Greek, Stavrakaki & Kouvava 2003; Varlokosta et al. 2006).
Interestingly, this pattern appears also in fluent aphasia (e.g. Varlokosta et al., 2006).
With respect to retrieval there has been found an effect of verb argument structure
(henceforth VAS) on verb retrieval. Transitive verbs were found to be more affected than
intransitive ones for agrammatic patients (Thompson et al., 1997). Thompson (2003),
however, found an heterogeneity in the intransitive verbs, namely that unergative
intransitive verbs are easier for agrammatic patients than unaccusative intransitive ones.
The difference between the two classes is that the unaccusative subject is a deep structure
object and, thus, has to move. On the other hand, Bastiaanse & van Zonneveld (2004) found
that the unaccusative intransitive version of verbs with alternating transitivity is more
difficult than their transitive one for agrammatic aphasics, but not for Wernicke/anomic
aphasics. The researchers claim that agrammatic patients have problems with the
grammatical encoding of the arguments, namely at the post-lemma level (Bastiaanse & van
Zonneveld, 2004).
This study aims at investigating the performance of Greek-speaking aphasic patients on the
production of transitive, unaccusatives and unergative verbs as well as at evaluating the
effect of inflectional marking on verb retrieval and VAS assignment.
To this end, we tested four patients, one agrammatic (AG), one anomic (AN) and two
Wernicke’s aphasics (W1, W2) whose diagnosis was made on the basis of the BDAE (adapted
for Greek) (Tsapkini et al., 2008) and 5 age matched control subjects. The tasks used were
the following:
1) Sentence elicitation with video stimuli, in which the participants were presented with
videos and had to describe what was happening. Correct responses were those
including (i) correct verb lemma and (ii) correct VAS retrieval.
2) Sentence elicitation and tense marking task, in which the participants were presented
with the same video stimuli preceded by a phrase prompting for a specific
tense/aspect (T/A) marking. They were instructed to combine the phrase and the
83
video in order to produce a correct sentence. Correct responses were those including
(i) correct verb lemma (ii) correct VAS retrieval and (iii) correct T/A marking.
The actions depicted in the videos could be described by unergative, unaccusative and
transitive verbs. There were 7 videos for unergative and transitive and 6 videos for
unaccusative verbs.
The control participants performed at ceiling in both tasks. The patients’ correct
performance on the three verb categories in both tasks is presented in Figure 1.
0
20
40
60
80
100
TASK 1-un
erga
tives
TASK 1-un
accu
sativ
es
TASK 1-tra
nsitiv
es
TASK 2-un
erga
tives
TASK 2-un
accu
sativ
es
TASK 2-tra
nsitiv
es
AGR
AN
W1
W2
Figure 1. Correctness percentages for unergative, unaccusative and transitive verbs in Task 1
and Task 2
In Task 1, all patients (except for W1) found unaccusative verbs the most difficult to produce.
We suggest that the difficulties the aphasic patients have with unaccusative verbs arise from
the mapping mismatch between the theme θ-role and the subject position. The performance
of the patients with respect to transitives, on the other hand, was very variable, so that these
findings cannot support a general effect of number of arguments on retrieval.
As for Task 2, AG’s performance dropped remarkably in comparison to Task 1, whereas the
performance of the fluent patients (except for W1) improved. In other words, there was a
facilitative effect of verb inflection on verb lemma and VAS retrieval only for two of the fluent
patients and not for the agrammatic patient.
Eliciting tense/aspect inflection activated the correct verb and VAS retrieval, hence the
significant increase of the performance of the two fluent patients on all verb classes.
Improvement in verb and VAS retrieval is correlated with prompting grammatical features at
least for some patients. This finding is compatible with studies indicating interdependence
between verb retrieval and grammatical feature marking (e.g. see a therapy study by
84
Weinrich et al., 1997). Whilst the tense and aspect feature of the verb in Greek are very
prominent-a property related to the rich morphological nature of Greek-they do not facilitate
lemma retrieval for the agrammatic patient presumably because of the deficits that these
patients show with tense and aspect.
We argue that production of unaccusative verbs is problematic across aphasia types. We
claim that this is related to the mapping mismatch between the patient theta-role and the
syntactic subject position. In addition, we suggest that verb inflection affects verb production
either facilitating it in fluent aphasia or hampering it in agrammatism.
References
Bastiaanse, R. & van Zonneveld, R. (2004). Broca’s aphasia, verbs, and the mental lexicon. Brain and
Language, 90, 198–202.
Stavrakaki, S., & Kouvava, S. (2003). Functional categories in agrammatism: Evidence from Greek.
Brain and Language, 86, 129-141.
Thompson C. K. (2003). Unaccusative verb production in agrammatic aphasia: the argument
structure complexity hypothesis. Journal of Neurolinguistics, 16, 151-167.
Thompson, C.K., Lange K.L., Schneider, S.L. & Shapiro L.P. (1997). Agrammatic and non-brain
damaged subjects’ verb and verb argument structure production. Aphasiology, 11 , 473-490.
Tsapkini, K., Emmanouil, A., Passalidou, G., Nassiopoulou, G., Vlahou, C.E., Potagas, K. (2008).
Prosarmogi tis diagnostikis Exetasis gia tin Afasia tis Bostonis – Syntomi morfi stin elliniki glossa kai
dimiourgia normon gia ton elliniko plithismo [Adaptation of BDAE – Short Version in Greek and
norm establishment for the Greek population] Paper Presentation in the Second Panhellenic
Conference of Cognitive Psychology, Thessaloniki, November 6-8, 2008.
Varlokosta, S., Valeonti, N., Kakavoulia, M., Lazaridou, M., Economou, A. & Protopapas, A. (2006).
The breakdown of functional categories in Greek aphasia: Evidence from agreement, tense, and
aspect. Aphasiology, 20, 723-743.
Weinrich M., Shelton, J.R., Cox, D.M, & McCall, D. (1997) Remediating production of tense
morphology improves verb retrieval in chronic aphasia. Brain and Language, 58, 23-45.
85
Syllable frequency effect in progressive apraxia of speech: a
case study
Marina Laganaro1, Odile Bagou1, Michèle Croisier2
1. University of Neuchâtel, Switzerland
2. Hôpital neuchâtelois, Neuchâtel, Switzerland
Introduction
Although apraxia of speech (AoS) has usually been described in degenerative language
disorders in association with non-fluent progressive aphasia, at least 20 clinical cases of
pure progressive apraxia of speech (PAoS) have been reported in the literature (Didic et al.,
1998; Duffy, 2006; Joseph et al. 2006). Impairment in AoS is usually ascribed to the level
of programming of speech gestures (phonetic encoding in the models of speech production).
Converging evidence in the literature point to a frequency organization of syllable-sized
gestural scores (Cholin, et al., 2006; Laganaro and Alario, 2006) and to an effect of syllable
frequency in the programming difficulties accompanying AoS after stroke (Aichert and
Ziegler, 2004; Steiger and Ziegler, 2008).
One central issue for both theoretical and diagnostic purposes is whether the same
characteristics are observed in AoS after focal lesion and in progressive AoS.
Here we present a single case study aimed at analyzing whether syllable frequency affects
the progressive disruption of speech production in PAoS.
Method
We present a 18 months follow-up study of a 68 year-old men presenting with PAoS. The
patient displayed progressive disruption of speech production, characterized by phonetic
errors, intersyllabic pauses, syllable lengthening and dysprosodia, without positive signs of
aphasia.
Error analyses and word and syllable duration measures were carried out on spontaneous
speech and word repetition at first examination and 18 months later. Across the two sessions
words and syllables with the same syllabic structure and word position were analyzed.
Results
Error rate increase and speech rate decrease were observed in the comparison of the two
assessment periods. More interestingly, syllable duration (1) did no more differentiate
according to its position in the word and (2) correlated negatively with syllable frequency in
the second assessment only (see Figure 1).
86
Figure 1: Duration of CV syllables
Conclusion
Progressive isolated AoS is affected by syllable frequency during the worsening of speech
production. This finding suggests that PAoS progressively displays the same features of
AoS after stroke and that its manifestation can be exploited for theoretical and clinical
investigations of phonetic encoding.
References
Aichert, I., & Ziegler, W. (2004). Syllable frequency and syllable structure in apraxia of speech. Brain
and Language, 88, 148–159.
Cholin, J., Levelt, W. J. M., & Schiller, N. O. (2006). Effects of syllable frequency in speech
production. Cognition, 99, 205-235.
Didic, M., Ceccaldi, M. & Poncet, M. (1998). Progressive Loss of Speech: a Neuropsychological Profile
of Premotor Dysfunction. European Neurology,39, 90–96
Duffy, J.R. (2006). Apraxia of speech in degenerative neurologic disease. Aphasiology, 20, 511–527
Josephs, K.A., Duffy, J.R. et al. (2006). Clinicopathological and imaging correlates of progressive
aphasia and apraxia of speech. Brain, 129, 1385–1398
Laganaro, M. & Alario, F.X. (2006). On the locus of syllable frequency effect. Journal of Memory and
Language, 55, 178-196.
Staiger, A. & Ziegler, W. (2008). Syllable frequency and syllable structure in the spontaneous speech
production of patients with apraxia of speech. Aphasiology, 22,1201-1215.
87
Clinical Use of Auditory Event-Related Potentials (AERPs) in
Children with Speech Output Disorders
Maassen, B.1,2,3, Pasman, J.4
1Medical Psychology, 2Paediatric Neurology, Radboud University Nijmegen Medical
Centre; 3Department of Neurolinguistics, University Groningen
4Department of Clinical Neurophysiology, Radboud University Nijmegen Medical Centre
The Netherlands
Introduction
It has been well documented that auditory processing deficits can play a role in the origin
and maintenance of developmental speech disorders. Both global and specific relations
between audition on the one hand, and speech and language production on the other have
been demonstrated. The objective assessment of auditory processing deficits in young
children with speech and language difficulties, however, is challenging when using auditory
processing tests whereby the child must understand test instructions and give a behavioral
response. For a wide range of specific auditory processing deficits electrophysiological
registrations have been suggested as a valid alternative assessment procedure. The present
study forms an extension of a series of previous studies in our laboratory, showing deviant
auditory event-related potentials (AERPs) in a heterogeneous group of children with speech-
language disorders and in children at risk of developmental dyslexia [1, 2]. Aim of the
present study was to further explore the contribution of auditory event-related potentials
(AERPs) for the clinical diagnosis of children, with a focus in this study on children with
speech output disorders. The diagnostic categories studies were idiopathic Developmental
Apraxia of Speech (DAS), and severe speech motor deficits due to developmental and/or
demonstrated neurological disorders. Most importantly, both these diagnostic categories
were referred because of their speech output difficulties, without clinical indication of
deficits in speech perception. Furthermore, electrophysiological registrations might prove
especially worthwhile for these diagnostic categories, because these children’s severe speech
output disorder pose serious restrictions on the behavioral response that can reliably be
obtained.
Methods
Ten children with Developmental Apraxia of Speech (DAS), and four children with severe
speech motor deficits due to developmental and/or demonstrated neurological disorders (all
between 6 and 9 years of age), were administered an extensive neuropsychological, speech-
language, auditory processing and clinical neurophysiological assessment (including
88
auditory event-related potentials, AERPs). AERPs were compared with those obtained from
normally developing children. The relation between deviant AERP patterns, behavioral
assessments of auditory processing, and phonological and phonetic analyses of speech were
examined.
Two AERP-protocols were administered. In the first protocol children were presented with a
standard tone of 500 Hz (S1: 85% of the presentations) and a deviant tone of 1000 Hz (S2:
15%). In the second protocol the standard stimulus was the word /pop/ (meaning: ‘doll’),
and the deviant was the word /kop/ (meaning ‘cup’). All children were administered the
tone-protocol; to all typically developing children and 11of the 14 children with a speech
disorder also the word-protocol could be administered.
Results
The children with speech disorders and typically developing children showed clear and
reproducible response patterns on the standard stimulus, with a rather stable P1-N2-pattern
at Fz and Cz. However, differences between groups were found with respect to the mismatch
response on the deviant stimuli (mismatch-negativity, MMN). The control group showed a
clearly recognizable MMN with a latency of 250 ms, consisting of a negative field with frontal
maximum combined with a positive field in posterior regions. MMN patterns were similar
for tone and word stimuli. The children with speech disorders demonstrated a deviant MMN
with lower amplitude and for some children longer latency. Furthermore, the spatial
distribution did not follow the anterior-negative and posterior-positive pattern found in the
controls. Of the children with speech disorders, 4 showed normal patterns, 2 showed
scattered and diffuse activity without clear negative or positive fields, and 8 demonstrated a
clearly deviant pattern consisting of a large positive field around the vertex.
Of the children with DAS, 1 had neither a poor score on behavioral tests for APD nor a
deviant MMN, 3 had adequate test scores but strongly deviant MNN, and 6 showed both
deviant test scores and deviant MMN. In two children the speech disorder was due to
operculum syndrome after encephalitis; both had normal APD scores, in one child combined
with normal MMN, in the other combined with deviant MMN. Of the two children with
primarily phonological disorder, one showed deviant APD scores and both had normal
MMN.
Discussion
Results showed an independent contribution of AERPs to the diagnostic characterization of
children with persistent articulation and speech motor difficulties. That is, some of these
children showed deviant AERPs in combination with poor performance on behavioral
assessments of auditory processing, but also dissociated clinical patterns were found
comprising adequate AERPs in combination with poor behavioral scores, or the reverse,
89
deviant AERPs in combination with normal behavioral scores. Most notorious is the positive
field around the vertex in the MMN of some of the children with speech disorders.
Discussion focuses on the functional interpretation of AERPs, possibly reflecting not only
clinical speech perception difficulties but also poor auditory control processes related to
clinical speech production difficulties.
References
[1] Maassen, B., Pasman, J. W., Nijland, L., & Rotteveel, J. J. (2006). Clinical use of AEVP- and AERP-
measures in childhood speech disorders. Clinical Linguistics and Phonetics, 20, 125-134.
[2] Van Leeuwen, T. H., Been, P. H., Kuijpers, C., Zwarts, F., Maassen, B., & Leij van der, A. (2006).
Mismatch response is absent in 2-month-old infants at risk for dyslexia. Neuroreport, 17, 351-355
90
Object clitics and reflexive pronouns in agrammatic Ibero-
Romance
Barcelona
Silvia Martínez-Ferreiro
Universitat Autònoma de Barcelona
The present study aims at characterizing agrammatics’ skills in object clitic and reflexive
pronoun production. To fulfil this aim, 5 Catalan, 5 Galician and 5 Spanish agrammatic
speakers together with 15 matched controls have been tested by means of an elicitation task
with pictures. Findings on typologically different languages indicate high attested rates of
clitic omission in pathological populations (Nespoulous et al. 1988; Miceli et al. 1989;
Reznik, Dubrovsky and Maldonado 1995; Stavrakaki and Kouvava 2003; Chinellato 2004;
Rossi 2007; Nerantzini 2008). This coincides with our results which show significant
differences in the performance of experimental vs. control subjects for both reflexive
pronouns and clitic forms (p < 0.01). Nevertheless, as illustrated in table (1), the former have
been found to be better preserved than the latter (85% correct reflexives vs. 41.54% object
clitics); something already documented in agrammatic comprehension (Grodzinsky, Wexler,
Chien, Marakovitz and Solomon 1993; Edwards and Varlokosta 2007 or Gavarró 2008;
among others).
Table 1: Correct Object clitic and Reflexive pronoun production in Ibero-Romance.
0
20
40
60
Catalan Galician Spanish
Correct Reflexive pronouns and Object clitics in Ibero-Romance
Reflexive pronouns
Object clitics
Our analysis departs from the theoretical framework granted by generative grammar
including the Minimalist Program (Chomsky 1995ss) and Cartographical proposals (Belletti
2002; Cinque 1999, 2002; Rizzi 1997, 2004). Recent studies on cliticizations postulate that
clitics are hosted in a functional projection residing between the CP and the IP-area (Raposo
and Uriagereka 2005; Ledgeway and Lombardi 2005). Hence, under structural
considerations (such as Friedmann and Grodzinsky 1997, 2000 Tree-Pruning Hypothesis),
they are highly susceptible of impairment in agrammatic aphasia. However, both object
91
clitics and reflexive forms have been claimed to share structural position what would render
the dissociation unpredictable. According to Uriagereka (1995), clitics need to be
referentially indexed with a pro (which is not present in the case of reflexive forms). We
explore the possibility that the additional presence of this phonologically null element in
non-reflexive forms is prejudicing their production, thus causing the dissociation.
Selected references:
Belletti, A. (ed.) (2002) Structures and Beyond: The Cartography of Syntactic Structures, Vol. 3,
New York: Oxford University Press.
Chinellato, P. (2004) Disturbi di sintassi nell'afasia non fluente: un'analisi linguistica
dell'agrammatismo italiano e dialettale, Phd. Dissertation, Università di Padova.
Chomsky, N. (1995) The Minimalist Program, Cambridge, MA: The MIT press.
Cinque, G. (1999) Adverbs and Functional Heads: A Cross-linguistic Perspective, Oxford: Oxford
University Press.
Cinque, G. (ed.) (2002) Functional Structure in DP and IP. The Cartography of Syntactic Structures,
Vol. 1, Oxford Studies in Comparative Syntax, Oxford: Oxford University Press.
Edwards, S. and S. Varlokosta (2007) ‘Pronominal and anaphoric reference in agrammatism’, Journal
of Neurolinguistics 20: 423-444.
Friedmann, N. and Y. Grodzinsky (1997) ‘Tense and agreement in agrammatic production: Pruning
the syntactic tree’, Brain and Language 56: 397-425.
Friedmann, N. and Y. Grodzinsky (2000) ‘Split inflection in neurolinguistics’. In M.A. Friedemann
and L. Rizzi (eds.) The Acquisition of Syntax: Studies in Comparative Developmental
Linguistics, New York: Longman, 84-104.
Gavarró, A. (2008) 'Binding and coreference in Catalan agrammatism', The Academy of Aphasia
Meeting, Turku, 20 de octubre.
Grodzinsky, Y.; K. Wexler, Y. C. Chien, S. Marakovitz and J. Solomon (1993) ‘The breakdown of
binding relations’, Brain and Language 45: 396-422.
Ledgeway, A. and A. Lombardi (2005) ‘Verb movement, adverbs, and clitic positions in Romance,’
Probus 17.1: 79-113.
Miceli, G.; C. Silveri, C. Romani and A. Caramazza (1989) ‘Variation in the pattern of omissions and
substitutions of grammatical morphemes in the spontaneous speech of so-called Agrammatic
patients’, Brain and Language 36: 447-492.
Nerantzini, M. (2008) Direct Object clitics in Greek Agrammatic Production: A single case study,
Unpublished MA thesis, University of Groningen, The Netherlands.
Nespoulous, J.L.; M. Dordain, C. Perron, B. Ska, D. Bub, D. Caplan, J. Mehler and A. R. Lecours
(1988) ‘Agrammatism in sentence production without comprehension deficits: Reduced
availability of syntactic structures and/or of grammatical morphemes? A case study’, Brain and
Language 33: 273-295.
Raposo, E. and J. Uriagereka (2005) ‘Clitic Placement in Western Iberian’. In G. Cinque and R. S.
Kayne (eds.) The Oxford Handbook of Comparative Syntax, New York: Oxford University
Press, 639-697.
92
Reznik, M., S. Dubrovsky and S. Maldonado (1995) ‘Agrammatism in Spanish: A Case Study’, Brain
and Language 51: 355-368.
Rizzi, L. (1997) ‘The Fine Structure of the Left Periphery’. In L. Haegeman (ed.) Elements of
Grammar, Dordrecht: Kluwer, 281-337.
Rizzi, L. (ed.) (2004) The structure of CP and IP. The cartography of Syntactic Structures, Vol. 2,
New York: Oxford University Press.
Rossi, E. (2007) Clitic Production in Italian Agrammatism, Phd. Dissertation, University of
Groningen.
Stavrakaki, S. and S. Kouvava (2003) ‘Functional categories in agrammatism: Evidence from Greek’,
Brain and Language 86: 129–141.
Uriagereka, J. (1995) ‘Aspects of the syntax of clitic placement in Western Romance’, Linguistic
Inquiry 26.1: 79-123.
93
Conversational analysis: A comparison between non brain-
damaged geriatric and middle-aged adults
Smita. Caren. Mathias¹, Pratibha. Karanth²
¹Dr.M.V.Shetty College of Speech and Hearing, Mangalore, India. Currently at University of Potsdam,
Berlin, Germany doing masters in clinical linguistics.
²Former Professor at Dr. M.V. Shetty College of Speech and Hearing, Mangalore, India, And Institute
of Speech and Hearing, Bangalore, India. Currently Director of The Com DEALL Trust, Bangalore,
India. (Centre for individuals with communication difficulties)
Previous literature highlights the fact that the majority of individuals with acquired language
disorders belong to the geriatric group. Prior studies within this population
have looked only at the different parameters of language separately, mainly using formal
testing tools. Moreover, these were done on disordered language, overlooking the
importance of normal variations in communication due to ageing. In the present study the
pragmatic strategies used by Indian English speaking non brain-damaged (NBD) geriatric
speakers were compared to those of middle aged speakers using Conversational Analysis
(CA) as a tool. This method gives an overall view of communication skills between two
partners and highlights the strengths and weaknesses of an individual in a natural context.
Although previous studies using this method have focused on language disordered subjects,
it is also necessary to understand these behaviors in normal ageing. This information is
critical in order to the separate the real effects of a disorder from that due to the natural
ageing process. This tool was chosen as it analyzes the sequential, collaborative constructions
within a conversation, specifically looking at what causes breakdowns during a conversation
and what repairs might be negotiated to overcome these trouble spots. Also during
conversations individuals tend to communicate better or least differently, than during formal
testing. Tests do not necessarily predict performance in everyday communication.
Additionally, this tool also gives the clinician information on all the five parameters of
language.
In the current study four normal geriatric persons above eighty years of age, of whom two
were males and two females, served as the experimental group. Similarly, a group of four
middle aged adults (40-50yrs), consisting of two males and two females, served as the
control group. The subjects were matched for age, gender, socio-economic status, education
and linguistic background. It is worth noting that all the subjects are Indian English
94
speakers, which makes this study particularly revealing as no studies in this area have been
done on this dialect, even though it has over ninety million speakers.
Subjects were selected only if they did not have any neurological impairment. For the older
group, The Mini Mental Status Examination Screening given by Folstein et al. (1975) was
used to rule out dementia. The proficiency of the subjects in English was matched using the
Australian second language proficiency rating scale (ASLPR) given by Ingram in 1985. The
data was gathered using a semi-structured interview with questions based on the Western
Aphasic Battery (Kertesz, 1982) and Boston Diagnostic Aphasic Examination (Goodglass &
Kaplan, 1983). The sessions were video recorded in order to take into account gestures and
facial expressions.
The conversations were divided into turns and further into utterances. Then types of trouble
sources, repairs, resolutions, and degree of success were analyzed, based on the definitions
given by Orange et al. (1996). Along with the above parameters, the Trouble source
sequences (TSR) were analyzed quantitatively and descriptively (Springer et al., 1998).
In the second part of the analysis, type token ratio (TTR) was studied. The utterances were
analyzed into open and closed class words based on the definition given by Yule in 2002, and
subsequently into nouns, verbs, adjectives and closed class words. This data was subjected
to statistical analysis using Mann Whitney U-Test, and multiple comparisons within the
groups using the same test with Bonferroni correction.
The results of the analysis show that for geriatric individuals the conversation time required
to produce 100 utterances was on an average much shorter compared to the younger group.
Also, the number of trouble sources and repairs were considerably more, compared to the
younger group.
The trouble sources found most frequently in the geriatric group were lexico-semantic and
discourse in nature. This probably indicates a decline in their cognitive abilities and memory.
Excessive verbosity was also seen, but this cannot be directly attributed to decline in the
linguistic and cognitive resources, since it may be a consequence of the loneliness often
encountered in ageing. The most common type of repair strategy used by this group was
repetition, elaboration and substitution. Substitution was not seen in the younger group.
Among the types of trouble source repair sequences used, the older group used more of the
Self Initiated-Self Repair sequences.
Most of the trouble source repairs found in the younger group was of the type “simple and
unsuccessful” because of the incomplete utterances they used. With the use of gestures, facial
95
expressions and contextual information the interlocutors were successful in comprehending
the information. This is an important difference in the conversation skills of the middle aged
subjects noted in this study. On the other hand, since the older group had clearly defined
trouble sources, most of them were successful in their attempts to repair them, either
initiated by the speaker themselves or by their partners.
The analysis of the type token ratio reveals that only for verbs the older group showed a
larger TTR, but that was not statistically significant. In all other conditions the younger
group scored higher than the older group. The geriatric group showed comparatively less
differences when classes of words were compared with each other.
In conclusion, qualitative changes were found in the communication behaviors of normally
ageing subjects, and these changes must be taken into account when evaluating language
disordered individuals. As a technique CA, even though time-consuming, gives a realistic
picture of the individual’s capacity to communicate, and whether his pragmatic skills are
successful or not. This method can be used to give feedback to the family and client as to
where trouble sources occur and how they can be repaired by the combined efforts of the
listener and the speaker. Future research can be done using CA to compare deficits across
languages in bilingual or multilingual communities. The validity of this preliminary study
could be enhanced if the same conversational aspects were studied across larger groups of
adults and geriatric speakers from different ages, with different language disordered groups.
Furthermore, CA could be used to assess children with language disorders.
96
Mapping language and action in the brain: Evidence from
aphasia and apraxia.
Mengotti P.1, Negri G.A.L.1, Corradi dell’Acqua C. 1, Trincia E.2, Tessari A.3,
Lunardelli A.4, Zadini A.4, Rumiati R.I.1
Sensorimotor theories of language have recently implied that motor representations are
necessary for processing language (e.g. Gallese & Lakoff, 2005). Studying patients suffering
from aphasia or apraxia offers a unique opportunity to test the predictions derived from
these theories. To date only a few studies (e.g. Papagno et al. 1993; Kertesz et al. 1984; De
Renzi et al. 1980) have been published that served this aim for they demonstrated that
aphasia and apraxia can double dissociate. However, these studies did not report an
exhaustive analysis of the lesions that gave rise to aphasia and apraxia.
Here we present data from 61 left-brain damaged patients who performed clinical tests
assessing linguistic (using the Aachener Aphasie Test, Luzzatti et al. 1996) and praxic
(imitation and tool use) abilities. Behavioral analysis at group level showed a significant
positive correlation between scores on linguistic and imitation tests and between scores on
imitation and tool use tests; no reliable correlation was found between scores on linguistic
tasks and tool use task. Single-case analyses showed that apraxic and aphasic deficits
dissociated: six patients were selectively impaired at language processing, whereas five
patients were selectively impaired at imitating and one patient at tool use. The lesion
analysis showed aphasia was associated with regions involving insular cortex and putamen,
and ideomotor apraxia with temporal white matter areas and thalamus. The patient who
showed a selective impairment in tool use had a lesion restricted to the putamen. In order to
establish the brain areas most affected in each deficit (controlling for the other two), a lesion
analysis at group level was performed using Voxel-based Lesion Symptom Mapping
technique (VLSM). Significant results were found for aphasia and ideomotor apraxia:
patients’ performance on linguistic tests was associated with damage of the superior
temporal regions (MNI coordinates x = -55, y = -11, z = -1) extending to the insular cortex (x
= -41, y = 4, z = -5); patients’ performance on the imitation test was associated with damage
of the inferior parietal cortex, including the angular gyrus (x = -37, y = -53, z = 50) and the
surrounding white matter (x = -28, y = -47, z = 41) (see Figure). These results replicate
previous neuropsychological observations by showing that linguistic and motor abilities are
independent and extend our knowledge of their neural correlates.
97
References
De Renzi, E., Motti, F., & Nichelli, P. (1980) Imitating gestures. A quantitative approach to ideomotor
apraxia Archives of Neurology 37, 6-10
Gallese, V., & Lakoff, G. (2005). The brain\\\'s concepts: the role of the sensory-motor system in
conceptual knowledge. Cognitive Neuropsychology 22 (3/4), 455-479.
Kertesz, A., & Ferro, J.M. (1984). Lesion size and location in ideomotor apraxia. Brain 107, 921-933
Luzzatti, C., Willmes, K., & De Bleser, R. (1996) Aachener Aphasie Test: Versione Italiana (Second
edition). Organizzazioni Speciali: Firenze
Papagno, C., Della Sala, S., & Basso, A. (1993). Ideomotor apraxia without aphasia and aphasia
without apraxia: the anatomical support for a double dissociation. Journal of Neurology,
Neurosurgery, and Psychiatry 56, 286-289.
98
Idiom comprehension in aphasic populations
Marcin Morawski
School of English, Adam Mickiewicz University, Poznan, Poland
The aim of the present paper is to provide an insight into the issue of idiom comprehension
in patients who are in the process of recovery from the syndrome of aphasia. Research in
figurative language comprehension has seen a robust development in the recent decades.
However, it has not been until quite recently that psycholinguists began to delve into the
aspect of metaphorical language comprehension in brain damaged populations. It was
observed that even though the ability to produce and understand language is recovered in
the majority of patients with cranial trauma, the impairment of some aspects of
comprehension may protract. The understanding of idioms, metaphors, similes and
proverbs, due to their specific, non-literal character, has been evidenced to pose a serious
problem to aphasic patients, as they fail to decipher the figurative meaning of the utterance,
and, instead, tend to process the message literally (Papagno et al. 2004).
In the present study three patients who suffered from aphasic disorder were tested for
comprehension of idioms by means of two multiple choice tasks. The stimuli for the
experiment comprised a battery of 60 idiomatic phrases in Polish, distributed equally across
six categories of idioms. In the first task the patients were supposed to choose a correct
paraphrase of a given idiom from the three presented options – a figurative paraphrase, a
literal paraphrase and a random sentence. The second task necessitated a choice of a correct
paraphrase of an idiom from three options – a figurative paraphrase, an unrelated
paraphrase containing one of the words from the idiom and a random sentence. Identical
tasks were also administered to three matched controls with no reported language deficits.
The obtained results corroborated the hypothesis that patients who are in the process of
recovery from aphasia encounter various pitfalls in the comprehension of idiomatic
language. Predominantly, they exhibit an inclination to choose the erroneous, literal
paraphrases of the presented idioms over their correct, idiomatic counterparts. This implies
that the intricacy of idiomatic phrases’ processing constitutes an immense difficulty for
patients diagnosed with aphasia. The feasible explanations for such a state of affairs include
the lack of control over suppression mechanisms (Cacciari et al. 2006) or the inability to
access the learned, idiomatic meanings (Papagno et al. 2007). Interestingly, each of the three
patients included in the study exhibited divergent levels of idiom comprehension
impairment. In the paper it is argued that this discrepancy stems from different etiologies
underlying the occurrence of aphasia in each case.
99
References
Cacciari, Cristina – Fabiola Reati – Maria Rosa Colombo – Roberto Padovani – Silvia Rizzo –
Constanza Papagno. 2006. “The comprehension of ambiguous idioms in aphasic patients”,
Neuropsychologia 44: 1305-1314.
Papagno, Constanza – Patrizia Tabossi – Maria Rosa Colombo – Patrizia Zampetti. 2004. “Idiom
comprehension in aphasic patients”, Brain and Language 89: 226-234.
Papagno, Constanza – A. Caporali. 2007. “Testing idiom comprehension in aphasic patients: The
effects of task and idiom type”, Brain and Language 100: 208-220.
100
Lexical and post-lexical deficits in bilingual anomia
Amaia Munarriz & Maria-José Ezeizabarrena
University of the Basque Country
Three major stages in the process of word retrieval are distinguished in traditional cognitive
models (Levelt, Roelofs, & Meyer, 1999): the semantic stage, for conceptual or semantic
information, the lexical stage for the word form corresponding to a certain concept, and
finally the phonological stage, where phonological information is programmed.
Consequently, word retrieval deficits appear in different forms, depending on which stage of
word retrieval is damaged, and three types of anomia are distinguished: semantic anomia,
lexical or pure anomia, and phonological anomia (Laine & Martin, 2006).
In this study we report on the oral production of an early, and highly proficient, Basque-
Spanish bilingual (AF) with lexical and phonological anomia attributable to aphasia. Her
good performance across several tasks suggests target-like linguistic comprehension and
perception at lexical and post-lexical levels. However, several target-deviant features of her
spontaneous oral production reveal some deficit at lexical and post-lexical phonological
levels. The goal of this presentation is to analyse the lexical and post-lexical deficits in order
to provide new evidence regarding the levels of representation of phonological information.
Two kinds of data have been considered in this longitudinal study to assess AF’s production:
1) spontaneous production data and 2) data obtained experimentally through a picture
naming task. Both kinds of data point in the same direction, namely, the bilingual
participant performs better in Spanish, the more frequently used language, than in Basque,
although both languages were acquired at an early stage. As far as lexical performance is
concerned, lower frequency of filler words has been observed in Spanish than in Basque in
AF’s spontaneous speech, and shorter reaction times and lower lexical errors have been
found in Spanish in the picture naming task. These results reveal more efficient lexical access
in the more frequently used language (Sp). With regard to phonological performance, fewer
phonological paraphasias and false beginnings are attested in Spanish both in the
spontaneous speech production and picture naming task. Thus, these results indicate a
better performance in Spanish than in Basque also at the phonological level.
Furthermore, detailed analyses were carried out in order to detect which variables exert an
influence on lexical and post-lexical representations. On the one hand, lexical deficit appears
to be sensitive to lexical frequency, as the frequency of names in picture naming task
101
negatively correlates with RTs and lexical errors in both languages. However, no influence of
neighbourhood density was found; words with high density of phonological neighbours are
not associated with lower RTs and lower lexical error rates. On the other hand, phonological
paraphasias are sensitive to phonological factors such as featural information, markedness
and syllable constituency. Distinctive features are interpreted to be specified at post-lexical
representation, since segments involved in substitutions differ mostly (75% in both
languages) in a single distinctive feature (voice, manner of articulation or place of
articulation). Markedness analysis has shown that a) coronals that are considered unmarked
predominate within added segments; and that b) in omissions, marked syllables (CVC) tend
to become unmarked (CV). In relation to syllable constituency, the vast majority of the
phonological paraphasias (90%) contains target-like metrical properties such as syllable
number and stress pattern.
According to these results, the influence of lexical frequency in lexical deficit indicates that
frequency is represented at the lexical level. Conversely, the influence of featural
information, syllable constituency and typological markedness at phonological deficit
indicates that these factors are represented at post-lexical phonological levels. The lack of
influence of neighbourhood density can be interpreted as induced by the phonological deficit
(cf.Goldrick and Rapp, 2007).
With regard to the bilingual brain, two results have been reported in this study: first, the
different degree of lexical access between Spanish and Basque indicates a more severe lexical
impairment in one language (Basque) than in the other (Spanish). This observation is
consistent with the separation of both lexical systems in the bilingual brain. Besides, this
difference corroborates that one of the main variables in pure anomia is frequency of use
(Laine & Martin, 2006), since AF speaks mostly in Spanish after the brain injury. Secondly,
phonological paraphasias produced in Spanish and Basque follow nearly the same pattern.
The similarity observed in the phonological impairment between both languages suggests
that Spanish and Basque phonological systems may be organized according to similar
governing principles. This result is not surprising since Spanish and Basque were AF’s two
first languages, and similar governing principles usually apply in languages learned under
the same conditions (Ullman, 2001). Moreover, Spanish and Basque phonological systems
are quite similar in terms of syllable structure and phonological inventories, and similarity
within languages of bilinguals has been mentioned as a factor which accounts for similarities
in the organization of these languages (Hernández, Caño, Costa, Sebastián-Gallés,
Juncadella & Gascón, 2008; Santesteban & Costa, 2006).
In summary, this study provides new evidence about the separation of lexical and post-
lexical phonological information in word retrieval processes (Goldrick & Rapp, 2007).
102
Moreover, the fact that Spanish, the more frequently used language, appears to be less
affected as compared to Basque poses new questions about the representation of lexical and
phonological systems in the bilingual brain (Paradis, 2004).
References
Goldrick, M., & Rapp, B. (2007). Lexical and post-lexical phonological representations in spoken
production. Cognition, 102, 219-260.
Hernández, M., Caño, A., Costa, A., Sebastián-Gallés, N., Juncadella, M., & Gascón-Bayarri, J. (2008).
Grammatical category-specific deficits in bilingual aphasia. Brain and Language, 107, 68-80.
Levelt, W. J. M., Roelofs, A., & Meyer, A. S. (1999). A theory of lexical access in speech production.
Behavioral and Brain Sciences, 22, 1-75.
Paradis, M. (2004). A Neurolinguistic Theory of Bilingualism. Amsterdam/Philadelphia: Studies in
Bilingualism.
Santesteban, M., & Costa, A. (2006). Does L1 syntax affect L2 processing. A study with highly
proficient early bilinguals. In B. Fernández & I. Laka (Eds.), Andolin gogoan. Papers in
honour of Professor Eguzkitza (pp. 817-834). Bilbo: EHU-ko argitalpen zerbitzua.
Ullman, M. T. (2001). The neural basis of lexicon and grammar in first and second language: the
declarative/procedural model. Bilingualism: Language and Cognition, 4(1),105-122.
103
Is all morphology morphology: Evidence from therapeutic
intervention
Karin Nault, Harald Baayen, Gary Libben
University of Alberta
Investigations of morphological impairment in aphasia reveal that patients display more
difficulty (i.e., morphological, phonological, and semantic paraphasias) with words that have
internal morphological structure than with simple words (e.g., Nasti & Marangolo, 2005;
Jarema, 2008).
We report on the results of a new therapeutic intervention for individuals with language
impairment after stroke. The Morphological Therapy Protocol (MTP; Nault in preparation) is
a computer-assisted speech therapy that specifically focuses on morphological deficits in the
speech of individuals with aphasia. The MTP is designed such that it addresses recovery of
morphological errors in inflected, derived, and compound words.
A key question in this study was how therapy in one domain may affect performance in other
morphological domains. We were able to address this question as the MTP is administered
with morphologically complex word treatment restricted to each morphological category
being treated in a distinct training module.
This paper reports on the results of four patients that underwent morphological therapy.
Spontaneous naming measurements were taken three times prior to the commencement of
therapy in weekly sessions as well as up to three weeks post therapy in addition to one
measurement at the two-month and one measurement at the three-month post-therapy
point. The MTP is a very short, but intense therapy with twelve treatment sessions of up to 1
hour administered daily over five days per week. During the treatment period, spontaneous
naming measurements were collected daily prior to therapy administration. This study
mainly reports on analyses of the naming measurements taken during therapy.
The results indicate that overall the therapy is effective in the treatment of morphological
deficits (p < 0.001). When analyzing how therapy in one morphological domain affects
performance in other morphological domains, we encountered an intriguing result. Training
derivational morphology (DM) significantly improves the probability of accurately producing
compound words (p < 0.001) and inflected words (p < 0.01); training inflectional
morphology (IM) significantly improves the probability of accurately producing compounds
(p < 0.000) and to a lesser degree, but still significantly, the probability of accurately
104
producing derivations (p < 0.01); and training compounding morphology (CM) significantly
improves the probability of accurately producing inflected words (p < 0.000) and derivations
(p < 0.01). Thus training in any morphological domain always significantly improves the
chances of accurately producing all of the other morphological domains. However,
interestingly, IM training influences most strongly the probability of correctly producing
compound words and conversely, CM training most strongly influences the probability of
correctly producing inflected words. Indeed, the improvement of inflected-word production
of was most defined during CM training.
These results force us to revisit the notion of “morphological deficit.” We discuss this pattern
of findings with reference to models of morphological processing and morphological
representation.
References
Jarema, G. (2008). Impaired morphological processing. In B. Stemmer and H. Whitaker (Eds.)
Handbook of neuroscience of language. Elsevier: Amsterdam.
Nasti, M., & Marangolo, P. (2005). When “macrocefalo (macrocephalous)” is read “minicervello
(minibrain)”: Evidence from a single case study. Brain and Language, 92, 212-218.
Nault, K. (in preparation). Morphological Therapy Protocol. Unpublished doctoral dissertation,
University of Alberta, Canada.
105
An examination of sub-types of semantic errors in word
production in aphasia.
Lyndsey Nickels1, Wendy Best², Lucy Williamson²,
Juliane Burmester³ and Maria Etzien³
1Macquarie University, Sydney; ²University College, London;
³Potsdam University, Germany
Semantic errors are one of the most common types of error produced in aphasic word
production. Coltheart (1980) made the distinction between shared feature errors which
share semantic characteristics with the target (e.g. ‘animal’ or ‘dog’ for ‘cat’) and associative
errors which do not. Associative errors are responses that are related to the target but do not
share semantic characteristics (e.g. desk -> school).
There has been relatively little attention paid in the literature regarding these different
subtypes of semantic errors. This is despite the fact that Coltheart (1980) argued that shared-
feature errors and associative errors may require different explanations. Where work has
been done, it has mostly been to examine the relationship between associative responses
produced by non-aphasic subjects and the semantic errors of people with aphasia (e.g.
Deloche et al., 1996; Huber, 1981; Rinnert and Whitaker, 1973).
More recently, there has been interest in the contrast between semantic errors produced by
those with non-progressive language impairments and individuals with semantic dementia.
Jeffries and Lambon Ralph (2006) found that most individuals with semantic dementia
produced only shared feature (coordinate and superordinate) errors whereas most people
with aphasia following stroke, produced both shared feature and associative errors. Jeffries
and Lambon Ralph interpret the associative errors (e.g. squirrel -> nuts’; lorry -> ‘diesel’) as
“strong but irrelevant associations”, and conclude “it is difficult to account for these errors in
terms of a loss of knowledge: instead, they might have resulted from a failure of controlled
semantic retrieval” (p2143). In contrast, Nickels (1997) argued that the fact that most
individuals produce both error types makes it more likely that the errors have the same
source of impairment. We therefore aimed to investigate this issue further in three case
series of people with aphasia. We first document the frequency of each type of semantic error
in individuals and the group, and investigate the correlations between different types of
semantic error. Then, we investigate the effects of psycholinguistic variables on the
production of shared feature and associative errors, under the hypothesis that if different
variables affect each error type, then this is suggestive of different levels of impairment.
106
Method
Picture naming responses were collected from 3 case series of people with aphasia on three
different sets of black and white line drawings. All three groups were selected on the basis of
having impaired word retrieval and/or production. Group A comprised10 individuals who
were tested on a set of 200 pictures, Group B 21 individuals and 130 pictures, and Group C
16 individuals tested on 118 pictures. For all three groups, the first full naming response for
each item was coded using the same criteria. For those responses where there was a single
word lexical response with a semantic relationship to the target, the responses were coded as
coordinate (within the same semantic category e.g. dog-> cat), superordinate (hypernym e.g.
dog -> animal), subordinate (hyponym, e.g. ) or associative (e.g. dog -> bark; dog -> kennel).
Multiword-semantically-related responses were coded separately but included in the broad
classification of semantic errors. Acceptable alternative responses were not classified as
semantic errors, these included correct subordinates, synonyms, and errors made by two or
more controls.
Results
The percentage of semantic errors produced was similar across the three groups (19%, 16%,
14%; including multiword responses). Table 1 shows the percentages of each type of semantic
error produced. The proportion of coordinate errors is relatively stable across each group,
however, there is more variation in the proportion of associative errors produced.
Data from Groups B & C were examined in further detail, looking at the correlations between
the production of shared feature (coordinate, superordinate and subordinate) and
associative response types. Two consistent factors emerged in the by-subjects analysis: those
individuals with more accurate naming tended to produce a greater proportion of semantic
errors relative to other error type, and there was a tendency for more associative errors to be
produced the more semantic errors were produced, suggesting a link with severity of
impairment.
When examining the patterns across items with groups B & C, once again the
correlations showed some consistent factors: First, items tended to produce more of one
kind of error than the other (there was a significant negative correlation between number of
shared feature errors and number of associative errors). This suggests that perhaps some
property of an item influences whether or not it is more likely to lead to a shared feature or
an associative error.
Second, there was no relationship between a higher proportion of semantic errors and a
higher proportion of either shared feature or associative errors, suggesting that the error
types are not distinguished in terms of difficulty of a particular item.
107
Table 1: Percentage of each type of semantic error produced as a proportion of total
semantic errors.
% of semantic errors
Group
A
Group
B
Group
C
coordinate 25 22 23
superordinate 6 4 10
subordinate 3 2 3
associative 25 18 33
semantic description: specific/close 10 9 15
semantic desription: general/distant 18 3 7
semantic description of a semantically related item 2 2 2
semantic then phonological (single word) 8 15 3
information from episodic memory 1 1 0
semantic other 2 5 3
Finally we examined the effects of four psycholinguistic variables on the production of
shared feature and associative errors using multiple regression: age of acquisition,
imageability, rated word familiarity and number of syllables. For shared feature errors,
Group A showed a significant effect of imageability, and groups B & C no significant effects.
For associative errors, all three groups showed significant effects of age of acquisition.
Further analyses will be reported from analysis of the factors affecting semantic errors for
individual participants and the implications of these results for theoretical accounts of the
origin of semantic errors discussed.
108
Evidence of Different Patterns of Semantic deficit in fluent
and non-fluent Aphasic Patients
Reza Nilipour and Fariba Yadegari
Department of Speech Therapy, University of Rehabilitation Sciences, Tehran
The question concerning the cortical locus of the processing of word meaning has been
addressed in many recent studies. Neuropsychological evidence suggests that the time course
and topography of cortical activation during word processing, in particular during the
processing of words of different categories, are based on functional webs for each category.
Also referents of particular word kind are relevant for determining the brain areas involved
in their processing. Therefore, words with different representational meaning may have
functional webs characterized by different topographies. In particular it has been suggested
that natural words vs. unnatural words or animal vs. man-made words have different
representations. In this paper we are reporting the behavioral responses of fluent and non-
fluent aphasic patients to a Naming Task consisting of different categories (Pulevermuller,
2001 &2002). The main question in this study is weather fluent and non-fluent aphasics with
different lesion sites behave differently in naming different word categories presented to
them in confrontation Naming Task.
Methodology
The subjects in this study are 18 (8 male and 10 female) native speakers of Persian who
became aphasic subsequent to their first CVA incidence. They were all right handed with the
age range of 37 to 80 years (Mean=63.27). They all had la facal left hemisphere lesion with
different sites and their post-onset time range was 1 to 37 months (mean=9 months).
Each subject’s linguistic impairments were initially assessed using the Persian Aphasia Test
(FAT, 1993). A sample of each subject’s connected speech was also collected to determine the
type of aphasic syndrome of each patient. Based on each patient’s general linguistic profile
and the results of the analysis of each patient’s connected speech sample they were grouped
into either fluent or non-fluent aphasic. Accordingly, six of the subjects were diagnosed as
fluent and 12 of them as non-fluent.
In order to investigate the behavioral responses of each group of patients to different word
categories a naming task consisting of a combination of fifty words of different categories
consisting of natural and un-natural objects (natural= 28 un-natural=32) names was
implemented as a confrontation naming task. The confrontation naming task was a line
drawing Picture Naming Test (Nilipour, 2002) developed and normed for elderly normal
109
speakers of Persian. The pictures were presented to each patient in random order in the
confrontation naming task.
Results and Conclusion
The behavioral data obtained from each group of patients were analyzed and compared
between the two groups for natural vs. un-natural and animal vs. man-made categories using
Wilcoxon Signed Rank Test. The results of the analysis indicated that the two groups did not
behave the same in responding to the two categories of objects. When comparing the
behavioral responses of the fluent patients to natural vs. non-natural objects, the data
indicated to a small difference which was not significant (z=.31, p=.753). But the responses
of non-fluent patients to natural vs. unnatural categories were not the same. Their responses
to natural vs. non-natural objects indicated to a significant difference (z=-2=2.062, p=.039).
Another area of significant difference between the behavioral responses of the fluent and
non-fluent subjects was between naming animals vs. man-made objects. The non-fluent
group responses to animals were much better as compared to man-made category and
showed a significant difference (z=-2.040, p=.041. But the fluent aphasic group responses
did not indicate to a significant difference between the two categories (z=-0.314, p=.753).
Based on the obtained data different patterns of semantic deficit and category specificity as a
reflection of different types of lesion in the two group of patients will be looked up and
discussed.
References
Pulevermuller, F. (2002) The Neuroscience of Language: On the Brain Circuits of Words and serial
Order. Cambridge University press.
Pulevermuller, F. (2001) Brain reflections of words and their meaning. Trends in Cognitive Sciences.
5, 517-24.
Nilipour, R. (1987) Farsi Aphasia Test. Iran University of Medical Sciences Press. Tehran, Iran.
Nilipour, R. (2002) Farsi Aphasia Naming Test. University of Rehabilitation Sciences Press. Tehran,
Iran.
110
A Retrospective Study: The characteristics of Patients with
Aphasia at DİLKOM
Sertan Özdemir, Bülent Toğram, Gülce Alev, Emel Aksoy, Mukadder Gülpınar
Anadolu University-DİLKOM, Turkey
The cases with speech and language disorders have been referred to DİLKOM since 2000,
the year the institution was founded. DİLKOM was affiliated to Anadolou University, Turkey
in Eskişehir, Turkey where Masters and Doctorate programs in SLP were initiated. The
Education, Research and Training Center for Speech and Language Disorders (DİLKOM)
together with the Department of Speech and Language Therapy Institute have structured the
programs according to the rules of associations such as ASHA, IALP and CPLOL but giving
more emphasis to research in clinical linguistics. These programs have led about 50 speech
and language pathologists start their professions so far in Turkey.
This presentation aimed to investigate the retrospective information about the adults with
aphasia who have referred to DİLKOM and assessed by Turkish Aphasia Test (ADD) so to get
some demographical characteristics of the cases. The information will cover the frequent
etiology for stroke, the previous and the existing troubles before and after stroke and
aphasia, frequently affected gender and age among the aphasic patients, the types of the
disorder and the most affected component of language, whether their professions
interrupted by aphasia, how they cope with their addictions like smoking and drinking after
stroke and aphasia, etc… .
The study includes 92 cases with aphasia and the retrospective study covers 3 years in
between 2005-2008.
111
Complexity Difficulties in Sentence Comprehension Among
Malay Adults with Aphasia
Razak, RA, A Aziz, MA & Hassan, M
Department of Audiology and Speech Sciences, Faculty of Allied Health Sciences
Universiti Kebangsaan Malaysia, Kuala Lumpur, Malaysia
Complexity in a sentence can be examined at two levels. At sentence level, a sentence is
more complex if the word order is non-canonical (i.e passive sentences) versus canonical (i.e
active sentences). Sentence such as ‘The girl was chased by the boy’ (OSV) is considered
more complex than sentence ‘The boy chased the girl’ (SVO order). The reversibility nature
of verbs adds to complexity. At word level, a verb and its argument structure with its
respective thematic role assignment contribute to complexity. The more argument a verb
has, the more complex it becomes. In this study, complexity in sentence comprehension of
Malay adults with aphasia is explored by looking at both sentence and word levels. Five
normal individuals and five individuals with aphasia were aged, education and L1- matched.
These two groups of subjects were given sentence–picture matching tasks. Results showed
that individuals with aphasia have poorer understanding of active and passive reversible
than that of normals (p < 0.05). Understanding of reversible passive sentences among
subjects with aphasia is found to be lower than that of active sentences (p < 0.05). It was
observed that increasingly complex argument structure of verbs affect comprehension level
in individuals with aphasia compared to their normal counterparts (p < 0.05). In addition,
based on Bonferroni post – hoc analysis, a sub – test of ANOVA, there are significant
differences between mean score of sentences with one, two and three arguments when
compared to sentences containing four arguments.
Key words : complexity difficulties, sentence comprehension, Malays adults with aphasia
112
Grammar learning by Borca’s agrammatic aphasics (a pilot
study)
Daria Schmidt
Joint European Masters in Clinical Linguistics, University of Potsdam, Germany, University
of Groningen, the Netherlands & University of Milano-Bicocca, Italy
e-mail: [email protected]
Introduction
In severe agrammatic Broca’s aphasia sensitivity to syntactically well-formed sentences
seems to be lost. Syntax involves the application of rules. The artificial grammar learning
(AGL) task is a useful paradigm for testing rule formation and application [1]. Previous
studies have shown that two specific artificial grammar types, namely FSG and PSG, are
processed differently, i.e. by distinct brain structures [2, 3]. The left Broca’s area is active in
long-distance dependency processing of PSG strings whereas the left posterior frontal
operculum subserves the processing of local, dependencies in FSG strings. This pilot study
investigates the grammaticality judgment performance of two aphasic patients when
learning FSG and PSG strings. We expected lower accuracy scores in the aphasic participants
for the PSG rules as they involve non-adjacent dependencies.
Methods
Participants were tested with a familiarization-discrimination paradigm in two sessions.
Similar to previous studies [3, 4, 5] sequences of consonant-vowel nonsense syllables (e.g., li
ra ti ka) synthesized with the MBROLA speech synthesizer were presented auditory to the
participants in two sessions using the presentation software DMDX. Each session included a
3-minute familiarization phase either with FSG or PSG strings and an 8-minute testing
phase with both string types. Responses were given via button-press and accuracy and
reaction times were recorded. Performance was analyzed using the d-prime and Xc-bias
measures of signal detection theory.
Results
Both aphasic participants were tested with PSG in the first session and the FSG in
the second session. Both participants showed higher accuracy scores in the PSG session
compared to the FSG session, although results for all sessions are at chance (65% vs. 44%
and 54% vs. 50%). So there is a tendency that they were better able to learn the PSG rules.
113
Even in the second session with the FSG familiarisation the participants favoured the PSG
strings over the FSG strings.
Conclusions
Despite low performance in the task on the whole, contrary to our hypothesis the aphasic
participants did not perform worse on PSG than on FSG, although the former type of
grammar seems to involve more complex syntactic operations. We speculated that if patients
suffer from syntactic impairment (what agrammatism seems to be) then, they would need
recruitment of extra resources for processing of this grammar. The study conducted is a pilot
study, the further research in the future with a bigger data set and an inclusion of an elderly
group of control subjects will give a more reliable data set. Although no conclusions could be
made on the basis of this pilot study, the results however demonstrated that aphasic patients
had generally higher scores with the PSG familiarization in comparison to the FSG on
various parameters, such as general acceptance of strings, percentage of correct responses
overall and correct yes and no responses.
References
1. Reber, A.S. (1967). Implicit learning of artificial grammars. Journal of Verbal Learning and
Verbal Behaviour, 6, 855-863.
2. Friederici, A. D., Bahlmann, J., Heim, S., Schubotz, R. I.,& Anwander, A. (2006). The brain
differentiates human and non-human grammars: Functional localization and structural
connectivity. Proceedings of the National Academy of Sciences of the United States of
America, 103, 2458–2463.
3. Friederici, Angela D. (2004): 'Processing local transitions versus long-distance syntactic
hierarchies'. In: Trends in Cognitive Sciences Vol. 8, No.6, and pp. 245_247.
4. Fitch, W. T., & Hauser, M. D. (2004). Computational constraints on syntactic processing in a
nonhuman primate. Science, 303, 377–380.
5. Fritzsche, T., Höhle, B., & Fitch, W. T. (2007). Vowels as category-cue in artificial grammar
processing, Paper presented at the International Linguistic Association 52nd Annual
Conference. Hunter College, CUNY, NY.
114
Spatial cognition and language in aphasia: crosslinguistic
perspectives
Efstathia Soroli1, Hélène Boudre2, Maya Hickmann1, Jean-Luc Nespoulous3, Thi
Mai Tran2
1Laboratoire Structures Formelles du Langage, CNRS & Université de Paris 8
2Laboratoire STL, CNRS & Université Lille Nord de France
3Laboratoire Jacques Lordat, Université de Toulouse Le Mirail
Despite a few crosslinguistic studies of aphasia (Menn & Obler, 1990; Goodglass & Gleason,
1991; Nespoulous, 1999), little is still known about universal vs. language-specific aspects of
patients’ linguistic deficits and compensatory strategies. The present research addresses this
question in the domain of spatial cognition, with particular attention to how aphasic patients
vs. control subjects represent motion events.
Languages differ in how they lexicalize and grammaticalize spatial information (Talmy,
2000). Recent psycholinguistic research suggests that such properties constrain how adults
and children organize spatial information to encode motion in discourse (Choi & Bowerman,
1991; Hickmann, et al. 2008; Slobin, 2004), thereby raising new questions concerning the
relation between language and cognition. With respect to the expression of motion events,
Germanic languages (satellite-framed) such as English express manner in the verb root and
path in satellites, while Romance languages (verb-framed) such as French lexicalize path in
the verb, leaving manner implicit or expressing it in the periphery of the sentence (eg. She
ran across the street vs. Elle a traversé la rue [en courant] Lit. ‘She crossed the road
running’). Such striking differences are of great relevance for the study of aphasic patients
who typically present dissociations between lexical and grammatical capacities and deficits.
Our experiment examined the impact of such typological constraints on how two French
aphasic patients (agrammatic vs. anomic) represented motion events in comparison to
control groups of neurologically intact French and English speakers in four controlled tasks:
non-verbal categorization, verbal categorization, production, and comprehension. In the two
categorization tasks subjects were shown triads of visual stimuli (videos or cartoons)
implicating motion events that varied with respect either to Path or to Manner and they had
to group them together relying on either type of information as criterion. Non-verbal
categorization involved triads of visual stimuli (one target, two choices), while verbal
115
categorization combined a linguistic target (a sentence describing the event) with two visual
stimuli. In both cases subjects’ choices and reaction times were measured. In the production
task subjects described motion events that were presented visually. The analysis examined
which of these two types of information were expressed (focus), by what verbal means (locus:
verbs, particles, prepositions, sentence structures), how much information was expressed
(density), and the particular strategies on which patients relied to compensate their deficits.
In the comprehension task both accuracy and reaction times were measured as subjects
selected one of two visual stimuli that correctly corresponded to a sentence they had heard.
French and English speakers were expected to show different performances as a function of
the typological properties of their language. English speakers should equally rely on Manner
and on Path information, given the compact structures provided by their language. In
contrast, French speakers should rely on Path information which is lexicalised in the verb
and constitutes the main linguistic information in focus. According to the weak hypothesis, it
was expected that this effect should only occur in verbal tasks (verbal categorization,
production) while the strong hypothesis predicted that a language effect should also occur in
tasks that did not involve language (non-verbal categorization).
Finally, with respect to the French aphasic patients, the following predictions were made: 1)
their deficit should affect their production, but not necessarily their comprehension; 2) their
categorization performance should depend on their language and on their particular deficit,
i.e. more path focus for the anomic and more manner focus for the agrammatic patient than
the control group in the same language; 3) their deficit should lead to particular difficulties
and strategies, resulting in different productions between the two patients (searching verbs
for the anomic patient, lack of some relational devices such as prepositions for the
agrammatic patient).
The results show that English control speakers used both Manner and Path to categorize
events following the prototypical structure of their language. In contrast, French controls
used Manner to categorize events significantly less often than English controls. Similar
results were found in the verbal categorization task despite an increase in the saliency of
Manner information in both languages. In the production task English speakers typically
encoded Manner in verbs together with Path in other devices within compact structures,
whereas French speakers focused on Path and mainly encoded this information in verbs.
When categorizing motion events, the agrammatic patient seemed to follow a compensatory
strategy that lead him to adopt an English-like pattern consisting of choosing equally
manner and path. In contrast, the anomic patient showed a strong preference for path-
directed choices, focusing on the most accessible information typically encoded by the verb
116
in his mother tongue. In the production task aphasics produced utterances of lower density
in comparison to controls. The comprehension task showed that patients had no
comprehension deficit, although their performances were slower than those of controls.
The present crosslinguistic experimental study suggests that typological properties of
languages can generate different ways of expressing motion events, as well as different
categorical choices, even in non-linguistic tasks. Aphasics seem to develop strategies that are
dependent both on the spatial system of their language and on their specific deficit. Further
data collection in progress will complement these case studies in a larger crosslinguistic
study that will include control subjects of other languages and aphasic patients in the same
language groups.
Choi, S. & Bowerman, M. (1991). Learning to express motion events in English and Korean: The
influence of language-specific lexicalization patterns. Cognition 41: 83-121.
Goodglass, H., Gleason, JB. (1991). Introduction. Brain and language 41 (2; Special Issue:
Crosslinguistic Studies of Aphasia): 121-122.
Hickmann, M., Tarrane, P. & Bonnet, Ph. (2008). Motion in first language acquisition: manner
and path in French and in English. Journal of Child Language (December 2008).
Menn, L. & Obler L.K. (1990). Language data and theories of agrammatism, in : L. Menn & L.K.
Obler, (Eds.), Agrammatic Aphasia. A cross language narrative sourcebook, Amsterdam: John
Benjamins Publishing Company.
Nespoulous, J-L. (1999). Universal vs language-specific constraints in agrammatic aphasia, in C.
Fuchs & S. Robert (eds.) Language diversity and cognitive representations, John Benjamins, 195-
207.
Slobin, D. (2004). The many ways to search for a frog: linguistic typology & the expression of
motion events. In S. Strömqvist & L. Verhoeven eds. Relating Events in Narrative. Vol 2, 219-257.
Mahwah, NJ: LEA.
Talmy, L. (2000). Toward a cognitive semantics. Volume 1: Concept structuring systems. Volume
2: Typology and process in concept structuring. Cambridge, MA: MIT Press.
KEY WORDS: Space, motion, language, aphasia
117
Assessing the Use of Speech Acts in Participants with
Aphasia through ADD
Bülent Toğram, İlknur Maviş, İlim Aksu, Ayşen Yargan
Anadolou University, Turkey
Center for Speech and Language Disorders (DİLKOM)
Introduction
Communicative competence implies knowledge of how to converse with different partners
and in different contexts. It is knowledge of who can say what to whom, in what way, where
and when and by what means (Prutting and Kirchner,1987). Pragmatics involves the
acquisition and use of such conversational knowledge and of the semantic rules necessary to
communicate intent within the context of a speech act, a theoretical unit of communication
between a speaker and a hearer (Lucas, 1980). Speech acts include making promises,
statements, requests, assertions, and so forth. Thus, pragmatics involves the interactional
aspects of communication including sensitivity to various aspects of social contexts; it is
simply an analysis of the use of language for communication.
Researchers have investigated pragmatic performance in individuals with aphasia (Avent &
Wertz, 1996; Holland, 1982; Prutting & Kirchner, 1987; Wilcox & Davis, 1977) and data
suggest that individuals with aphasia maintain a high level of pragmatic appropriateness
despite their linguistic impairments (Avent & Wertz, 1996; Prutting & Kirchner, 1987).
Method
Participants of the study were a total of 282 neurologically normal adults and 92 adults with
aphasia subsequent to a first, single left-hemisphere stroke. Participants were administered
the Pragmatic sub-test in the ADD (Maviş & Toğram,2009). The sub test measures 10
pragmatic aspects of language use in which each correct reply was scored by 2 points. The
research statements aimed to elicit what the participants say in given situations such as what
they will say to an ill person on a visit day, what they will say to the one who sneezes, what
they will say to a neighbor when they meet in the morning, etc. The purpose of the study was
to evaluate whether pragmatic performance and functional communication were
significantly related with language impairment, age, gender, and/or education of the
participants. The results were studied in detail.
118
References
Prutting,C.A. and Kirchner, D.M. 1987. Clinical appraisal of the pragmatic aspects of language.
Journal of Speech and Hearing Disorders, 52, 105-119,
Lucas,E. 1980. Semantic And pragmatic language disorders. Assessment and remediation.
Rockville,MD. : Aspen.
Maviş & Toğram,2009. Afazide Dil Değerlendirme Testi (ADD). Ankara:Detay yayıncılık
119
“Frog where are you?” Narration Analysis in Turkish
Participants with and without Aphasia
Müge Tunçer, Özlem Ünal, Aysun Şan, İlknur Maviş
Anadolu University, Turkey
Center for Speech and Language Therapy (DİLKOM)
Introduction
Narrative is a discourse type in which the speaker sequences events in a relationship between
them, using a set of grammatical rules and conveying the meaning. Narratives may be oral or
written and be affected by age, gender, culture, time and context. Assessing and describing
the narrative discourse is an important component in aphasia research and intervention. It
has been frequently used in the studies of children’s language development, but the use in
the field of aphasia is recent.
Assessment of narrative discourse is important for aphasic individuals, because production
of narratives is necessary for relaying the relationships between events and characters in
everyday life, and because these patients are impaired in narrative discourse. Narratives are
used for assessment at pre and post treatment, are intended to reveal aphasic patient’s
language characteristics and identify impaired abilities for treatment goals. Studies on
narrative production of individuals with aphasia have reported disruption in producing
sequences of events and actions and a reduction of language complexity at both sentence and
discourse level (Foka-Kavalieraki et al., 2008; Armstrong, 2000).
Method
Ten adults who experienced aphasia and ten healthy individuals, matched for gender and
age, participated in the study. Narrative productions were elicited by the wordless picture
story, “Frog, Where Are You?” (Mayer, 1969). Data will be analyzed for semantic cohesion
and coherence in narrative discourse. In this purpose, we investigated T units, complex T
units, reference items, lexical selections (repetitions, synonyms, superordinates, collocatives,
vs.), conjunctions, substitutions (nominal, verbal, causal), and ellipsis as mentioned by
Chapey (1994).
References
Foka-Kavalieraki, P., Kakavoulia, M., Economou1, A., Varlokosta, S., Routsis, C., Kasselimis, D.,
Potagas, C., Evdokimidis, I. & Protopapas, A. (2008). A comprehensive approach to the analysis of
120
narrative discourse production by Greek speakers with aphasia. Science of Aphasia IX – Chalkidiki,
Greece, 20-25 September, 2008.
Armstrong, E. (2000). Review: Aphasic discourse analysis: The story so far. Aphasiology, 14, 875-
892.
Mayer, M. (1969). Frog, where are you? New York: Dial Books for Young Readers.
Chapey, R. (1994).The assessment of language disorders in adults. In Language Intervention
Strategies in Aphasia and Related Neurogenic Communication Disorders. Third Edition. Ed. Roberta
Chapey. Marryland, USA.
121
Lexical therapy for apraxia of speech
Varley, R.A., Cowell, P.E., Dyson. L., Roper, A.H., Inglis, A.L., Whiteside, S.P.
Department of Human Communication Sciences, The University of Sheffield
[email protected], (44 114) 2222449;
31 Claremont Crescent, Sheffield, S10 2TA
Research Question
The underlying impairment in apraxia of speech (AOS) remains poorly understood. The view
that AOS is a post-lexical impairment motivates micro-structural therapies that train
articulatory gestures and then provide practise in combining gestures to form cohesive
syllables. Although there is evidence that people with apraxia are capable of developing an
increased repertoire of articulatory gestures under such therapy regimes, there is less
evidence that such approaches generalise from treated syllables (Wambaugh, 2004). An
alternative view of AOS is that it represents an impairment in activating consolidated motor
plans for higher frequency utterances. Such a view motivates therapies that train words and
other utterances. We report preliminary results for the effectiveness of a computerised
intervention for AOS, based on whole-word therapy. The research employed a two-period
cross-over design in which participants completed a speech program, but also a non-speech
sham intervention. In addition, the treatment employed a set of principles that aimed to
enhance motor learning. These included intensive sensory-perceptual stimulation prior to
production in order to activate mirror neurone networks; the use of error–reducing
strategies to encourage the practice of fluent behaviour; and increasing treatment intensity
by allowing participants to self-administer via use of a software program.
Design
50 people with AOS and 20 people with aphasia but no AOS will be recruited to the study. All
will be six months post-stroke. Participants received two therapies. Both were self-
administered via computer and each lasts for six weeks, with a four week break in between
programs. One therapy was a speech program, and the other was a sham visual-perceptual
training program. Participants were randomly assigned to speech or a sham visuo-perceptual
first. Two baseline measures were taken prior to therapy. Outcomes were measured after
each therapy and eight weeks after the final therapy in order to determine maintenance of
any behavioural gains. Behavioural change was measured on three groups of words: (1)
122
treated items; (2) phonetically- and frequency-matched untreated controls; (3) phonetically-
distinct but frequency-matched untreated controls. In addition, samples of connected speech
were collected before and after treatment to determine generalisation beyond single word
level. Finally, performance on untreated control behaviours (written word-picture matching
and spoken sentence-picture matched) was determined.
Results
We report the results from a pilot study intervention of six participants with apraxia of
speech. The results revealed positive effects on speech fluency, in particular on word
duration. Decreases in word duration followed the speech program for treated and matched
items. These effects were observed for both patient sub-groups regardless of whether speech
treatment was administered before or after sham intervention. Behavioural gains were
maintained up to 8-18 weeks after the withdrawal of the speech program, although
improvement appeared fragile for some of the more severely affected patients. The sham
intervention resulted in no consistent pattern of change in speech behaviours as measured
by duration, despite similar levels of usage to the speech program. There was no evidence of
change on the untreated control behaviours.
Discussion
Self-administered therapy via a software program appears to be a viable option in increasing
treatment intensity for post-stroke speech and language impairments. Similarly, preliminary
data indicate that lexical therapy for apraxia of speech is a useful therapeutic option
particularly as it targets real words and functionally useful vocabulary rather than isolated
articulatory gestures; however is not yet clear whether the therapy outcome is due to the
reduction of apraxic impairment or increased activation of underlying lexical networks.
Subsequent study of anomic but non-apraxic individuals will help to clarify this issue. The
results of the treatment study cannot be used in support of claims regarding the underlying
impairment in apraxia of speech. The positive effects found in this pilot study might, for
example, be due to increased intensity of practice. Segment gestural approaches that
achieved similar treatment intensity might also find equivalent outcomes.
References
Wambaugh, J. & Nessler, C. (2004). Modification of sound production treatment for apraxia of
speech: Acquisition and generalisation effects. Aphasiology, 18 (5/6/7), 407-427.
123
Tense/Aspect Category in Fluent and Nonfluent German
Aphasia
Roma Wieczorek4, Walter Huber , Robert Darkow
Neurolinguistics, Department of Neurology, University Hospital, RWTH Aachen
University
Objectives & Background
The aim of this study is to describe the linguistic performance regarding sentence
comprehension and production of tense/aspect category from the morphological and
semantic point of view, in fluent and non-fluent German aphasia. As the category of aspect is
not overtly marked (Musan, 2001), but rather reflected by means of tense morphology in
German (Van Hout et al., 2005 ; Musan, 2001), the difference between the ongoing action
(Präsens→Imperfektiv) and the completed past action (Perfekt→Perfektiv) is tested (Wischer
& Habermann, 2004).
The examples illustrating the tense/aspect contrast in the German language:
� Sie malt das Bild. [imperfective (or neutral) Aspect / Present Tense]
� Sie hat das Bild gemalt [perfective Aspect / Present Perfect Tense]
Special interest is taken in observing any differences or similarities in verbal processing
between fluent and non-fluent aphasia, since in majority of cases the data of non-fluent
aphasia only are available in the literature and the debate upon the potential
morphosyntactic or morphosemantic nature of tense / aspect deficits tends to remain open
(Faroqi-Shai & Dickey, 2009).
Methods
The following four tasks were used in the experiment:
- Auditory comprehension task, based on the proper choice of the picture from the four
provided
- Sentence completion task , with gaps to be filled in with the correctly inflected form
- Oral production task , focusing only on the imperfective / perfective contrast
- Semantic-ordering task, with pictures depicting future, present and past situations
4 Contact person : Roma Wieczorek, Abt. Neurolinguistik, Neurologische Klinik, Universitätsklinikum der RWTH Aachen, Pauwelsstr. 30, D-52074 Aachen, E-mail: [email protected]
124
The task design comprised of 38 pictures and aimed at showing the tense/aspect contrast
with the use of specific morphosyntactic structures. The pictures presented were
randomised.
The last non-linguistic task functioned as a cognitive control tool, where the patients were
asked to put the future, present and past activities in the proper order.
All tasks were carried out with the control group and no significant linguistic error patterns
were noted.
Results
Seven German (one with global, three with Broca’s , one with Wernicke’s, and two with
anomic aphasia) patients were tested. The significant difference regarding correct aspect
application was observed in the comprehension and two production tasks in the case of the
two (Broca’s and Wernicke’s) patients with chronic, moderate aphasia. On the basis of error
analysis it was shown that the Broca’s aphasic patient comitted mostly the errors of a
morphological origin in the two production tasks. As expected, it was noted that the patient
found it particularly difficult to provide the perfective target items in the oral production task
(p = .000) .The Wernicke’s aphasic patient, on the other hand, regularly confused aspect in
both production tasks as well as in the comprehension one. Contrary to Broca’s aphasic
individual, the Wernicke’s patient made almost no morphological errors in the
corresponding production tasks.
Discussion
We believe that “aspect mixed up” phenomenon cannot be explained only in terms of the
complexity of German verbal morphology, since also the imperfective present forms were
replaced by the morphologically more complex perfective items. On the other hand, the non-
fluent patient who commited morphological errors, did not seem to have problems with the
proper choice of aspect. Finally, the observed problem with aspect differentiation in the case
of the non-fluent patient does not seem to be rooted in some cognitive impairments, as all
the patients (including the global aphasic one) performed relatively well in the control non-
linguistic task.
References
Faroqi-Shah, Y. & Dickey, M.W. 2009. On-line processing of tense and temporality in agrammatic
aphasia. Brain & Language, 108, pp.97-111
Musan, R. 2001. The Present Perfect in German: outline of its semantic composition. In Natural
Language & Linguistic Theory 19 ; pp.355-401
Van Hout, A., de Swart, H., Verkuyl, J.H. 2005. Introducing Perspectives on Aspect. In J.H. Verkuyl,
de Swart, H. & van Hout, A. (Eds.). Perspectives on Aspect. the Netherlands : Springer
125
Wischer, I. & Habermann, M. 2004. Der Gebrauch von Präfixverben zum Ausdruck von
Aspekt/Aktionsart im Altenglischen und Althochdeutschen. In ZGL 32 , pp. 262-285
126
Sentence Comprehension in Turkish Broca’s Aphasia: An
integration problem
Tuba Yarbay Dumana, Nermin Altınokb, Neşe Özgirginb, and Roelien Bastiaansea
a Department of Linguistics, Graduate School for Behavioral and Cognitive
Neurosciences (BCN), University of Groningen, The Netherlands
b Ankara Physical Medicine and Rehabilitation Center, Turkey
Introduction
Turkish agrammatic speakers with Broca’s aphasia have more problems producing sentences
in derived order (object scrambling, subject relatives and object relatives) than sentences in
base order (Yarbay Duman et al., 2007, 2008). Word order errors and morphological errors
are produced in sentences in derived order. Word order and morphology are closely related
in Turkish. For example, in object relative sentences, the word order changes but so does the
case of the subject: it changes from nominative to genitive. The data of Yarbay Duman et al.
(2008) on the production of clauses with base and derived word order show that when
Turkish agrammatic speakers fail to produce the right order, they adapt case to the order
they produce. Apparently, there is an interaction between word order and case that Turkish
agrammatic speakers are aware of when producing sentences. The question that arises is
whether an interaction between word order and case which is similar to the one found in
production can be observed in sentence comprehension. Therefore, the interaction between
word order and case in sentence comprehension is investigated.
Word order problems are a key feature of sentence comprehension in Broca’s aphasia for
languages without overt case marking (e.g. Grodzinsky, 1995; Bastiaanse & Edwards, 2003).
In this study, patients with Broca’s aphasia were tested with a comprehension test where
sentences with base and derived order varied in their use of case. Both sentences with base
case assignment, that is subject = nominative and object = accusative (base order clauses,
clauses with object scrambling and subject relatives) and sentences with non-base case
assignment (object relatives: subject = genitive; object = nominative and passives = two
nominatives) were included to evaluate the interaction between word order and case (see
Table 1A for a summary and Table 1B for specific examples of sentence types). Only a few
sentence comprehension studies have been performed in languages where case information
is decisive (Turkish: McWhinney et al., 1991; German: Burchert et al., 2003). The results
show that word order influences sentence comprehension and the influence of case is less
decisive.
127
Table 1A. Summary of sentence types: thematic structure, movement (in italics) and
case morphology
Table 1B: Sentence Types and Examples
(1) Base Order
kadın-laragent adam-ıtheme kurtar-ıyor
women-plr/nom man-acc save-asp/3s agr
‘the women save the man’
(2)Object
Scrambling
adam-ıi-theme kadın-laragent ti kurtar-ıyor
man-acc women-plr/nom save-asp/3s agr
‘the man, the women save’
(3)Subject relatives
ti adam-ı theme kurtar-an kadın-lariagent (burada)
man-acc save-asp women-plr/nom (are here)
‘the women who save the man (are here)’
(4) Object relatives
kadınlar-ın agent ti kurtar-dığ-ı adami-theme (burada)
the women-plr/gen save-asp-agr the man-nom (are here)
‘the man whom the women save (are here)’
(5) Passives adam theme kadın-laragent tarafından kurtar-ıl-ıyor
the man-nom women-plr/nom by save-pass-asp/3s agr
‘the man is saved by the women’
An element that moves leaves a trace (t) behind, which is co-indexed (i) with the antecedent in 2, 3
and 4.
Thematic Structure Case Agent Case Theme
Base Order (BO) Agent-Theme nominative accusative
Scrambling (SCR) Theme-Agent nominative accusative
Subject Relative (SR) [NPTheme-Agent] nominative accusative
Object Relative (OR) [NPAgent-Theme] genitive nominative
Passives (PASS) Theme-Agent nominative nominative
128
Methods
Subjects
Seventeen Turkish speakers with Broca’s aphasia (seven female; mean-age: 49.8; right-
handed) were tested. 15 were with a left CVA and the others with a brain-trauma and brain-
tumor). All (except one) had right-hemiplegia and were at least four months post-onset. The
aphasia type was established with the Gülhane Aphasia Test (Tanrıdağ, 1993) and confirmed
by the speech therapist on the basis of the spontaneous speech. Ten native speakers of
Turkish with no language impairment history participated (and performed at ceiling) on the
test.
Materials
A spoken-sentence-to-picture-matching task was developed with five semantically reversible
sentence types (20 items in each).
The participants were asked to point to one of the four pictures that matched the spoken
sentence: (1) target picture; (2) picture with role reversals; (3) picture in which roles are
correct, but the number of agents/themes was wrong (hereafter, number); (4) picture in
which both the roles and the number of agents/themes were incorrect (hereafter, role
reversal + number). For the sentence ‘the women save the man’ the target is at the left upper
corner (see Figure 1).
Results
The Repeated Measures ANOVA showed a statistically significant effect for order: F(1.16) =
19.3, p<.000. The comprehension of sentences with agent-theme order (BO + OR) was better
than sentences with theme-agent order (SCR-SR + PASS). There was a statistically
significant main effect of case as well: F(1.16) = 192, p<.000. The comprehension of
sentences with base case information (BO + SCR-SR) was better than sentences in which this
type of information was absent (OR + PASS).
There was a significant interaction between order and case: F(1.16) = 5.75, p<.029. Although
there was an effect of order when two clauses had normal case information (F(1.16) = 18.1,
129
p<.046), this effect was lost when the normal case information was absent (F(1.16) = 3.13,
p<.096). A sentence in agent-theme order (BO) was easier to comprehend than a sentence in
theme-agent order (SCR-SR) only when both have base case information. However, when
base case information was lacking, the sentences in agent-theme order (OR) were as difficult
as the sentences in theme-agent order (PASS). Consequently, the following order of difficulty
was observed:
Base Order > Object Scrambling = Subject Relative > Object relative = Passive. Qualitatively,
the patients predominantly made role reversal errors.
Discussion
The interaction between case and word order shaped three performance levels. (1) A clause is
comprehended best when there is both base case and base word order information (base
order clauses). (2) The performance drops if there is base case information but word order is
derived (object scrambling and subject relatives). (3) A clause is comprehended the poorest
when there is neither base case information nor base word order (object relatives and
passives). Apparently, there is an interaction between case and word order information.
When either of these is not in the base form, comprehension diminishes. When neither is in
the base form, performance drops even further.
These findings complement earlier findings on speech production that show more difficulties
with derived word order (Yarbay Duman et al., 2007, 2008). Furthermore, Yarbay Duman et
al. (2008) shows that object relative clauses are more problematic for some patients than
subject relative clauses: the order is derived and case is non-basic in object relatives. When
either word order or case information is non-basic, problems arise. In this comprehension
test, it becomes unclear who the agent is and who the theme is. In production, sentences
with derived order and non-base case are difficult to produce and Broca patients tend to
produce sentences in base order and base case. Overall, if information from different sources
(e.g. syntax, morphology) needs to be integrated, both production and comprehension
decreases. For this study, that means that integrating word order information and case
information to assign thematic roles to the NPs is difficult. Previous studies show that
integrating information from other levels can also be problematic (see Yarbay Duman &
Bastiaanse, 2009 for problems in integrating semantic (time-reference) information into
(morpho)syntax (verb inflection)).
130
References
Bastiaanse, R. & Edwards, S. (2004). Word order and finiteness in Dutch and English Broca’s and
Wernicke’s aphasia. Brain and Language, 89, 91–107.
Burchert, F., De Bleser, R., & Sonntag, K. (2003). Does morphology make the difference? Agrammatic
sentence comprehension in German. Brain and Language, 87, 323–342.
Grodzinsky, Y. (1995). A restrictive theory of trace deletion in agrammatism. Brain and Language, 50,
27–51.
MacWhinney, B., Osmán-Sági, J., & Slobin, D.I. (1991). Sentence comprehension in aphasia in two
clear case marking languages. Brain and Language, 41, 234–249.
Tanrıdağ O. (1993). Afazi. Ankara: GATA Basımevi.
Yarbay Duman, T. & Bastiaanse, R. (2009). Time reference through verb inflection in Turkish
agrammatic aphasia. Brain and Language,108, 30 –39.
Yarbay Duman, T., Aygen, G., & Bastiaanse, R. (2008). The production of Turkish relative clauses in
agrammatism: Verb inflection and constituent order. Brain and Language, 105, 149–160.
Yarbay Duman, T., Aygen, G., Özgirgin, N., & Bastiaanse, R. (2007). Object scrambling and finiteness
in Turkish agrammatic production. Journal of Neurolinguistics, 20, 306-331.