NEUROCASE2012, iFirst, 1–16

A case of hyperthymesia: rethinking the role of theamygdala in autobiographical memory

Brandon A. Ally1,2,3, Erin P. Hussey1, and Manus J. Donahue2,4,5

1Department of Neurology, Vanderbilt University, Nashville, TN, USA2Department of Psychiatry, Vanderbilt University, Nashville, TN, USA3Department of Psychology, Vanderbilt University, Nashville, TN, USA4Department of Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA5Department of Physics and Astronomy, Vanderbilt University, Nashville, TN, USA

Much controversy has been focused on the extent to which the amygdala belongs to the autobiographical memory(AM) core network. Early evidence suggested the amygdala played a vital role in emotional processing, likelyhelping to encode emotionally charged stimuli. However, recent work has highlighted the amygdala’s role in socialand self-referential processing, leading to speculation that the amygdala likely supports the encoding and retrievalof AM. Here, cognitive as well as structural and functional magnetic resonance imaging data was collected from anextremely rare individual with near-perfect AM, or hyperthymesia. Right amygdala hypertrophy (approximately20%) and enhanced amygdala-to-hippocampus connectivity (>10 SDs) was observed in this volunteer relativeto controls. Based on these findings and previous literature, we speculate that the amygdala likely charges AMswith emotional, social, and self-relevance. In heightened memory, this system may be hyperactive, allowing formany types of autobiographical information, including emotionally benign, to be more efficiently processed asself-relevant for encoding and storage.

Keywords: Amygdala; Hyperthymesia; Superior memory; Hypermnesia; Autobiographical memory.

The mysteries of human memory have intriguedphilosophers, scientists, and society in generalfor centuries. Why is it that some memoriesendure the test of time, while others are seem-ingly lost within days or weeks? Understandingthe neurophysiology of memory can help to elu-cidate the difference between these two scenarios.Autobiographical memory (AM) has received agreat deal of attention over the past decade. Many

This work was supported by National Institute on Aging grant R01 AG038471 (BAA), the Vanderbilt University Institute of ImagingScience, and the Vanderbilt University Department of Neurology. The authors are eternally indebted to HK and his grandmother fortheir patience and willingness to cooperate during the many trips to the laboratory. Also, many thanks to HK’s treating physicians; TomDavis, MD, Jennifer Najjar, MD, and Eric Pina-Garza, MD at Vanderbilt University Hospital for providing insight and verificationinto HK’s history and development. Without the referral from Dr. Davis, this project would never have been completed. We also thankDonna Butler and Victoria Morgan for experimental assistance.

Address correspondence to Brandon A. Ally, PhD, Memory Disorders Research Lab, Department of Neurology, A-0118 MedicalCenter North, 1161 21st Avenue South, Nashville, TN 37232, USA. (E-mail: brandon.ally@vanderbilt.edu).

believe AM to be uniquely human, allowing usto maintain a sense of self, as well as to simu-late and predict future events (Schacter, Addis, &Buckner, 2007). More specifically, AM is a com-plex phenomenon, dependent on the delicate inter-action of episodic memory for ‘what’, ‘where’,and ‘when’, semantic memory for factual knowl-edge, visual imagery, emotion, self-reflection, men-tal time travel, and executive control functions

c© 2012 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

http://www.psypress.com/neurocase http://dx.doi.org/10.1080/13554794.2011.654225

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

2 ALLY ET AL.

(Cabeza & St. Jacques, 2007; Conway & Pleydell-Pearce, 2000), which collectively provoke the sub-jective perception of re-experiencing a past event(Conway, 2009). Clinically, AM abnormalities havebeen implicated early in neurodegenerative pro-cesses such as Alzheimer’s disease and related geri-atric cognitive disorders (Addis, Sacchetti, Ally,Schacter, & Budson, 2009). As such, neuroscien-tists have attempted to elucidate the brain networksthat support AM, with the motivation that a morethorough knowledge of AM would have broad rele-vance for understanding human brain function, andcould translate to the clinic where memory disor-ders comprise a growing public health concern.

The majority of neuroimaging studies investi-gating AM suggest a large core brain network,involving hippocampus, parahippocampal gyrus,medial and ventrolateral prefrontal cortex, pre-cuneus, retrosplenial and posterior cingulate cor-tices, lateral temporal cortex, and temporo-parietaljunction (Cabeza & St. Jacques, 2007; Schacteret al., 2007). Additionally, the amygdala andsensory-perceptual areas, such as occipital cortex,are recruited during AM encoding and retrieval(Cabeza & St. Jacques, 2007; Maguire, 2001;Svoboda, McKinnon, & Levine, 2006), but the sig-nificance of these additional structures to the corenetwork has been debated (Markowitsch, 1992;Markowitsch & Staniloiu, 2011; Svoboda et al.,2006). Previous behavioral and neuroimaging inves-tigations of AM have been performed over a rangeof individuals with normal or reduced AM per-formance. However, limited information is avail-able regarding individuals with elevated AM per-formance. In fact, there has been only one casereport in the literature of an individual with near-perfect AM, otherwise described as autobiograph-ical hypermnesia or hyperthymesia (Parker et al.,2006). Parker et al. (2006) describe a female, AJ,in her 40s whose perfect AM dominates her life.AJ spends excessive amounts of time reliving pastevents with great detail and accuracy. Althoughthere has been great media attention surroundingAJ, and possibly others with perfect AM, therehas been very little investigation into the structuraland functional differences in the brains of individu-als with hyperthymesia relative to normals. Indeed,such a detailed examination of structural and func-tional brain differences may help to elucidate theunderpinnings of healthy AM, and perhaps haverelevance in translational studies by providing pos-sible targets for therapy in patients with memorydisorders.

Here, we performed intellectual, cognitive, andneuroimaging studies with HK, a 20-year-old manwith autobiographical hypermnesia. Aside frombeing only the second case reported in the sci-entific literature, HK’s medical history makes thestudy of his superior AM unique. He was bornprematurely at 27 weeks and suffered retinopa-thy of prematurity (ROP), resulting in completeblindness. Although visual imagery is thought tobe at the core of autobiographical re-experiencing,patients born blind are believed to show rela-tively little differences in the quality or detail oftheir AMs (Eardley & Pring, 2006). To under-stand the underpinnings of HK’s superior AM,we undertook three phases of investigation. First,HK underwent an extensive battery of intellectualand memory testing to understand the uniquenessof his AM performance relative to other cogni-tive functions. Second, to understand the accuracyof HK’s AM, we queried at least four unremark-able dates from each year of his life. Events, whichwere multiple for any given date, were verifiedthrough diaries kept by HK’s grandmother, inter-view with his family, electronic medical records atVanderbilt University Hospital, and the Internet.We also developed an AM interview based on pre-vious work in the field to understand the episodicand semantic contributions to his AM, as well asa structured interview to understand HK’s subjec-tive experience of AM retrieval. Finally, to under-stand the anatomical and hemodynamic substratesunderlying his AM, HK and healthy volunteersunderwent structural and function connectivityneuroimaging analysis. To our knowledge, this isthe first report of such a broad cognitive andneuroimaging examination of an individual withsuperior AM, and the results of this investigationprovide novel insight into the mechanistic originsof memory in humans.

METHODS

Intellectual and memory assessment

To assess IQ/Intellect in HK, subtests that donot require vision from the Wechsler AdultIntelligence Scales – IV were administered over one2-hour session. Approximately 1 week later, theWechsler Memory Scales – IV and CaliforniaVerbal Learning Test – II were administered overone 2-hour session to assess HK’s memory. Resultswere compared with well-known normative scores

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 3

for these tests in age and sex-matched cohorts(Delis, Kramer, Kaplan, & Ober, 2000; Wechsler,2008).

Autobiographical memory assessment

The AM assessment was broken into three parts.First, we collected at least four dates from everyyear of HK’s life since his first memory in 1993.For each date, we gathered at least three factsfrom HK’s family, medical records, and histori-cal facts regarding the Nashville area, where HKresided. The entire interview consisted of 80 datesspanning two separate sessions, 10 of which wererepeated across sessions to help assess consis-tency in HK’s AM. Percent correct was calcu-lated by dividing the total number of correctfacts given by HK by the total number of factsfor each event. Second, to assess the semanticand episodic contributions to his AM, we devel-oped a semi-structured questionnaire based onthe Autobiographical Memory Interview (AMI;Kopelman, Wilson, & Baddeley, 1990; Levine,Svoboda, Hay, Winocur, & Moscovitch, 2002)and the Test Episodique de Mémoire du Passéautobiographique (TEMPau; Piolino, Belliard,Desgranges, Perron, & Eustache, 2003). We selectedthree time periods, consisting of 6-year blocks,beginning from the age of three (3–8, 9–14, 15–20).Cues were general events that had occurred in HK’slife. For example, for the Adolescent time period,one of the cues was, ‘Tell me about an event thatinvolved sports and your grandmother betweenin the ages of 9 and 14.’ HK’s responses wererecorded, transcribed, and scored for semantic andepisodic details using previously established crite-ria (Levine et al., 2002). Finally, we administered astructured interview to understand HK’s subjectiveexperience of his AMs. For examples of questionsand responses for each of the AM interviews, seeAppendix.

Magnetic resonance imaging (MRI)

In addition to the above cognitive testing, we per-formed structural and functional MRI on HK anda group of age and sex-matched controls. All volun-teers provided informed, written consent and werescanned at 3.0T (Philips Medical Systems, Best,The Netherlands) using volume body coil trans-mission and phased-array (8-channel) head coilreception.

Structural imaging

Standard T1-weighted structural imaging(3D turbo gradient echo; 1 × 1 × 1 mm3;TR = 9.1/TE = 4.6 ms) was performed in both HKas well as a cohort (n = 30) of approximately age-matched (29 ± 4 years) healthy male volunteers.Additionally, T2-weighted (2D turbo spin echo;0.5 × 0.5 × 4 mm3; TR = 3000/TE = 80 m) andT2-weighted Fluid Attenuated Inversion Recovery,FLAIR (turbo inversion recovery; 1 × 1 × 5 mm3;TR = 9000/TE = 120 ms), sequences wereperformed on HK for pathology classification.

Functional connectivity imaging

HK, as well as a sub-group of the controlvolunteers (n = 10; age = 29 ± 5 years) werescanned using a baseline blood oxygenation level-dependent, BOLD (gradient echo; 3 × 3 × 4 mm3;TR/TE = 3000/35 ms; 120 time points), approach.Volunteers were instructed to lie in the scan-ner awake, with their eyes closed. These datawere acquired with the intent of assessing func-tional connectivity within and between corticaland subcortical structures. Owing to HK beingborn blind, and the corresponding difficulty inperforming memory encoding tasks in the scan-ner, memory encoding tasks were not specificallyperformed.

Analysis

Volumetric analysis

Total gray matter (GM), white matter (WM), andcerebrospinal fluid (CSF) volume were quantifiedin mL from T1-weighted structural scans using ahidden Markov random field model and an associ-ated Expectation-Maximization algorithm (Zhang,Brady, & Smith, 2001) and routines provided bythe Oxford Functional MRI of the Brain (FMRIB)software library (FSL). Additionally, the volumeof subcortical structures believed to have rele-vance to memory networks, including hippocam-pus, amygdala, thalamus, caudate, putamen, andpalladium, were quantified separately in left andright hemispheres using model-based segmenta-tion algorithms available within the FMRIB inte-grated registration and segmentation tool, FIRST(Patenaude, Smith, Kennedy, & Jenkinson, 2011).For inter-subject comparison, subcortical volumeswere normalized by total intracranial tissue volume

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

4 ALLY ET AL.

to reduce bias from head size discrepancies. Finally,vertex analyses (Patenaude et al., 2011) wereperformed to identify the spatial locations ofvolumetric differences between HK’s subcorticalstructures and controls.

Functional connectivity imaging

BOLD data were corrected for motion, base-line drift, and co-registered to a standard spacebrain atlas (Montreal Neurological Institute atlas;2 mm). Seven known network hubs (Tomasi &Volkow, 2011) comprising PCVP, inferior parietal,cuneus, postcentral cortex, cerebellum, thalamus,and amygdala were detected on an individual sub-ject basis using a seed voxel analysis (Table 2; seedvoxel coordinates). Voxel-wise t-scores, describingthe extent to which a given voxel time course signif-icantly correlates with the seed voxel time course,were calculated. A composite map was createdby summing the t-scores from all subject-specificmaps and thresholding at the 99% confidence inter-val (robust range). This mask was then appliedto all volunteer data and t-values were recordedwithin the mask for each volunteer, thereby giv-ing a measure of intra-network connectivity. Next,connectivity, measured as t-score, between eachvoxel and the functional hub seed voxel were deter-mined. Voxels in which the t-score of HK’s mapwas more than four standard deviations from themean control t-score map value were defined assignificant. Therefore, connectivity within majorfunctional hubs as well as between the functionalhubs and all voxels were determined.

Case study considerations

The exceptional nature of HK’s memory clearlywarrants a thorough investigation of any possi-ble unique brain anatomy and/or neurophysiol-ogy. However, the unique nature of HK is suchthat AM characterization must be accomplishedwithin the limitations of a case study, as it isnot possible to perform group-level comparisonsowing to the small number of individuals withhyperthymesia. Therefore, extreme care must betaken to ensure identical scanner hardware andgradient configuration, pulse sequence parameters,and co-registration performance when evaluatingchanges between a single volunteer and a controlgroup. The former two requirements were ensuredby scanning all volunteers on the same scanner andsoftware version (Rel. 2.6.3.4). For co-registration,

12-parameter affine transformations were used andcare was taken, through both detailed visual inspec-tion and ensuring similar cost and similarity func-tion values for HK and controls, to ensure that co-registration of HK’s structural and low-resolutiongradient echo data were of comparable quality asthe control data.

RESULTS

Intellectual and memory assessment

Table 1 demonstrates that HK performed withinthe average range of intelligence. On the WechslerAdult Intelligence Scales – Fourth Edition, hisVerbal Comprehension Index Score was 97 and hisWorking Memory Index Score was 95 (mean = 100,SD = 15). He also performed in the averagerange on measures of episodic memory. On theWechsler Memory Scales – Fourth Edition, HKperformed within ±0.5 SDs from the group meanon the Auditory Memory, Immediate Memory,and Delayed Memory Indices. He was also within±0.5 SDs on all measures of the California VerbalLearning Test – II. For intellectual and memoryperformance data, please see Table 1.

Autobiographical memory assessment

The AM assessment was broken into three parts toexamine the development, accuracy, and quality ofHK’s AMs. HK reported his first autobiographicalrecollection to be from December 17th, 1993,when at the age of 3.5 years, his father put him ina red satchel and carried him around the houselike Santa Claus. As can be seen in Figure 1,for dates between this first memory until his10th year of life, HK shows a relatively steadyincrease in accuracy for autobiographical events.Accuracy takes a noticeable jump to near 90% in2001 at age 11. From that point forward, HK’srecollection of autobiographical events is nearperfect. Additionally, 10 dates were re-queried ina second session. HK’s answers were extremelysimilar for these dates across sessions, with nomissed events or content, placing his consistencyacross a month-long delay at 100%. Related tothe development and quality of HK’s AMs, wedeveloped a semi-structured questionnaire basedon the AMI (Kopelman et al., 1990; Levine et al.,2002) and the TEMPau (Piolino et al., 2003). Threetime periods of HK’s life were explored: 3–8 yearsold (childhood), 9–14 years old (adolescence), and

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 5

TABLE 1Results of HK’s intellectual and memory testing

IQ/Scaled score Percentile SS

HK’s performance on the WAIS-IV intellectual testing. (Mean Index score = 100, SD = 15; meanScaled score = 10, SD = 3)

Verbal Comprehension Index 95 37 −0.33Similarities 9 37 −0.33Vocabulary 8 25 −0.67Comprehension 10 50 0.00

Working Memory Index 97 42 −0.28Digit Span 10 50 0.00Arithmetic 9 37 −0.33Letter-Number Sequencing 9 37 −0.33

IQ/Scaled score Percentile SS

HK’s performance on the WMS-IV and the CVLT-II memory testing. (Mean Index score = 100,SD = 15; mean Scaled score = 10, SD = 3)

Auditory Memory Index 102 55 +0.25Logical Memory I 11 63 +0.33Logical Memory II 11 63 +0.33Verbal Paired Associates I 9 37 −0.33Verbal Paired Associates II 11 63 +0.33

Immediate Memory Index 100 50 0.00Delayed Memory Index 107 68 +0.45

California Verbal Learning Test-II Raw score Percentile SDTrial 1 Free Recall 6/16 31 −0.5Trial 2 Free Recall 9/16 31 −0.5Trial 3 Free Recall 11/16 50 0.0Trial 4 Free Recall 12/16 50 0.0Trial 5 Free Recall 14/16 69 +0.5Trail 1–5 Total Recall 52 58 +0.2List B Free Recall 5/16 16 −1.0Short Delay Free Recall 13/16 69 +0.5Short Delay Cued Recall 14/16 84 +1.0Long Delay Free Recall 13/16 50 0.0Long Delay Cued Recall 14/16 69 +0.5Long Delay Recognition 16/16 69 +0.5

Figure 1. Accuracy of HK’s AM performance. Bars represent percent correct for autobiographical events. Numbers in parenthesesreport HK’s age for the year above, starting at 3 years of age in 1993.

15–20 years old (recent). Results, using previouslyestablished scoring criteria to examine the contribu-tion of semantic and episodic information (Levineet al., 2002) to HK’s AM, reveal a significant

increase in episodic details for the adolescent andrecent AMs compared to the childhood AMs (seeFigure 2). It should be noted here that in additionto semantic-based memories from his childhood

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

6 ALLY ET AL.

Figure 2. Quality of HK’s AM performance. Bars represent the proportion of episodic details to the overall number of details for thethree periods of HK’s life.

years, there is likely contamination from externalsources (e.g., family member rehearsing memories).For example, although HK could perceive light andsome color information in his right eye very early inhis life, it is likely that the description of the satchelas ‘red’ in his first memory was gathered froman external source while rehearsing with familymembers. This type of contamination has beenreferred to as suggestibility (Schacter, 1999), andis common in children where the actual memory isaccurate, but embellishment has been provided dueto suggestibility (see Ceci, 1995 for review).

In addition to the objective measure of the qual-ity of HK’s AMs, we also developed an interview tounderstand HK’s subjective experience of his AMs.He reports that he is able to relive memories inhis mind as if they just happened. HK stated thateverything about his memory, including sounds,smells, and emotions, are vividly re-experiencedwhen he remembers a particular event in time, andhe described his AMs as being in the first personapproximately 90% of the time. He stated that thereis no difference in the vividness of his recollectionbetween events that occurred when he was five andevents that he experienced within the past month.HK also reported that sounds, smells, emotions,and news events act as cues for his AM, trigger-ing the retrieval of past events with similar contexts.He claims that he does not dwell on his memories,but it becomes part of his daily routine to wake upand think about that particular date in his own his-tory. HK reported that he cannot stop AMs fromcoming into consciousness, and bad memories arerecalled just as often as good memories. However,HK stated that he can stop thinking about his mem-ory at any given time, and he tends to focus on onlythe positive memories. For examples of questions

and responses for each of the AM interviews, seeAppendix.

Magnetic resonance imaging (MRI)

The results of the neuroimaging studies reveal apattern of structural and functional connectivityuniqueness that likely contributes to HK’s height-ened AM. Figure 3 shows a subsample of slicesfrom the structural scanning of HK; the col-ored arrows show regions of pathology, presentingprimarily as white matter pathology and dam-age to the optic radiations as a result of hisprematurity. A more detailed volumetric analysis(Figure 4) reveals significantly reduced total tis-sue volume in HK (1019 mL) relative to controls(1249 ± 29 mL). Additionally, a volumetric analysisof subcortical structures shows general reductionin subcortical volumes, with the noted exceptionof the right amygdala, which is fractionally larger(18 ± 6%) than the control volume. This find-ing is for an approximately age-matched (n = 30;age = 29 ± 4 years) cohort of male volunteers,however we found this trend to be preserved whenstructural data from a larger (n = 74) mixed-sexcohort was analyzed as well (right amygdala frac-tional increase = 15 ± 5%). Volume renderings ofthe amygdala are shown in Figure 4c.

Table 2 shows the seven hubs used for func-tional connectivity analysis, which include bothcortical and subcortical networks, and were selectedfor analysis based on the robust presence of thesenetworks in functional connectivity density map-ping procedures (Tomasi & Volkow, 2011). Theseed voxel coordinates used for map generation(Figure 5) are displayed, as well as the functional

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 7

Figure 3. Structural neuroimaging. Representative slices from FLAIR, T2- and T1-weighted structural MRI acquisitions in HK.Blindness is secondary to damage to the optic radiations (blue arrow). Cystic encephalomalacia is present in the left dorsal lateralthalamus (red arrow) and slight periventricular leukomalacia (yellow arrow) is observed. These findings are consistent with an individualborn >10 weeks premature.

connectivity, reported as t-score, for the controlpopulation and HK. Notice that the postcentraland thalamic networks have increased connectiv-ity in HK, which may be attributed to height-ened somatosensory awareness concurrent to blind-ness (Shu, Liu, Yonghui, Chunshui, & Jiang,2009). Alternatively, the posterior cingulate/ventralprecuneus (PCVP) network, which is comprisedof regions that constitute the default mode net-work (DMN), was reduced in HK relative tocontrols.

To further understand the amygdala volumetricfinding and any potential unique functional roleof the amygdala in HK, we investigated func-tional connectivity between the amygdala and otherbrain structures. Figure 6 shows the regions wherethe connectivity between the amygdala networkhub is significantly increased in HK relative tothe control population. Notice that the connectiv-ity is most different in the right hippocampus, aswell to a lesser extent in more distal white matterregions.

DISCUSSION

A detailed examination of AM performance in anindividual with hyperthymesia, as well as structural,functional, and metabolic neuroimaging has helpedto possibly elucidate underpinnings of heightenedAM. With respect to the development of AM, HKdemonstrated an increase in both accuracy andepisodic detail during the transition from childhoodto adolescence. Specifically, from the period of 9 to12 years of age, we observed a sharp increase inthe accuracy of HK’s AMs, as well as an increasein episodic details compared to semantic details inhis AMs for the 9–14-year adolescent time period.These findings are generally consistent with workexamining the development of AM. Specifically,while evidence suggests that children as young asthree exhibit a capacity for episodic memory, theytend to only recollect semantic fact-stating nar-ratives from their pasts (Fivush, 2011; Nelson &Fivush, 2004). It has been speculated that themental time-travel aspect of AM does not develop

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

8 ALLY ET AL.

Figure 4. Volumetric analysis. Volumetric analysis reveals smaller cortical (a) and subcortical (b) volumes in HK. Subcortical fractionalvolumes are defined as structure volume normalized to total brain tissue volume. Notice the increased right amygdala fractional volumerelative to all other subcortical structures analyzed. (c) Volume rendering (below) of statistical variation in amygdala volume inlaid onstandard space brain atlas. The color bar describes the F-statistic for which HK’s amygdala volume is larger than that of the controlpopulation. On right, the right amygdala rending is shown (three views), with cooler colors describing regions of increased growthrelative to the control population.

TABLE 2Voxel coordinates (MNI) used for seed voxel analysis indefining each of the separate network hubs, along with

within network connectivity (t-scores) for controls and HK

x y z t-scoreControl t-scoreHK

PCVP 4 −55 29 12.9 ± 1.7 8.8∗Inferior Parietal −38 −57 39 11.0 ± 5.2 8.9Cuneus −24 −83 15 12.2 ± 7.9 9.2Postcentral 20 −48 59 9.9 ± 6.6 50.1∗Cerebellum −9 −56 −27 7.4 ± 2.4 10.0Thalamus −12 −20 8 8.0 ± 2.1 26.9∗Amygdala 24 −5 −18 14.8 ± 13.5 14.7

∗Regions where HK’s t-scores are significantly (>2 SD) differ-ent from those of controls.

until around the age of 11 (Piolino et al., 2007),and that autobiographical recollections from chil-dren this age struggle to meet criteria for true AM(Fivush, 2011). Supporting this hypothesis, oth-ers have found that not until the age of 12 canchildren link events in their life to an accurateautobiographical timeline (Habermas & de Silveira,2008). In fact, 8-year-old children are slightly abovechance at accurately judging the order of the eventsthat occurred more than a few months in the past(Friedman, 2003). These behavioral findings arealso consistent with neuroimaging work revealingthat brain regions associated with the AM net-work lack functional connectivity until 7–9 years

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 9

Figure 5. Within-network functional connectivity. The sevenfunctional network hubs analyzed, corresponding to (a) poste-rior cingulate/ventral precuneus (PCVP), (b) inferior parietal,(c) cuneus, (d) postcentral, (e) cerebellum, (f) thalamus, and(g) amygdala. The color bar represents the cumulative t-scoredistribution, thresholded at the 99 percentile robust range overall volunteers. All colored voxels shown were used for definingthe network masks, and each mask was applied separately toeach volunteer to assessing intra-network functional connectiv-ity within the specified network hub.

Figure 6. Functional connectivity from HK’s right amygdala.Coronal and sagittal slices are shown above, as well as repre-sentative axial slices below. The color bar describes the numberof standard deviations HK’s t-scores vary from those of the con-trol population. Notice the increased connectivity to the righthippocampus.

of age (Fair et al., 2008). Further, DMN, whichheavily overlaps with the AM network, has beenreported not to fully develop until 9–12 years of age(Thomason et al., 2008).

The development of HK’s AM is also very sim-ilar to that of AJ, the first individual with autobi-ographical hypermnesia presented in the literature(Parker et al., 2006). AJ reported her first memoryas being from when she was 18–24 months in age,and she vividly remembers autobiographical detailsfrom late in her third year of life. AJ recalls thather memory ‘changed’ at the age of eight, whenshe was traumatized by the move of her family.At this point, she began organizing and recordingher memories, and stated that she constantly forcedherself to relive her experiences. AJ first becameaware of her superior memory around the age of12. She can recall most, but not all days betweenthe ages of 8 and 13. For events at the age of 14 andbeyond, her AM is almost automatic (Parker et al.,2006). Similarly, HK can vividly remember eventsas early as 3 years of age, and reports first becom-ing aware of his superior memory at age 11, withnear automaticity by the age of 14.

Cognitive neuroscientists report structural andconnectivity changes associated with brain develop-ment may be responsible for changes in AM per-formance (Bauer, Burch, Scholin, & Guler, 2007).Longitudinal studies suggest a steady increase incortical gray matter volume in memory-related

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

10 ALLY ET AL.

areas until approximately the age of 16 (Gieddet al., 1999). Limbic subcortical areas also showdevelopmental growth (albeit much slower), withrecent work suggesting a sharp increase in amyg-dalar volume between the ages of 11 and 13 asso-ciated with increased testosterone during puberty(Neufang et al., 2009). This time period is highlyconsistent with when a noticeable jump in HK’sAM accuracy was observed. Researchers also high-light a dramatic increase in white matter volumeand connectivity, particularly to and from memory-related limbic areas, during the first 10 years of life(Pfefferbaum et al., 1994). Two findings from thecurrent neuroimaging study stand out with respectto the structure and functional connectivity withinthe AM network. First, HK’s right amygdala issignificantly larger in volume compared to con-trol subjects. Second, the functional connectivitybetween the amygdala and hippocampus, as well asdistal cortical and subcortical regions in HK, is sig-nificantly increased compared to controls (greaterthan 10 SDs above the control group mean).

Although there has been debate as to whetherthe amygdala falls in the AM core network(Markowitsch, 1992; Markowitsch & Staniloiu,2011; Svoboda et al., 2006), recent work has cer-tainly highlighted the importance of the amygdalato encoding AMs (Greenberg, Rice, Cabeza, Rubin,& Labar, 2005; Spreng & Mar, 2012). Most recently,it has been shown that amygdala activity at encod-ing predicts the subjective vividness of an episodicmemory, regardless of its emotional valence orarousal (Kensington, Addis, & Atapattu, 2011;Phelps & Sharcot, 2008), leading to the hypothesisthat the amygdala is critical in relaying biologi-cal and social significance to AM (Markowitsch& Staniloiu, 2011). Indeed, it has been proposedthat the subjective sense of remembering invariablyinvolves an emotional re-experiencing of an event(Rubin & Berntsen, 2003; Welzer & Markowitsch,2005), and this emotional aspect serves as the foun-dation for episodic AMs (Piolino, Desgranges, &Eustache, 2009). Further, emotionally benign infor-mation may be processed in an affective mannerdue to its self-relevance (Markowitsch & Staniloiu,2011; Rameson, Satpute, & Lieberman, 2010).Consequently, it has been posited that the amygdalaacts as the hub for processing sensory informa-tion of biological, social, and self-importance forencoding and subsequent storage in neocorticalareas (Markowitsch & Staniloiu, 2011). It is likelythat the enhanced functional connectivity betweenamygdala and hippocampus in HK allows for

information to be easily processed as self-relevantand bundled by amygdala–hippocampal interactionfor encoding and storage.

The interactions of amygdala and cortical areashas been highlighted by studies of functionalconnectivity, which show that the amygdala isconnected to nearly 90% of all cortical areas(Cole, Pathak, & Schneider, 2010; Young, Scannell,Burns, & Blakemore, 1994), making it an excellentcandidate for increasing the likelihood that mem-ories are properly stored and retrieved (Ritchey,Dolcos, & Cabeza, 2008). Indeed, right amygdala–hippocampal connectivity to medial prefrontalregions has been shown to support memoryencoding of high self-relevance or self-involvement(Muscatell, Addis, & Kensinger, 2010). Moreover,in a study comparing the retrieval of true AMsand the retrieval of fictitious episodes, the rightamygdala was activated when retrieving the trueAMs, whereas the retrosplenial/precuneus areawas activated during the retrieval of the fictitiousepisodes (Markowitsch et al., 2000). This right lat-eralized effect has also been seen in the retrieval offield memories, or memories experienced in the firstperson (Eich, Nelson, Leghari, & Handy, 2009).Similarly, HK, whose right amygdala is enlargedcompared to controls, reports that approximately90% of his AMs are experienced in the first person.This is in comparison to the general population,who report approximately 66% of AMs being inthe first person (Sutin & Robins, 2008). It has beenargued that the increase in right amygdala activ-ity during first person retrieval perspective reflectsthe higher degree of subjective emotionality (Eichet al., 2009).

Of course, changes in structure and function ofamygdala have also been linked to other pathologicchanges in memory. For example, amygdalar vol-umes are increased in individuals who are exposedto chronic stress due to post-traumatic stress dis-order (PTSD) (Roozendaal, McEwen, & Chattarji,2009), and right amygdala over-activation has beenimplicated in the retrieval of traumatic memoriesin these patients (Driessen et al., 2004). Memoriesassociated with PTSD are persistent and com-monly rehearsed by these individuals, despite theirdesire to end the rumination. It has been proposedthat hypermnesia, particularly for emotional events,may be amygdala-dependent and varies as a func-tion of noradrenergic-glucocorticoid input into theamygdala (Hurlemann et al., 2005). In contrast,individuals with right medial temporal lobectomiesproduced fewer emotional AMs compared to

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 11

individuals with left medial temporal lobectomies(Buchanan, Tranel, & Adolphs, 2005). These stud-ies provide further evidence that the amygdala iscritically involved in encoding and retrieval of AMs,but our understanding of the mechanism of suchretrieval is relatively limited.

While HK’s structural and connectivity resultsmay enhance our understanding of superior AM,we would be remiss to not acknowledge his uniquesituation and inherent limitations of the currentstudy. HK’s occipital regions are highly active andwell connected during rest, and studies of blindindividuals have shown recruitment of occipitalareas on tasks of episodic memory (Raz, Amedi,& Zohary, 2005). Though there have been fewstudies of AM in the blind, recent literature suggeststhat blind participants demonstrated no differencesin memory specificity compared to sighted indi-viduals, even if the retrieval cue is dependent onvisual imagery (Eardley & Pring, 2006). Futurework could potentially recruit born blind individ-uals with normal AM to compare HK’s structuraland functional differences to this control group.We also acknowledge that unique case studies suchas HK are not easily translated or generalizable tothe normal population. The current results shouldbe interpreted with caution, but continue to pro-vide evidence that the amygdala is heavily involvedin AM. Further, perhaps the present findings canhelp to guide future regions of brain stimulation inmemory-disordered populations, with the goal ofimproving memory function. Indeed, brain stimula-tion to deep, subcortical memory-related structureshas shown very early promise in patients with AD(Laxton et al., 2010).

In conclusion, we provided a detailed exami-nation of AM performance as well as structuraland functional neuroimaging in an individual withhyperthymesia with the motivation of elucidatingthe mechanistic origins of AM. The behavioraldata show an increase in accuracy and episodiccontribution associated with AM, paralleling thecourse of AM development proposed in normalindividuals. However, there was a sharp increasein accuracy and the number of episodic detailsassociated with the transition from childhood toadolescence, which could indicate the time intervalassociated with pathologic developmental changesrelated to brain structure and physiology. Theneuroimaging data reveal HK’s right amygdala tobe nearly 20% fractionally larger than normals, inthe face of significantly reduced gray and whitematter volumes. Additionally, HK has significantly

increased connectivity between his right amygdalaand hippocampus, as well as distal cortical andsubcortical regions. We posit that the amygdala,particularly the right amygdala, plays a vital rolein AM encoding and retrieval, likely by charg-ing AMs with emotion or social and self-relevance(Markowitsch & Staniloiu, 2011). We further specu-late that in HK, this system is hyperactive, resultingin emotionally benign information being processedin a self-relevant affective manner (Rameson et al.,2010). While the results of this unique case study donot provide direct evidence for the underpinningsof normal memory function, the present inves-tigation provides significant support for previoushypotheses as to the role of the amygdala in AMperformance as well as provides the basis of poten-tial future targeted therapies for patients with mem-ory disorders.

Original manuscript received 6 September 2011Revised manuscript accepted 4 December 2011

First published online 24 April 2012

REFERENCES

Addis, D. R., Sacchetti, D., Ally, B. A., Schacter, D. L.,& Budson, A. E. (2009). Episodic simulation offuture events is impaired in mild Alzheimer’s disease.Neuropsychologia, 47, 2660–2671.

Bauer, P. J., Burch, M. M., Scholin, S. E., & Guler,O. E. (2007). Using cue words to inform the distri-bution of autobiographical memories in childhood.Psychological Science, 18, 910–916.

Buchanan, T. W., Tranel, D., & Adolphs, R. (2005).Emotional autobiographical memories in amnesicpatients with medial temporal lobe damage. Journalof Neuroscience, 25, 3151–3160.

Cabeza, R., & St. Jacques, P. (2007). Functionalneuroimaging of autobiographical memory. Trends inCognitive Sciences, 11, 219–227.

Ceci, S. J. (1995). False beliefs: Some developmentand clinical considerations. In D. L. Schacter (Ed.),Memory distortion: How minds, brains, and societiesreconstruct the past (pp. 91–128). Cambridge, MA:Harvard University Press.

Cole, M. W., Pathak, S., & Schneider, W. (2010).Identifying the brain’s most globally connectedregions. NeuroImage, 49, 3132–3148.

Conway, M. A. (2009). Episodic memories. Neuro-psychologia, 47, 2305–2313.

Conway, M. A., & Pleydell-Pearce, C. W. (2000). Theconstruction of autobiographical memories in the self-memory system. Psychological Review, 107, 261–288.

Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A.(2000). California Verbal Learning Test – SecondEdition. San Antonio, TX: Pearson Publishing.

Driessen, M., et al. (2004). Different fMRI activation pat-terns of traumatic memory in borderline personality

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

12 ALLY ET AL.

disorder with and without additional posttraumaticstress disorder. Biological Psychiatry, 55, 603–611.

Eardley, A. F., & Pring, L. (2006). Remembering the pastand imagining the future: A role for nonvisual imageryin the everyday cognition of blind and sighted people.Memory, 14, 925–936.

Eich, E., Nelson, A. L., Leghari, M. D., & Handy, T. C.(2009). Neural systems mediating field and observermemories. Neuropsychologia, 47, 2239–2251.

Fair, D. A., Cohen, A. L., Dosenbach, N. U. F., Church,J. A., Miezin, F. M., Barch, D. M., Raichle, M. E.,Petersen, S. E., & Schlaggar, B. L. (2008). Thematuring architecture of the brain’s default network.Proceedings of the National Academy of Science of theUnited States of America, 105, 4028–4032.

Fivush, R. (2011). The development of autobiographicalmemory. Annual Review Psychology, 62, 559–582.

Friedman, W. J. (2003). The development of a differenti-ated sense of the past and the future. Advances in ChildDevelopment and Behaviour, 31, 229–269.

Giedd, J. N., Blumenthal, J., Jeffries, N. O., Castellanos,D. X., Liu, H., Zijdenbos, A., Paus, T., Evans, A. C.,& Rapoport, J. L. (1999). Brain development dur-ing childhood and adolescence: A longitudinal MRIstudy. Nature Neuroscience, 2, 861–863.

Greenberg, D. L., Rice, J. J., Cabeza, R., Rubin, D. C.,& Labar, K. S. (2005). Co-activation of the amygdala,hippocampus, and inferior frontal gyrus during auto-biographical memory retrieval. Neuropsychologia, 43,659–674.

Habermas, T., & de Silveira, C. (2008). The develop-ment of global coherence in life narratives acrossadolescence: Temporal, causal, and thematic aspects.Developmental Psychology, 44, 707–721.

Hurlemann, R., et al. (2005). Noradrenergic modula-tion of emotion-induced forgetting and remembering.Journal of Neuroscience, 25, 6343–6349.

Kensinger, E. A., Addis, D. R., & Atapattu, R. K.(2011). Amygdala activity at encoding correspondswith memory vividness and with memory for selectepisodic details. Neuropsychologia, 49, 663–673.

Kopelman, M. D., Wilson, A. E., & Baddeley, A. (1990).The Autobiographical Memory Interview. Bury St.Edmunds, UK: Thames Valley Test Co.

Laxton, A. W., Tang-Wai, D. F., McAndrews, M. P.,Zumsteg, D., Wennberg, R., Keren, R., et al. (2010).A phase I trial of deep brain stimulation of memorycircuits in Alzheimer’s disease. Annals of Neurology,68, 521–534.

Levine, B., Svoboda, E., Hay, J. F., Winocur, G., &Moscovitch, M. (2002). Aging and autobiograph-ical memory: Dissociating episodic from semanticretrieval. Psychology and Aging, 17, 677–689.

Maguire, E. A. (2001). Neuroimaging studies of autobi-ographical event memory. Philosophical Transactionsof the Royal Society of London, B: Biological Science,356, 1441–1451.

Markowitsch, H. J. (1992). Intellectual functions andthe brain. An historical perspective, Toronto, ON:Hogrefe & Huber Publishing.

Markowitsch, H. J., & Staniloiu, A. (2011). Amygdalain action: Relaying biological and social significanceto autobiographic memory. Neuropsychologia, 49,718–733.

Markowitsch, H. J., Thiel, A., Reinkemeier, M.,Kessler, J., Koyuncu, A., & Heiss, W. D. (2000).Right amygdalar and temporofrontal activationduring autobiographic, but not during fictitiousmemory retrieval. Behavioural Neurology, 12,181–190.

Muscatell, K. A., Addis, D. R., & Kensinger, E. A.(2010). Self-involvement modulates the effective con-nectivity of the autobiographical memory network.Social, Cognitive and Affective Neuroscience, 5, 68–76.

Nelson, K., & Fivush, R. (2004). The emergence ofautobiographical memory: A social cultural develop-mental theory. Psychological Review, 111, 486–511.

Neufang, S., Specht, K., Hausmann, M., Gunturkun,O., Herpertz-Dahlmann, B., Fink, G. R., & Konrad,K. (2009). Sex differences and the impact of steroidhormones on the developing human brain. CerebralCortex, 19, 464–473.

Parker, E.S., Cahill, L., & McGaugh, J. L. (2006). A caseof unusual autobiographical remembering. Neurocase,12, 35–49.

Patenaude, B., Smith, S. M., Kennedy, D. N., &Jenkinson, M. A. (2011). Bayesian model of shapeand appearance for subcortical brain segmentation.Neuroimage, 56, 907–922.

Pfefferbaum, A., Mathalon, D. H., Sullivan, E. V.,Rawles, J. M., Zipursky, R. B., & Lim, K. O. (1994).A quantitative magnetic resonance imaging study ofchanges in brain morphology from infancy to lateadulthood. Archives of Neurology, 51, 874–887.

Phelps, E. A., & Sharot, T. (2008). How (and why)emotion enhances the subjective sense of recollec-tion. Current Directions in Psychological Science, 17,147–152.

Piolino, P., Belliard, S., Desgranges, B., Perron, M., &Eustache, F. (2003). Autobiographical memory andautonoetic consciousness in a case of semantic demen-tia. Cognitive Neuropsychology, 20, 619–639.

Piolino, P., Desgranges, B., & Eustache, F. (2009).Episodic autobiographical memories over thecourse of time: Cognitive, neuropsychologicaland neuroimaging findings. Neuropsychologia, 47,2314–2329.

Piolino, P., Hisland, M., Ruffeveille, I., Matuszewski,V., Jambaque, I., & Eustache, F. (2007). Do school-age children remember or know the personal past?Consciousness and Cognition, 16, 84–101.

Rameson, L. T., Satpute, A. B., & Lieberman, M. D.(2010). The neural correlates of implicit and explicitself-relevant processing. NeuroImage, 50, 701–708.

Raz, N., Amedi, A., & Zohary, E. (2005). V1 activation incongenitally blind humans is associated with episodicretrieval. Cerebral Cortex, 15, 1459–1468.

Ritchey, M., Dolcos, F., & Cabeza, R. (2008). Roleof amygdala connectivity in the persistence of emo-tional memories over time: An event-related FMRIinvestigation. Cerebral Cortex, 18, 2494–2504.

Roozendaal, B., McEwen, B. S., & Chattarji, S. (2009).Stress, memory and the amygdala. Nature ReviewsNeuroscience, 10, 423–433.

Rubin, D. C., & Berntsen, D. (2003). Life scripts help tomaintain autobiographical memories of highly pos-itive, but not highly negative, events. Memory andCognition, 31, 1–14.

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 13

Schacter, D. L. (1999). The seven sins of memory. Insightsfrom psychology and cognitive neuroscience. TheAmerican Psychologist, 54, 182–203.

Schacter, D. L., Addis, D. R., & Buckner, R. L. (2007).Remembering the past to imagine the future: theprospective brain. Nature Reviews Neuroscience, 8,657–661.

Shu, N., Liu, Y., Yonghui, L., Chunshui, Y., & Jiang, T.(2009). Altered anatomical network in early blindnessrevealed by diffusion tensor tractography. PLoS One,4, e7228.

Spreng, R. N., & Mar, R. A. (2012). I remember you: Arole for memory in social cognition and the functionalneuroanatomy of their interaction. Brain Research,1428, 43–50. doi: 10.1016/j.brainres.2010.12.024.

Sutin, A. R., & Robins, R. W. (2008). When the ‘I’looks at the ‘me’. Autobiographical memory, visualperspective, and the self. Consciousness and Cognition,17, 1386–1397.

Svoboda, E., McKinnon, M. C., & Levine, B. (2006).The functional neuroanatomy of autobiographicalmemory: A meta-analysis. Neuropsychologia, 44,2189–2208.

Thomason, M. E., Chang, C. E., Glover, G. H., Gabrieli,J. D. E., Greicius, M. D., & Gotlib, I. H. (2008).Default-mode function and task-induced deactiva-tion have overlapping brain substrates in children.Neuroimage, 41, 1493–1503.

Tomasi, D., & Volkow, N. D. (2011). Association betweenfunctional connectivity hubs and brain networks.Cerebral Cortex, doi: 10.1002/hbm.21252 (21 March2011).

Wechsler, D. (2008). Wechsler Adult Intelligence Scales –Fourth Edition. San Antonio, TX: Pearson Publishing.

Welzer, H., & Markowitsch, H. J. (2005). Towards abio-psycho-social model of autobiographical memory.Memory, 13, 63–78.

Young, M. P., Scannell, J. W., Burns, G. A., & Blakemore,C. (1994). Analysis of connectivity: Neural systemsin the cerebral cortex. Reviews in the Neuroscience, 5,227–250.

Zhang, Y., Brady, M., & Smith, S. (2001). Segmentationof brain MR images through a hidden Markov ran-dom field model and the expectation-maximizationalgorithm. IEEE Transactions on Medical Imaging,20, 45–57.

APPENDIXFROM THE DATES IN HISTORY

AUTOBIOGRAPHICAL QUESTIONNAIRE

Examiner: Can you tell me what happened duringyour day on January 2nd, 2001?

HK: Laughs. I got up around 10:15 thatmorning, my grandmother cut my hair,we got dressed, and my grandmotherwent to work. She had the 4–8 shiftthat night. But Mr. James Biddeford,my friend and mentor, picked me upand we went to the Jeff Fisher show.Jeff Fisher had a two-hour show a

couple of days, 5 days I think, beforethe Titans played the Baltimore Ravensin the playoffs. The Titans ended upgetting beat 24-10. I got to sit inJeff Fisher’s lap and I had my pic-ture taken with him. I also got anautograph from Frank Wycheck andI met Dwight Lewis. Dwight Lewis iswas a columnist for the Tennessean.We ended up eating dinner there, atApplebee’s. James’ wife, Bridgette, hadto leave early because she was expect-ing a call from Julia, her daughter.Around 8:00 everyone left, and thenwe went back to the house to meetmy grandmother. She and I watchedthe nightly news, I listened to SeanHannity on talk radio, and I ended upgoing to bed around 10:00 pm. It wasan exciting day.

Examiner: Can you tell me what happened duringyour day on March 19th, 2003?

HK: Oh, that’s when the war in Iraq started.I had spent the night at the TennesseeSchool for the Blind the night before,in the cottage. That was a Wednesdaynight. After my grandmother pickedme up from school, we had dinnerat home. Spinach Alfredo and noo-dles. After dinner, I was watchingStar Search on Channel 5. At around8:45 pm they broke in and said that UStroops were shooting in target housesthinking Saddam Hussein mighthave been in there. Three civiliansended up being killed that day, Iraqicivilians.

Examiner: Can you tell me what happened duringyour day on September 22nd, 2003?

HK: Well, nothing exciting happened thatday. I went to school, then went to ther-apy at Vanderbilt. Ellen Diamano hadme walking on the treadmill and work-ing on my flexibility on this big ball allnight. We ate dinner in the VanderbiltCourtyard Café. I had spinach casse-role. That night I watched EverybodyLoves Raymond. I was excited becauseit was the first episode of the new sea-son. I think it was the eighth season.I think there might have also been areport on the news of a suicide bomberin Baghdad. There was a car explosionnear the UN.

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

14 ALLY ET AL.

Examiner: Do you think you could tell me whathappened on the episode of EverybodyLoves Raymond?

HK: Yeah, Ray and Debra went golfing.So they could spend more time as acouple. They ended up getting in thisbig fight on the 13th green, and Rayaccused Debra of coming just so shecould ruin the only thing he loves todo. They end up making up and Debraadmits that she hates golf and Ray canjust do it alone or with his brotherRobert.

Examiner: Can you tell me what happened duringyour day on February 5th, 2008?

HK: There were tornadoes that day inMacon and Sumner Counties, andwe were under a watch in Nashville.During the storms, there was a natu-ral gas explosion at a plant in SumnerCounty that night. I remember beingat school. I was scared. We were inthe closet under shelter for two peri-ods of time. The first one was dur-ing school between 9:15 to 10:00 forthe first tornado warning. Then wewere up again in shelter at the cot-tage from 12:00 am to 3:00 am forthe second tornado warning. The sirenwas screaming and I remember beingreally hot. It was also Super Tuesdayin Tennessee. Mike Huckabee ended upwinning the Republican Primary andHillary Clinton won the DemocraticPrimary. I really wanted to vote thatday, but it was a couple months beforemy 18th birthday, so I couldn’t. Whenmy grandmother asked me what Iwanted for my 18th birthday, I told herall I wanted was to vote!

FROM THE AUTOBIOGRAPHICAL MEMORYINTERVIEW BASED ON THE TEMPAU

Example response from a memory fromAdolescence (9–14 years old):

Examiner: Tell me about a day involving sportsfrom when you were between the agesof 9 and 14.

HK: Well, the first year I made it to theJunior Olympics was in 1999. It was

Wednesday October 13th, and it waskind of warm in Kentucky for that timeof the year, like 70 degrees. I ran the60 meter dash and the standing longjump, where I basically jumped intothis big sandpit, and I also threw thesoftball throw. I was so happy, I didreally good. I was 2nd place in the60 meters, 3rd place in the standinglong jump, and 3rd place in the softballthrow. A friend’s dad brought me backto my school, where my grandmotherand her friend Janey picked me up.My grandmother was so proud of me,she let us stop at McDonald’s on theway back home. I had chicken nuggetswith fries, and then we went straighthome. We had to rush because Janeywas leaving to go back to Florida thenext day. I was tired and went straightto bed.

STRUCTURED INTERVIEW TO UNDERSTANDHK’S SUBJECTIVE AM EXPERIENCE

(1) How do you remember dates and events thathappened to you?

HK: They just come into my mind. I can just pic-ture it as if I was there again. Especially whenanniversaries come around. That day of theanniversary, I just think back to what I wasdoing, what the weather was like, who I waswith, and so-and-so. I just remember it.

(2) Do you remember only things that interest youor do you seem to remember just about anythingthat happened on a particular date regardless ofwhether you are interested in it?

HK: No, I remember everything that happensduring my day. All of it is easy to remember.I feel like I am a walking computer some-times. The information just gets stored in mybrain. It can get distracting, but I can let itgo too.

(3) Do you ever think about bad things?

HK: I remember bad things too. I just don’t dwellon the bad things. My grandmother hasalways taught me to focus on the positive,and that’s what I do.

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

AMYGDALA-BASED SUPERIOR AUTOBIOGRAPHICAL MEMORY 15

(4) Do you think about memories that happened toyou a lot?

HK: Well, I think about them quite a bit.Especially if it is something that affectsme. A lot of times, when someone men-tions something to me, it triggers a memory.I like telling my grandmother what certainanniversaries are. Like I’ll think about whatwe did 5 years ago. You know, she also takesme to her appointments so that I can help herto remember her stuff too (laughs).

(5) What percentage of the time do you think thatyou are thinking about your memories?

HK: I’d say about 30 percent of the time. I’vegotten better about it. I think when I wasyounger I used to do it more than now.It would get me in trouble in school.

(6) When these events come to you mind, how doyou experience them? Do you experience them asjust a fact or do you actually feel like you areexperiencing the event all over again?

HK: Well, kind of both. I can remember all kindsof facts. But when I think about somethingfrom the past, and event or something, I feellike I am right back in that situation. Therereally is no difference in when it happenedand when I remember it.

(7) When you are experiencing these memories, areyou right back inside your own eyes, or are youmaybe looking down on the scene.

HK: I am usually experiencing it just like it hap-pened, right through my own eyes. (laughs)Of course, I’m blind, so I don’t see it, but itfeels like I am right back there. I can sort ofsee thing in my mind though.

(8) What percentage of time are you lookingthrough your own eyes?

HK: I would say 80 to 90 percent of the time. I dohave the experience like I am another personseeing something else happen. Like I am sit-ting there reliving a memory outside of mybody. But it is usually like I am there again.

(9) If I were to ask you to remember something thathappened when you were 5 or 6, does your mem-ory experience differ from something that happenedlast month?

HK: No. Really there is no difference.

(10) You know, there is another person that has per-fect memory too. She says that sometimes she feelslike there is two television sets in her head. Oneplaying back memories, and one where she is in thepresent. Do you ever feel like this?

HK: Yes, Jill Price. And I think there is some guyin Wisconsin or something. Yes, sometimesI feel like that. Mostly it is one or the otherthough.

(11) Some research has shown that maternal remi-niscing or reliving about the past with their childreninfluences memory. Did you ever relive your experi-ences with your grandmother when you were little?

HK: (laughs) Funny you should ask that. WhenI was little, my grandmother would ask meevery night in bed to tell her everything aboutmy day. So every night, I would tell her every-thing from the time I woke up to the time thatI was going to bed. It was our thing.

(12) You seem to get the day of the week right everytime we ask you about a particular date in history.Do you ever use math to calculate what day of theweek a certain date fell on or do you simply knowit?

HK: No, I just know it.

(13) You also seem to know the weather for a par-ticular date. How is it that you can remember theweather so well?

HK: Well, that is the first thing I do when I wakeup, listen to the weather on talk radio. I havealways liked the weather. We actually used towatch the Weather Channel in science class.I always wanted to learn about weather.

(14) At what age did you start to realize that youcould remember dates and events like you do?

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12

16 ALLY ET AL.

HK: About 10 years old. I remember I was sittingin my chair listening to sports radio and mygrandmother asked me why I was smiling.I told her that 3 years earlier on that date,we went to the Brentwood Rotary Club andhad breakfast with everyone. She looked itup on her calendar and I was right. I guessmy memory started to get really good whenI was 10, 11, or 12.

(15) Do things like smells or sounds ever triggermemories for you?

HK: Oh yeah, they all do. Sounds, smells, evenwhen I feel something or taste something itcan bring back a memory of a similar time.The news really acts as a trigger for me.

(16) Do you collect anything?

HK: Well, I guess I collect events and memories(laughs). I have had a collection of coins inthe past.

Dow

nloa

ded

by [

VU

L V

ande

rbilt

Uni

vers

ity],

[B

rand

on A

lly]

at 0

7:03

23

Apr

il 20

12