RESEARCH ARTICLE

Lari Vainio Æ Rob Ellis Æ Mike Tucker

Ed Symes

Manual asymmetries in visually primed grasping

Received: 8 March 2005 / Accepted: 24 January 2006 / Published online: 18 February 2006� Springer-Verlag 2006

Abstract Previous research has shown that the taskirrelevant size of familiar objects facilitates compatibleprecision and power grip responses. The present studyexamined whether the task irrelevant size of novel ob-jects produces the same compatibility effect. However,the main objective of the study was to investigate whe-ther visually primed precision and power grips aremanually asymmetric. Experiment 1 showed that the sizeof a novel prime object does facilitate compatible pre-cision and power grips, even when both the object itselfand the grasp type are irrelevant to the current task.However, this effect was only found when the precisiongrip was made with the right hand (RH) and the powergrip was made with the left hand (LH). When these gripswere made with the opposite hands, the effect was ab-sent. Experiment 2 replicated the LH bias for large ob-jects and the RH bias for small objects when power andprecision grip responses were replaced with simple LHand RH button-press responses. It appears that the twohemispheres are specialised with regard to precision andpower compatible objects.

Keywords Asymmetries Æ Precision Æ Power Æ Grasp ÆAffordance

Introduction

Recently there has been a great deal of interest in themechanisms underlying the planning and control ofvisually guided movements. Researchers have shownmanual asymmetries in visually guided movements.Furthermore, the reach and grasp components of reach-to-grasp movements may be controlled in part bydifferent central structures (Jeannerod 1981). Despite

evidence of manual asymmetries in visually guidedreaching, manual asymmetries in grasping have not beenstudied to the same extent. The principal goal of thispaper, therefore, was to examine manual asymmetries ingrasping. More specifically, this paper investigatedwhether the programming of precision and power grips(as engineered through visual priming rather than visualguidance), might be lateralised in the brain.

Virtually all people prefer one hand to the other inmaking skilled movements. A majority of the populationare more proficient with their right hand (RH) than theirleft hand (LH). The laterality of manual movements hasbeen thought to be the product of the specialisation ofeach hemisphere for different cognitive, visual and/ormotor information processing functions (e.g. Goodale1990). A goal-directed manual aiming task (Woodworth1899) has been one of the most common methods inresearch on manual asymmetries in visually guidedmovements. This has demonstrated faster and moreaccurate aiming movements of the RH (e.g. Fisk andGoodale 1985; Elliott et al. 1993), and a RH superiorityin making small adjustments to the movement trajectoryas the hand approaches the target location (e.g. Mies-chke et al. 2001). This is often attributed to a greaterability of the left hemisphere in processing the percep-tual and/or motor information required for motor con-trol during ongoing movements (e.g. Annett et al. 1979).Alternatively, it has been suggested that the RH systemmay be more proficient at the utilisation of kinestheticfeedback (Woodworth 1899). Neurophysiological andneuropsychological research suggests that the lefthemisphere is associated with the computation of manycognitive-motor processes such as the selection of motorprograms for sequential movements (e.g. Kimura andArchibald 1974). However, a dominant arm advantagein reaching accuracy is not evident during ‘‘ballistic’’(low-precision, high-speed) movements and could beobserved only when the precision requirements of a taskare increased (e.g. Todor and Cisneros 1985).

In a manual aiming task, participants point at atarget. However, more typical manual actions involve

L. Vainio (&) Æ R. Ellis Æ M. Tucker Æ E. SymesSchool of Psychology, University of Plymouth,Drake Circus, PL4 7AA, Plymouth, EnglandE-mail: lari.vainio@plymouth.ac.ukTel.: +44-1752-233146

Exp Brain Res (2006) 173: 395–406DOI 10.1007/s00221-006-0378-x

reaching and grasping. Jeannerod (1981) separated hu-man prehension into two independent motor programs,reaching and grasping, that involve separate brain re-gions. More recently, Jeannerod et al. (1995) have sug-gested that planning the reach component in prehensionmovements is largely based on analysing the spatialattributes of the target object such as distance anddirection. In contrast, planning the grasp itself is largelybased on an analysis of the object’s intrinsic propertiessuch as size (see also Roy et al. 2002 for an argumentfavouring partially independent and inter-related visuo-motor channels).

Support for the distinct visuo-motor channel hypoth-esis comes from evidence that separate circuits transformvisual information into motor codes in reaching andgrasping. In monkeys, the connections between theanterior intraparietal area (AIP) in the posterior parietalcortex (PPC) and the F5 neurons in the premotor cortex,code information relevant to grasping, namely intrinsicobject properties such as size and shape (e.g. Jeannerodet al. 1995). In sharp contrast, the ventral intraparietalarea (VIP) in the PPC codes object position and orien-tations in peripersonal space. This information about atarget is passed to the F4 neurons in the premotor cortex,which represents the arm’s goal position and is thenresponsible for setting up the initial reach program(Colby et al. 1993). Thus, the connections between areathe VIP and the F4 form a circuit that transforms infor-mation relevant to reaching, namely visual informationabout an object’s position (see Rizzolati et al. 1998 for areview). Human AIP is a likely homologue of macaqueAIP, an area with neurons that are activated by theviewing and grasping of specific shapes. Like macaqueAIP, humanAIP demonstrates activation during both thevisual and somatomotor phases of a delayed graspingtrial (Culham 2004). In light of this kind of evidence, it ispossible that when people plan and execute actionstowards objects of different positions, sizes, orientationsand shapes, the spatial attributes of the target areprimarily analysed for programming reaches, whereas theintrinsic attributes of the target are primarily analysed forprogramming grasps.

Napier (1956) divided grips into precision and powergrips from a functional and a phylogenetic perspective.The precision grip (the use of a thumb–index grip) hasdeveloped in primates for manipulation of small objectswhereas the power grip has developed for holding andgrasping larger objects with high stability. There is someevidence that a precision grip engages neural circuitsthat are different from those engaged during power grips(e.g. Ehrsson et al. 2000). Interestingly, some research inmonkeys suggests manual asymmetries in computingprecision and power grips. For example, Hopkins et al.(2002) showed that in chimpanzees the RH is morefrequently used in making precision grip. Furthermore,the dominant and non-dominant hands have specialroles in bi-manual manipulation movements (the stabi-lizing function of the non-dominant hand and themanipulative function of the dominant hand). The

optimal hand organisation in bi-manual movementsmight contribute to pressure for an anatomical separa-tion of the two grip types.

This present paper asks whether, similar to reaching,grasping also has manual asymmetries in humans. Spe-cifically, this paper investigates whether or not our pro-gramming of power and precision grips is manuallyasymmetric. In normal reach-to-grasp paradigms, it isdifficult to measure the respective roles of reaching andgrasping in manual asymmetries. However, because thepresent investigation concerns manual asymmetry ofprecision and power grip programming, it was essentialthat reach programming could be ruled out. In requiringan experimental task that only required participants toplan and execute power or precision grips, with noreaching involved, this study used the stimulus-response(S-R) compatibility paradigm presented by Tucker andEllis (2001). In this paradigm, the size of the viewed objectfacilitates precision and power grip responses that aremade to some object property, even though the object sizeis irrelevant and participants are not grasping the targetobject. This paradigm is discussed in more detail below.

Object affordances

Kinematic studies (Jeannerod 1988; Jakobson andGoodale 1991; Goodale et al. 1994; Gentilucci 2002)show that object affordances (i.e. action-relevant attri-butes of objects) influence both the selection of the typeof grip and the grasp kinematic implementation. Morerecently, Tucker and Ellis (1998, 2001, 2004) and Ellisand Tucker (2000) have demonstrated in a series of S-Rcompatibility studies that a viewed object (or photo-graph of an object) automatically facilitates responsesthat are compatible with the action-relevant attributes ofthat object (such as size and orientation). This occurs,despite there being no intention to act on the objectitself. Of most relevance to present purposes is a studyby Tucker and Ellis (2001), which examined precisionand power grip activation initiated by the size of a seenobject. In this study, participants viewed graspable ob-jects that belonged to ‘‘manufactured’’ or ‘‘natural’’categories. Half of the objects in both categories weresmall and would normally be grasped with a precisiongrip (e.g. grape, screw) and half were large and wouldnormally be grasped with a power grip (e.g. cucumber,hammer). Participants held two response devices, eachequipped with an inlaid micro-switch, simultaneously intheir dominant hand. The precision grip device wassmall and square, and was held between the index fingerand the thumb. The power grip device was larger andcylindrical, and was held in the palm of the hand bywrapping the remaining fingers around it. Participantswere asked to judge whether the object belonged to amanufactured or natural category by squeezingthe precision or power device. It was found that thoseresponses that were compatible with the presented objectsize were significantly faster and more accurate. For

396

convenience, we shall refer to this finding as the‘‘object-size effect’’. Tucker and Ellis (2001) also ob-served the object-size effect in a bi-manual task whenone response device was held in one hand and the otherdevice was held in the opposite hand. The object-sizeeffect has also been observed when participants catego-rized names (i.e. written words) of the objects that couldbe normally grasped with precision grip or power grip(Tucker and Ellis 2004) suggesting that the effect oper-ates at the level of grip planning.

We suggest that the object-size effect is consistentwith the possibility that action-relevant object properties(in this case size) activate already existing motor repre-sentations of an object’s afforded actions. This, in turn,suggests that the motor system has a central role in vi-sual object representation, whether or not an action thatis primed by the viewed object is actually executed.Murata et al. (1997) have presented electrophysiologicalevidence supporting this view. They demonstrated thatneurons in the F5 and the AIP, which were shown toform a parieto-frontal circuit that transforms visualinformation for grasp computing (as mentioned above),fire during the grasping and additionally discharge to thepresentation of graspable objects, even when no imme-diate action upon the object is allowed. Fagg and Arbib(1998) have developed a neural network model, in whichthe connections between these same areas in the PPCand premotor cortex play a principal role in the gener-ation of object affordances. Furthermore, the view thatactions are encoded as a part of object representation isalso supported by brain-imaging studies (e.g. Martinet al. 1995; Chao and Martin 2000; Grezes and Decety2002; Handy et al. 2003).

Finally, it should be mentioned that, in the object-sizeeffect, the grip is not ‘‘visually guided’’ in the same sensethat movements are guided in traditional visually guidedmotor tasks because, in the object-size task, participantsare not required to grasp the target object. Rather thesize of the target object facilitates precision and powergrip responses that are performed with separate devices.However, we presume that, in the object-size effect, theinfluence of the object size on grasp reflects the sameunderlying action programming mechanisms, which areoperating in grasp programming of traditional visuallyguided reach-to-grasp tasks. Consistent with this view,human AIP is activated in the object-size effect task(Grezes et al. 2003). Thus, we assume that these neurons(at least partly) underlie the grasp potentiation in theobject-size effect, and consequently the effect reflects agrasp plan in the reach-to-grasp movement. Further-more, we propose that the object-size effect reflects theplanning of the optimal, final arm posture of grasping inprehension. Therefore, given that the two hands showdifference in actions requiring precise motor control, wepredict that object size may afford asymmetrically dif-ferent hands. Thus, we expect precision grip objects toafford the grip type and the hand (the precision grip andthe RH, respectively).

Experiment 1

Given that the RH is normally more accurate in visuallyguided reaching, and given that this accuracy decreasesif the precision requirements of the task are lowered, wedecided to examine whether similar asymmetries couldbe observed with visually primed grasping. It was hy-pothesised that the RH would show superiority over theLH for precision grips, but not for power grips. In orderto test this hypothesis, the following response assign-ments were made: half of the participants held a preci-sion grip device in their LH and a power grip device intheir RH (mapping 1) and the other half held a precisiongrip device in their RH and a power grip device in theirLH (mapping 2).

While the primary aim of this experiment was to ex-plore manual asymmetries relating to power and preci-sion grips, there were several secondary aims.

1. Object priming: the object-size effect of Tucker andEllis (2001) was observed when participants were re-quired to categorise a viewed object. Such categori-sation required participants to focus attention on theobject. However, it is not clear whether the same ef-fect would be observed when the allocation ofendogenous attention to the object is minimal orabsent. Therefore, it is particularly interesting toexamine whether an action-relevant property of anobject (in this case size) could influence responseswhen the viewed object is task irrelevant (i.e. it doesnot require endogenous attention). The currentexperiment examines this aspect of attention by pre-senting participants with task-irrelevant prime ob-jects, and asking them to respond to a target arrowthat is superimposed over the prime.

2. Response dimensions: in Tucker and Ellis (2001)participants were asked to respond with a precisionor power grip to the object category. The grasp typewas therefore a task-relevant response dimension,and consequently participants were likely to codetheir responses explicitly as precision and powergrips. The current experiment investigates whetherviewed objects facilitate the precision and power gripresponses even though participants are instructed torespond with their LH or RH. Thus, the grasp type isa task-irrelevant response dimension, and partici-pants are not likely to code their required responsesexplicitly as precision and power grips. This taskarrangement minimizes the influence of cognitivefactors related to grip selection.

3. Novel objects: in Tucker and Ellis (2001) participantshad to categorise familiar objects. Familiar objectshave semantic associations. Indeed, as the Tuckerand Ellis (2004) study suggests, even categorizing anobject name (rather than a photograph of that object)can produce the object-size effect (presumably be-cause participants have semantically derived the sizeof the object). It has not been established whether the

397

purely visual size-related information of an object canfacilitate compatible grasps. We therefore investi-gated whether the object-size effect would be forth-coming using novel objects that did not carrysemantic connotations. The object set consisted ofrealistic three-dimensional (3D) objects never previ-ously seen by the viewer (see Fig. 1).

4. Time course: Finally, the time course of responseactivation in the object-size effect has not been re-ported in relation to two responding hands. The LHand RH might show differential time courses for re-sponse activation. We therefore examined the timecourse of response activation by varying the onsettime between the prime object and target.

Method

Participants

Forty-four participants took part in the experiment andwere each run in individual 25 min sessions. All werestudents at the University of Plymouth and receivedcourse credit for their participation. Informed consentwas obtained from each subject prior to commencing thetask. All participants reported having normal or cor-rected-to-normal vision and were naive as to the purposeof the experiment. Only the data of the right-handedparticipants was included in the analysis. However,participants were not told that only the right-handers’data would be analysed. All but two participants signedthe participation form using their RH. The participants’explicit report about their handedness (asked after

signing the participation form) was consistent with theobserved writing hand. The two participants who signedthe form using their LH also completed the experimentand were credited for their participation but their datawere not used in the analysis. This arrangement wasassumed to minimise the chance that participants wouldfalsely report their handedness for receiving the credit.

Apparatus and stimuli

The display and timing was controlled by a RM-Accelerator-Intel: Pentium 2 processor computer, inter-faced to a Mitsubishi Diamond Pro900u 19in. colourmonitor. There were two response devices, each equip-ped with an inlaid micro-switch. The precision gripdevice was small and square (1.3·1.3·0.7 cm3) and thepower grip device was larger and cylindrical (11 cmlong, 1.8 cm diameter). As the switches depressed ineach device, there was noticeable tactile and auditoryfeedback. The prime stimuli consisted of twenty-fourcomputer generated 3D objects (see example in Fig. 1).Each object had a slightly different ‘‘wood’’ texture andeach object had a slightly different variation of a naturalbrown wood colour. Half of the objects were small andtherefore more suitable to be grasped with a precisiongrip (they subtended a visual angle of approximately2.3� vertically and 2.9� horizontally). Small objects werein the shape of a ball, cone or cylinder. Half of theobjects were large and would normally be grasped with apower grip (they subtended a visual angle of approxi-mately 17.1� vertically and 4� horizontally) and con-formed to a grasp-appropriate shape. Large objects werein the shape of a cylinder or capsule shape. Objects haddifferent shapes and surface textures to increase thelikelihood of automatic attentional capture by a novelattribute (see Ruz and Lupianez 2002 for a review ofattentional capture). In addition to the prime objectstimuli, there was a centrally located black fixation cross(1·1�), and two centrally located black target arrows(1·1�), one pointing left and the other right (these wereinterchanged in a randomised order). All stimuli werepresented against a white background and presented onthe monitor at a resolution of 1,024·768 pixels.

Design and procedure

Participants sat in front of a monitor in a dimly illu-minated room with their eyes 50 cm from the centre ofthe monitor. The height of the monitor was adjusted sothat each participant was looking directly at the centreof the display. Participants held the power grip device intheir RH and the precision grip device in their LH(mapping 1; 21 participants were run in this mapping),or vice versa (mapping 2; 21 participants were run in thismapping). Participants were familiarized with theswitches. They were shown how to squeeze the precisiongrip device with the index finger and thumb, and thepower grip device with a whole hand grasp. ParticipantsFig. 1 Example prime object stimuli used in experiments 1 and 2

398

were instructed to keep both their arms on the table onwhich the monitor was placed (40 cm apart and 25 cm infront of the monitor). The leads of both response deviceswere attached to the table so that the participant’s armplacement was consistent.

Each trial was initiated by a black fixation cross.After 1,500 ms the fixation cross was replaced by arandomly selected prime object, which was presented inexactly the same central location as the fixation cross.Prime objects appeared standing vertically and thereforethey were equally compatible with a RH or LH grasp.Three stimulus-onset asynchrony (SOA) conditions(150, 300, 600 ms) determined the duration of the primeobject presentation. After the SOA period the target (theleft or right pointing arrow) was displayed over theprime object in the same location that the fixation crosshad previously occupied. The target arrow changed backinto the fixation cross after 180 ms. The prime objectand the fixation cross were presented until the partici-pant responded. Participants were instructed to respondas quickly as possible with their RH when they saw theright-pointing arrow. Similarly, participants were in-structed to respond with their LH when they saw theleft-pointing arrow. The participant was asked to focusupon the central point through the whole experiment.Participants understood that maintaining fixation at thecentral locus was the most efficient strategy whenattempting to detect a brief target. In addition, partici-pants were told that the objects that were displayedbefore the appearance of the target were absolutelyirrelevant to the task and therefore could be ignored.Error responses were immediately followed by a short‘‘beep’’-tone from the computer. Participants were timedout if they did not respond within 3,000 ms. A halfminute break divided the experiment into three blocks.Each block consisted of a different set of object stimuli.The objects were randomly assigned to one of the threeblocks. During the break, the monitor displayed textthat indicated the length of the break and instructionsfor carrying on with the experiment. The experimentaldesign is illustrated in Fig. 2.

Results

Response times

The experiment consisted of 432 trials. Reaction times(RTs) were cropped for each participant’s data. RTs twostandard deviations from each participant’s overallmean were discarded (2.6%). Condition means for theremaining data were computed. Error trials were ex-cluded from the analysis. Condition means were sub-jected to a repeated measures ANOVA with the withinparticipants factors of prime object size (small or large),SOA (150, 300 or 600 ms), grip type (precision orpower), and the between participants factor of mappingrule (M1: RH/power grip, LH/precision grip and M2:RH/precision grip, LH/power grip).

The analysis revealed two significant main effects andtwo significant two-way interactions that were of sec-ondary interest to this study. There was a main effect ofgrip type, F(1,40)=12.59, P=0.001, MSE=6,113.64;the power grip responses were made faster (M=278 ms)than the precision grip responses (M=285 ms). Addi-tionally, there was main effect of SOA, F(1.7,67.4)g=20.79, P<0.001, MSE=4,694 (‘‘g’’ superscript indicatesthe use of Greenhouse–Geiser adjustments to the de-grees of freedom for effect tests). Responses were madeslower in SOA 150 ms (M=287 ms) than in SOA300 ms (M=277 ms) or SOA 600 ms (M=280 ms), Inaddition, the analysis revealed two two-way interactions.First (size · mapping), in mapping 1, participants madefaster responses when the prime object was large(M=273 ms) rather than small (M=276 ms), and inmapping 2, participants made faster responses when theprime object was small (M=287 ms) rather than large(M=289 ms), F(1,40)=5.16, P=0.029, MSE=581.75.

The analysis revealed a statistically significant two-way interaction between object size and grip type,F(1,40)=23.21, P<0.001, MSE=1,804.23. However,this interaction differed in the two mappings because theanalysis also revealed a statistically significant three-wayinteraction between prime object size, grip type and

Fig. 2 A schematic of thedesign for experiments 1 and 2.Note that the design ofexperiment 2 included one moreSOA condition (1,000 ms)

399

mapping, F(1,40)=27.74, P<0.001, MSE=2,156.26.This interaction is of most interest because the primaryaim of this experiment was to establish whether or notprecision and power grip planning is associated withmanual asymmetries. Therefore, a separate analysis(a repeated measures ANOVA) of the simple interactioneffect of size by grip at each mapping was carried out.

Mapping 1 The analysis did not reveal a statisticallysignificant interaction between prime object size and griptype [F(1,20)=7.84, P=0.668, MSE=0.189] indicatingthe absence of an object-size effect. The mean RTs inmapping 1 by grip type and prime object size are dis-played in Fig. 3.

Mapping 2 Most importantly, this analysis revealed astatistically significant two-way interaction betweenprime object size and grip type [F(1,20)=34.65,P<0.001,MSE=3,952.65] indicating an overall advantage of pre-cision grip (RH) responses for small prime objects andpower grip (LH) responses for large prime objects. Par-ticipants made faster precision (RH) grip responses whenprime objects were small (M=284 ms) rather than large(M=294 ms). Similarly, participants made faster power(LH) grip responses when prime objects were large(M=283 ms) rather than small (M=290 ms). The meanRTs in mapping 2 by grip type and prime object size aredisplayed in Fig. 4.

Supplementary analysis

We carried out the supplementary analysis in order toexplore whether the time course of the grip congruencyeffect was different for the different grip types. Theoriginal analysis only allowed testing the overall size-by-grip-type congruency effect across SOAs. Therefore,

we replaced the object size factor by the congruencybetween object size and grip type. Because we wereinterested in time courses of the response activationbetween grip types in mapping 2, we carried out aseparate analysis for mapping 2. The supplementaryanalysis was restricted to mapping 2 as there was noevidence for any object-size effect in mapping 1.Therefore, the condition means were subjected to arepeated measures ANOVA with the within partici-pants factors of size-grip congruency (congruent orincongruent), SOA (150, 300 or 600 ms), grip type(precision or power). This analysis revealed that thethree-way interaction between congruency, grip typeand SOA was significant in mapping 2, F(2,40)=12.38,P<0.001, MSE=456.21. Figure 5a, b display thedevelopment of the precision grip (RH) priming effect(Fig. 5a) and diminishment of the power grip (LH)priming effect (Fig. 5b).

Errors

The mean error rate was 2.8%. Analysis of the per-centage error rates revealed a statistically significantmain effect of grip type. Participants made more errorswith the precision grip (M=3.5%) than the power grip(M=2.1%), F(1,40)=21.77, P<0.001, MSE=266.76.In addition, the analysis revealed a statistically signifi-cant two-way interaction between grip type andmapping. In mapping 1, participants made less errorswith the power (RH) grip (M=1.7%) than with theprecision (LH) grip (M=4.1%). Similarly, in mapping 2,participants made less errors with the power (LH) grip(M=2.4%) than with the precision (RH) grip(M=3.0%), F(1,40)=7.60, P=0.009, MSE=93.14. Theanalysis also revealed a statistically significant three-wayinteraction between prime object size, grip type andmapping, F(1,40)=15.75, P<0.001, MSE=129.58. Thisinteraction is of interest, and therefore a separate

Fig. 3 Mean RTs (Experiment 1) in Mapping 1 by response [right-hand (RH)/power grasp and left-hand (LH)/precision grasp] andprime object size (small and large)

Fig. 4 Mean RTs (Experiment 1) in Mapping 2 by response [right-hand (RH)/precision grasp and left-hand (LH)/power grasp] andprime object size (small and large)

400

analysis (a repeated measure ANOVA) of the simpleinteraction effect of size by grip at each mapping wascarried out.

Mapping 1 The analysis revealed a statistically signifi-cant two-way interaction between prime object size andgrip type. Participants made fewer errors with the power(RH) grip when the prime object was small (M=1.5%)rather than large (M=2.0%). Similarly, participantsmade fewer errors with the precision (LH) grip when theprime object was large (M=3.5%) rather than small(M=4.63%), F(1,20)=5.39, P=0.031, MSE=46.57.This was true across all SOAs [size · SOA · grip,F(2,40)=2.29, P=0.115, MSE=25.07]. This result sug-gests that when the grip type corresponds with the primeobject size, participants make more errors. The meanerror percentage in mapping 1 by response and primeobject-size is displayed in Fig. 6.

Mapping 2 This analysis revealed a pattern of resultssimilar to those found for response times in mapping 2.

The analysis revealed a statistically significant two-wayinteraction between prime object size and grip type.Participant made fewer errors with the power (LH) gripwhen the prime object was large (M=1.8%) rather thansmall (M=3.1%). Similarly, participants made fewererrors with the precision (RH) grip when the primeobject was small (M=2.5%) rather than large(M=3.5%), F(1,20)=10.99, P=0.003, MSE=11.15.This was true across all SOAs [size · SOA · grip,F(2,40)=.71, P=0.498, MSE=4.81]. The mean errorpercentage in mapping 2 by response and prime objectsize is displayed in Fig. 7.

Interestingly, the patterns of the two-way interactionbetween grip type and prime object size were opposite inthe two mappings. Therefore, it seems more likely thatthe statistically significant two-way interaction found inboth mappings of the error data reflects an interactionbetween hand of response and prime object size ratherthan simply an interaction between grip type and primeobject size. That is, participants appeared to favourresponding with their RH when the prime object wassmall and with their LH when the prime object waslarge.

Discussion

In accordance with our primary hypothesis, we foundmanual asymmetries in visually primed grasping. Theanalysis of RTs revealed the object-size effect only inmapping 2 when precision grip responses were madewith the RH and power grip responses were made withthe LH. The size of the prime object did not seem toinfluence planning in mapping 1. Furthermore, theanalysis of errors revealed that participants were moreaccurate responding with their RH when the prime ob-ject was small and with their LH when the prime objectwas large regardless of which grip device was held ineach hand. This result is consistent with our predictionthat object-size effect can be linked not only to optimalgrip type but also to optimal hand that would perform

Fig. 5 a The development of the object-size effect (congruentresponses) in mapping 2 of experiment 1 for precision grips (RH)across the three SOAs. b The diminishment of the object-size effect(congruent responses) in mapping 2 of experiment 1 for power grips(LH) across the three SOAs

Fig. 6 Mean error percentage (Experiment 1) in Mapping 1 byresponse and prime object size

401

that grip if the object would be reached-and-grasped to.Finally, the object-size effect, which was observed inmapping 2, had a differential time course in relation toLH and RH responses. The affect that large prime ob-jects had on LH power grip responses appeared to de-velop and diminish rapidly after the onset of the primeobject, whereas the affect that small prime objects hadon RH precision grip responses appeared to developgradually.

Experiment 2

The results of experiment 1 suggested that a prime ob-ject’s size only facilitated grasping when the power gripdevice was held in the LH and the precision grip devicewas held in the RH. Furthermore, the error data in bothgrip type-to-hand mappings suggested that participantsfavoured responding with their RH when the primeobject was small and with their LH when the primeobject was large in both grip type-to-hand mappings.

The possibility that the pattern of errors in experi-ment 1 reflects the facilitation of the hand (optimal ingrasping the different sized objects) rather than grip isintriguing. A preference for responding with a particularhand (e.g. right) when primed by a particular object size(e.g. small), points to the possibility of hemisphericspecialisation in object coding for action selection. Largeobjects may be primarily coded in the right hemisphere(which controls the LH system), and small objects maybe primarily coded in the left hemisphere (which con-trols the RH system). This suggests that grip responseswere not primed in mapping 1 (Experiment 1) because ofthe conflict between execution operations and planningoperations (i.e. the execution of the grip and the plan-ning of the very same grip were processed in oppositehemispheres). Recall from the introduction the proposalthat action-relevant object properties (in this case size)activate already existing motor representations of anobject’s afforded actions. The optimal motor represen-tation for small objects would be precision grip plan andRH plan. Experiment 2 tested the possibility that theobject size does not afford only the optimal grip type but

also the optimal hand of response by replacing powerand precision grip responses with simple LH and RHbutton-press responses. The following manual asym-metries were predicted: (1) Small prime objects (thatafford grasping with a precision grip) should facilitateRH button-presses; (2) Large prime objects (that affordgrasping with a power grip) should facilitate LH button-presses.

Method

Participants

Twenty-two new participants took part in the experi-ment and were each run in individual 25 min sessions.All were students at the University of Plymouth andreceived course credit for their participation. Informedconsent was obtained from each subject prior to com-mencing the task. All participants reported having nor-mal or corrected-to-normal vision and were naive as tothe purpose of the experiment. Experiment 2 used thesame procedure for testing participants’ handedness asexperiment 1. All participants signed the participationform using their RH and additionally explicitly reportedthat they were right handed.

Apparatus and stimuli

Apparatus, fixation point, prime objects and target ar-rows were same that those used in experiment 1 (notethat power and precision devices were not used in thisexperiment).

Design and procedure

The design and procedure were identical to that of thefirst experiment except that participants responded tothe target arrow by pressing the corresponding left ‘‘z’’or right ‘‘2’’ key of a standard computer keyboard withthe corresponding index finger. Furthermore, in addi-tion to the three same SOA conditions (150, 300,600 ms), which determined the duration of object pre-sentation in experiment 1, one additional SOA condition(1,000 ms) was introduced to the design. This was tofurther examine manually asymmetrical time courses ofthe facilitatory effect found in experiment 1 (mapping 2).

Results

Response times

The experiment consisted of 432 trials. Data processingand analysis were the same to the first experiment. The2SD cut-offs removed 2.5% of the data as outliers.Condition means were the within participants factors ofprime object size (small or large), SOA (150, 300, 600 or

Fig. 7 Mean error percentage (Experiment 1) in Mapping 2 byresponse and prime object size

402

1,000 ms), and hand of response (right or left). Theanalysis revealed statistically significant main effectsof SOA (SOA 1—376 ms; 2—364 ms; 3—359 ms;4—362 ms), F(3,63)=23.71, P<0.001, MSE=4,834.99and hand of response (left—373 ms; right—358 ms),F(1,21)=11.76, P=0.003, MSE=20,336.16. In addition,the analysis revealed a statistically significant two-wayinteraction between prime object size and SOA,F(3,63)=17.59, P<0.001, MSE=1,571.52. Participantsresponded faster in longer SOAs than in shorter SOAswhen the prime object was small (SOA150: M=383 ms;SOA300: M=366 ms; SOA600: M=356 ms; SOA1000:M=361 ms). This was not the case when objects werelarge (SOA150: M=369 ms; SOA300: M=363 ms;SOA600: M=362 ms; SOA1000: M=364 ms).

Most importantly, the analysis revealed a statisticallysignificant two-way interaction between prime objectsize and hand of response. Participants made faster RHresponses when the prime object was small (M=357 ms)rather than large (M=359 ms). Similarly, participantsmade faster LH responses when the prime object waslarge (M=370 ms) rather than small (M=376 ms),F(1,21)=22.49, P<0.001, MSE=1,603.08. The meanRTs by hand of response and prime object size are dis-played in Fig. 8.

Supplementary analysis

We carried out the supplementary analysis in order toexplore whether the differential time courses of thepriming effect between grip types, which was observed inexperiment 1, would be replicated in this second exper-iment. As already stated, the original analysis only al-lowed testing the overall size by grip congruency effectacross SOAs. Therefore, we replaced the object sizefactor by the congruency between object size and handof response. The results of experiment 1 suggested that

RH responses are facilitated by small objects and LHresponses are facilitated by large objects. Therefore, inthis case, we mean by congruency the predicted rela-tionship between object size and hand of response (LH-large objects/RH-small objects). The condition meanswere subjected to a repeated measures ANOVA with thewithin participants factors of size-hand congruency(congruent or incongruent), SOA (150, 300, 600 or1,000 ms), and hand of response (left or right). Thisanalysis revealed that the three-way interaction betweencongruency, SOA and hand of response was significant,F(3,63)=17.59, P<0.001, MSE=1,571.52. Figure 9a, bdisplay the development of the RH priming effect(Fig. 9a) and diminishment of the LH priming effect(Fig. 9b). The pattern of these time course differences isconsistent with the results of experiment 1.

Errors

The mean error rate was 2.5%. Analysis of the per-centage error rates did not reveal any statistically sig-nificant main effects or interactions.

Discussion

The results of experiment 2 documented manual asym-metries and hand-related time courses that were con-sistent with the results of experiment 1. The button-pressresponses of the RH were primed predominantly bysmall objects while the button-press responses of the LHwere primed by large objects. In addition, time coursesof these effects developed in the same way in the currentexperiment as they had developed when responses wereperformed with grip devices in the previous experiment(mapping 2). The results of experiment 2 support theview that performance benefits are associated with theRH when the prime object is small, and with the LHwhen the prime object is large.

General discussion

The main finding of the present study was that (task-irrelevant) small objects that require more precise graspplanning prime RH responses while large objects that donot require such precise grasp planning prime LH re-sponses. In experiment 1, prime object size was observedto influence the speed of precision and power grip re-sponses in mapping 2 when the precision grip was madewith the RH and a power grip was made with the LH. Incontrast, in mapping 1, when the grips were held in theopposite hands, the size of the prime object was notobserved to influence responses. Therefore, there seemto be manual asymmetries in visually primed precisionand power grips. The LH bias for large objects and RHbias for small objects was replicated in experiment 2when power and precision grip responses were replaced

Fig. 8 Mean RTs (Experiment 2) by response (RH and LH) andprime object size (small and large)

403

with simple LH and RH button-press responses. Asmentioned in the Introduction, studies examining visu-ally guided reaching indicate that the RH is superior inskilled and precise movements. This RH advantageseems to decrease when the precision requirements of thetask decrease. Our data, which showed manual asym-metries in visually primed grasping, is consistent withthese earlier suggestions of the nature of manual asym-metries. This view assumes that precision grip planningmay be biased to the left hemisphere. In contrast, theright hemisphere may code predominantly manual re-sponses for large objects that do not require precisegrasp planning.

In addition to finding manual asymmetries in RTs,the error data of experiment 1 suggested that the LHsystem had been facilitated automatically by viewing aprime object that afforded a power grip (e.g. a largeobject), whereas the RH system had been facilitated

automatically by viewing a prime object that afforded aprecision grip (e.g. a small object). In other words, theerror data suggested that responses are performedpreferably with the RH when small prime objects areviewed and with the LH when large prime objects areviewed, regardless of whether the hand was holding theprecision or power grip. The RT data of experiment 2supported the view that the object size does not affordonly the optimal grip type but also the hand.

The AIP area in both macaques and humans has beensuggested to have an important role in selecting the griptype when visual information is used to preshape thehand to grasp an object. Culham et al. (2001) examinedthe laterality of fMRI activation in right-handed sub-jects who either grasped objects with a precision grip orimagined grasping them using one hand or the other.The activation in the AIP was contralateral for the RHand bilateral for the LH. The authors suggested that thepredominance of activation seen in the left AIP in allconditions may reflect the special role that the lefthemisphere plays in the visual control of skilled move-ments with either hand. However, as mentioned in theIntroduction, the AIP appears to play an important rolenot only in the grasp control but also selection becausethe AIP neurons discharge to the presentation ofgraspable objects, even when no immediate action uponthe object is allowed (Murata et al. 1997). Our results arein line with the assumption that the left hemisphere has aspecial role in the initial planning of the visually primedprecision grip. In turn, this bias may lead to the primingeffect of only RH-precision grip responses. In addition,our data suggests that this lateral bias of grip planningto the left hemisphere might be linked particularly toprecision grip programming because the visual primingof power grip responses was associated with LH re-sponses.

Secondary findings

First, the results of experiment 1 and 2 suggested thatleft and RH responses were associated with differentialtime courses of the object-size effect. The performanceadvantage associated with the LH-power grip begins tobuild shortly after the onset of the prime object anddecreases with longer presentation times of the primeobject. In contrast, the performance advantage associ-ated with the RH-precision grip begins to build slightlylater and increases with longer presentation times of theprime object. This may be attributed to a greater abilityof the left hemisphere in processing perceptual infor-mation for generating the motor program. Interestingly,investigations into the influence of intermittent vision onmanual aiming have revealed that the RH system maybe better in using visual information than the LH systemin manual control (Elliott et al. 1994). It is possible thatthe same visuo-motor channel, which benefits controlof the RH, also leads to the gradually developingobject-size effect associated with the RH-precision grip

Fig. 9 a The development of the (object size-hand of response)congruency effect of experiment 2 for RH responses across the fourSOAs. b The diminishment of the (object size-hand of response)congruency effect of experiment 2 for LH responses across the fourSOAs

404

responses. However, the object-size effect ultimately re-flects manual planning whereas the aforementioned RHsuperiority in accessing visual information in an inter-mittent vision task is linked to manual control. Inaddition, planning and control may be subserved byseparate visual centres in posterior parietal lobes(Glover 2004). Nevertheless, irrelevant object sizeinformation has been observed to influence not only thegrip selection (Tucker and Ellis 2001) but also reaching-grasping kinematics (e.g. Gentilucci 2002). Thus, thesuggested link between the differential time courses thatwere found in the present experiment and the superiorityof the RH in movement control, whilst speculative, is inprinciple possible.

Second, the object-size effect was observed with task-irrelevant prime objects, suggesting that action-relevantobject properties can trigger an associated action planeven when endogenous attention to the object is minimalor perhaps even absent. Previous research has shownthat irrelevant object orientation can facilitate left–rightresponses when they correspond with the left–right ori-entation of an object (Phillips and Ward 2002). All ofthis evidence suggests that the object, which is not theintended target of response, is capable of priming actioncomponents that are associated with the object.

Third, prime objects facilitated precision and powergrip responses even when the grasp type was a task-irrelevant response dimension (recall that participantswere instructed to make left or right responses—the factthat these responses employed different grip types wasincidental to the participant). This suggests that theobject-size effect can be observed even when cognitivefactors related to grip selection are minimized. Thissuggests that the potentiation of a grasp component bythe object’s size in the object-size effect may operatedirectly, bypassing cognitive processing in responseselection. As already stated in the Introduction, theevidence from the AIP-F5 circuit, which suggested itplayed an important role in affordance generation,showed that the same neurons in this circuit that fire, forexample, during specific grasping, also fire selectivelywhen a monkey is passively viewing a graspable object.In our experiment, the object-size effect is observed evenwhen participants are simply viewing the prime objectwithout need to do anything with it. In fact, participantsare not even required to attend to the prime. The factthat both effects, the object-size effect and the automaticactivation of the AIP-F5 circuit, are observed under‘‘passive viewing’’ conditions supports the view that thehuman AIP neurons (human AIP is a likely homologueof macaque AIP) underlie at least partly the object-sizeeffect.

Fourth, the results showed that the size of a primeobject automatically facilitates size compatible responsesdespite the object having no semantic associations. Thisdemonstrates that the object-size effect can be extractedfrom purely visual object characteristics. The FARSmodel (Fagg and Arbib 1998), which was discussed inthe Introduction, proposed that visual information for

grasp planning can be inputted to the circuit, which isresponsible for affordance generation, via the dorsal andventral stream (see Ungerleider and Mishkin 1982;Milner and Goodale 1995, for a review of the dorsal andventral streams). Research suggests that the dorsalstream provides purely visual action-relevant informa-tion for action planning (Milner and Goodale 1995)whereas the ventral stream provides semantic action-relevant information (Jeannerod et al. 1994). Based onthis evidence, it may be speculated that the object-sizeeffect observed in the current study might be linked tothe dorsal stream processes.

Conclusion

Evidence from behavioural research suggests RH supe-riority in the control of visually guided reaching. How-ever, this superiority appears to decrease when the taskrequires less precise actions. Research into manualasymmetries has typically employed the manual aimingparadigm, revealing underlying mechanisms of the reachcomponent of a visually guided reach-to-grasp move-ment. The present study has revealed that visuallyprimed grasps conform to the same pattern of manualasymmetry found in visually guided reaching. The lefthemisphere seems to have a special role in planningprecision grips whereas the right hemisphere has a spe-cial role in planning power grips.

Acknowledgements This work was supported by ESRC researchgrant RES000220942. We thank two anonymous reviewers forextensive comments concerning this article.

References

Annett J, Annett M, Hudson PTW, Turner A (1979) The control ofmovement in the preferred and the non-preferred hands. Q JExp Psychol 31:641–652

Chao LL, Martin A (2000) Representation of manipulable man-made objects in the dorsal stream. Neuroimage 12:478–484

Colby CL, Duhamel JR, Goldberg ME (1993) Ventral intraparietalarea of the macaque: anatomical location and visual responseproperties. J Neurophysiol 69:902–914

Culham JC (2004) fMRI investigations of visually-guided grasping.In: Kanwisher N, Duncan J (eds) Functional neuroimaging ofvisual cognition—attention and performance XX. UniversityPress, Oxford, pp 417–438

Culham JC, Woodward SL, Milner AD, Gati JS, Menon RS,Goodale MA (2001) Laterality of fMRI activation in area AIPduring grasping and imagined grasping of visual targets. Inpaper presented at the annual meeting of the Society for Neu-roscience

Ehrsson H, Fagergren A, Jonsson T, Westling G, Johansson RS,Forssberg H (2000) Cortical activity in precision- versus power-grip tasks: an fMRI study. J Neurophysiol 83:528–536

Elliott D, Roy EA, Goodman D, Carson RG, Chua R, Maraj BKV(1993) Asymmetries in the preparation and control of manualaiming movements. Can J Exp Psychol 47:570–589

Elliott D, Chua R, Pollock BJ (1994) The influence of intermittedvision on manual aiming. Acta Psychol 85:1–13

Ellis R, Tucker M (2000) Micro-affordance: the potentiation ofcomponents of action by seen objects. Br J Psychol 91:451–471

405

Fagg AH, Arbib MA (1998) Modeling parietal-premotor interac-tions in primate control of grasping. Neural Netw 11:1277–1303

Fisk JD, Goodale MA (1985) The organization of eye and limbmovements during unrestricted reaching to targets in contra-lateral and ipsilateral space. Exp Brain Res 60:159–178

Gentilucci M (2002) Object motor representation and reaching-grasping control. Neuropsychologia 40:1139–1153

Glover D (2004) Separate visual representations in the planningand control of action. Behav Brain Sci 27:3–78

Goodale MA (1990) Brain asymmetries in the control of reaching.In: Goodale MA (ed) Vision and action: the control of grasp-ing. Ablex, Norwood, NJ, pp 14–32

Goodale MA, Meenan JP, Bulthoff HH, Nicolle DA, Murphy KJ,Racicot CI (1994) Separate neural pathways for the visualanalysis of object shape in perception and prehension. Curr Biol4:604–610

Grezes J, Decety J (2002) Does visual perception of object affordaction? Evidence from a neuroimaging study. Neuropsycholo-gia 40:212–222

Grezes J, Tucker M, Armony J, Ellis R, Passingham RE (2003)Objects automatically potentiate action: an fMRI study of im-plicit processing. Eur J Neurosci 17:2735–2740

Handy TC, Grafton ST, Shroff NM, Ketay SB, Gazzaniga MS(2003) Graspable objects grab attention when the potential foraction is recognized. Nat Neurosci 6:421–427

Hopkins WD, Cantalupo C, Wesley MJ, Hostetter AB, Pilcher DL(2002) Grip morphology and hand use in chimpanzees (Pantroglodytes): evidence of a left hemisphere specialization inmotor skill. J Exp Psychol Gen 131:412–423

Jakobson L, Goodale MA (1991) Factors affecting higher-ordermovement planning: a kinematic analysis of human prehension.Exp Brain Res 86:199–208

Jeannerod M (1981) Intersegmental coordination during reachingat natural visual objects. In: Long J, Baddeley A (eds) Attentionand performance IX. Erlbaum, Hillsdale, NJ, pp 153–168

Jeannerod M (1988) The neural and behavioural organization ofgoal-directed movements. Clarendon, Oxford

Jeannerod M, Decety J, Michel F (1994) Impairment of graspingmovements following a bilateral posterior parietal lesion.Neuropsychologia 32:369–380

Jeannerod M, Arbib MA, Rizzolatti G, Sakata H (1995) Graspingobjects: the cortical mechanisms of visuomotor transformation.Trends Neurosci 18:314–320

Kimura D, Archibald Y (1974) Motor functions of the left hemi-sphere. Brain 97:337–350

Martin A, Haxby JL, Lalonde FM, Wiggs CL, Ungerleider LG(1995) Discrete cortical regions associated with knowledge ofcolor and knowledge of action. Science 270:102–105

Mieschke PE, Elliott D, Helsen WF, Carson RG, Coull JA (2001)Manual asymmetries in the preparation and control of goal-directed movements. Brain Cogn 45:129–140

Milner AD, Goodale MA (1995) The visual brain in action.University Press, Oxford

Murata A, Fadiga L, Fogassi L, Gallese V, Raos V, Rizzolatti G(1997) Object representation in the ventral premotor cortex(area F5) of the monkey. J Neurophysiol 78:2226–2230

Napier J (1956) The prehensile movements of the human hand.J Bone Joint Surg 38B:902–913

Phillips J, Ward R (2002) S-R correspondence effects of irrelevantvisual affordance: time-course and specificity of response acti-vation. Vis Cogn 9:540–558

Rizzolati G, Luppino M, Matelli M (1998) The organization of thecortical motor system: new concepts. Electroencephalogr ClinNeurophysiol 106:283–296

Roy AC, Paulignan Y, Meunier M, Boussaoud D (2002) Prehen-sion movements in macaque monkey: effects of object size andlocation. J Neurophysiol 88:1491–1499

Ruz M, Lupianez J (2002) A review of attentional capture: on itsautomaticity and sensitivity to endogenous control. Psicologica23:283–309

Todor JI, Cisneros J (1985) Accommodation to increased accuracydemands by the right and left hands. J Mot Behav 17:355–372

Tucker M, Ellis R (1998) On the relations between seen objects andcomponents of potential actions. J Exp Psychol Hum PerceptPerform 24:830–846

Tucker M, Ellis R (2001) The potentiation of grasp types duringvisual object categorization. Vis Cogn 8:769–800

Tucker M, Ellis R (2004) Action priming by briefly presented ob-jects. Acta Psychol 116:185–203

Ungerleider LG, Mishkin M (1982) Two cortical visual systems. In:Ingle DJ, Goodale MA, Mansfield RJW (eds) Analysis of visualbehavior. MIT Press, Cambridge, MA, pp 549–586

Woodworth RS (1899) The accuracy of voluntary movement.Psychol Rev 3(Whole 13):1–14

406