Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Neuropsychologia

http://d0028-39

n CorrSensorim

E-malessandandrea.sthierry.p

1 Th2 Pr

de Laus

PleasNeur

journal homepage: www.elsevier.com/locate/neuropsychologia

Dissociating effect of upper limb non-use and overuse on space andbody representations

Michela Bassolino a,1,2, Alessandra Finisguerra a,1, Elisa Canzoneri b, Andrea Serino b,c,Thierry Pozzo a,d,n

a Robotics, Brain and Cognitive Sciences, Istituto Italiano di Tecnologia, Genova 16163, Italyb Laboratory of Cognitive Neuroscience, Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerlandc Department of Psychology, University of Bologna, 40127 Bologna, Italyd IUF, INSERM U1093 Cognition, Action et Plasticité Sensorimotrice, Université de Bourgogne, Dijon 21078, France

a r t i c l e i n f o

Article history:Received 1 June 2014Received in revised form7 November 2014Accepted 21 November 2014

Keywords:Peripersonal spaceBody representationImmobilizationPlasticityAction

x.doi.org/10.1016/j.neuropsychologia.2014.11.032/& 2014 Published by Elsevier Ltd.

esponding author at: IUF, INSERM U1093 Cotrice, Université de Bourgogne, Dijon 21078

ail addresses: michela.bassolino@epfl.ch (M. Bra.finisguerra@iit.it (A. Finisguerra), elisa.canzerino@epfl.ch (A. Serino), tpozzo@u-bourgogozzo@iit.it (T. Pozzo).ese authors contributed equally to this workesent address: Center for Neuroprosthetics, Éanne (EPFL), Grand-Champsec 90, CH-1951 Si

e cite this article as: Bassolino, M., etopsychologia (2014), http://dx.doi.or

a b s t r a c t

Accurate and updated representations of the space where the body acts, i.e. the peripersonal space (PPS),and the location and dimension of body parts (body representation, BR) are essential to perform actions.Because both PPS and BR are involved in motor execution and display the same plastic proprieties afterthe use of a tool to reach far objects, it has been suggested that they overlap in a unique representation ofthe body in a space devoted to action. Here we determined whether manipulating actions in space,without modifying body metrics, i.e. through immobilization, induces a dissociation of the plasticproperties of PPS and BR. In 39 healthy subjects we evaluated PPS and BR for the non-used right limb andthe overused left limb before and after 10 h of right arm immobilization. We observed that non-usereduces PPS representation around the immobilized arm, without affecting the metric representation (i.e.perceived length) of that limb. In contrast, overuse modulates the metric representation of the free arm,leaving PPS unchanged around that limb. These results suggest that the plasticity in PPS and BR dependson different mechanisms; while PPS representation is shaped as a function of the dimension of the actingspace, metric characteristics of BR are forged on a complex interplay between visual and sensorimotorinformation related to the body. This behavioral dissociation between PPS and BR defines a new scenariofor the role of action in shaping space and body representations.

& 2014 Published by Elsevier Ltd.

1. Introduction

To properly reach an object positioned in front of the body, thebrain needs to represent both size and location of the involvedbody parts and the space lying in between. Neuroscientists havestudied such body and space representations for many years. Onone side, since no direct sensory signals inform the brain of themetrics of different body parts (Longo and Haggard, 2010), a bodyrepresentation (BR) is generated from the integration of the so-matosensory, proprioceptive and kinesthetic signals coming from

28

ognition, Action et Plasticité, France.assolino),oneri@epfl.ch (E. Canzoneri),ne.fr,

.cole Polytechnique Fédéraleon, Switzerland.

al., Dissociating effect of upg/10.1016/j.neuropsychologi

skin, joints and muscles with visual information (De Vignemont,2010; Medina and Coslett, 2010; Serino and Haggard, 2010). On theother side, the representation of the action space has been studiedin monkeys (e.g. Rizzolatti et al., 1997) and humans (Farnè andLàdavas, 2000; Holmes et al., 2007; Serino et al., 2007) throughthe interaction between somatosensory information and visual oracoustic inputs, specifically when these occur within a limited areaaround the body, the peripersonal space (PPS). The encoding of thespatial position of external stimuli in a body-centered frame ofreference facilitates the “possibility to act in space,” in terms ofapproaching (Rizzolatti et al., 1997) and defensive responses(Graziano and Cooke, 2006). BR and PPS refer, by definition, todifferent sectors of the space: the former would be limited to thebody, whereas the latter includes the space surrounding the bodyitself (e.g. Cardinali et al., 2009a). However, previous reports havehighlighted that PPS and BR jointly support efficient motor beha-viors (e.g. Gallese and Sinigaglia, 2010) and show similar plasticeffects. For instance, they both extend after tool-use (Bassolinoet al., 2010; Canzoneri et al., 2013a; Cardinali et al., 2009b; Farnèand Làdavas, 2000; Holmes et al., 2007; Iriki et al., 1996; Maravita

per limb non-use and overuse on space and body representations.a.2014.11.028i

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

and Iriki, 2004; Maravita et al., 2002; Serino et al., 2007; Spositoet al., 2012) and rather at present there is no empirical evidence tosupport the dissociable effects on PPS and BR. Here, we in-vestigated the hypothesis of a potential dissociation between PPSand BR, by manipulating “the possibility of acting in space”without modifying the body structure. For this purpose, right armmovements were limited for 10 h through immobilization (Bas-solino et al., 2012) in a total of 39 healthy right-handed partici-pants. Given that it has been shown that the non-use of one arm issystematically associated with a compensatory overuse of the freelimb (Avanzino et al., 2011), in this study we then evaluated thePPS representation and the perceived length of both arms beforeand after immobilization. If PPS and BR rely on dissociable plasticmechanisms, upper limb non-use and overuse would differentlyimpact on PPS representation and on the perceived arm length.Alternatively, analog effects of immobilization on PPS and BRwould indicate a complete overlap in plastic effects of the tworepresentations.

2. Materials and methods

2.1. Subjects

The study included a total of 39 healthy subjects (20 males, 19females; age: 24.6373.09 years; range: 20–30 years). The parti-cipants were randomly assigned to 3 groups: Experiment 1, Ex-periment 3 and Experiment 3 (Exp. 1, Exp. 2 and Exp. 3), eachcomposed of 13 subjects matched for age and gender, as describedin the section below “Task Selection”. All the participants wereright-handed, as determined using the Edinburgh HandednessInventory (Oldfield, 1971). All of them had normal or corrected-to-normal vision and hearing, with no previous history of sensory ororthopedic problems for the upper limbs. The subjects, naive tothe purpose of the study, provided written informed consent andreceived an attendance fee at the end of the experiment. The studyprotocol was approved by the local ethics committee (ASL-3,“Azienda Sanitaria Locale”, Genoa) and was performed in ac-cordance with the Declaration of Helsinki.

2.2. Immobilization

The experiments were conducted for 2 consecutive days (thefirst day as baseline: PRE-test; the second one after immobiliza-tion: POST-test). The participants were tested around 6 p.m. Dur-ing both days, subjects spent 10 working hours in the laboratoryunder experimenter's visual control, performing daily life activities(i.e. reading or working at the computer). Particularly, the parti-cipants did not use any type of tool to act in the far space, such as acomputer mouse (Bassolino et al., 2010). On the second day, sub-jects were required not to use their right arm from the morning (8a.m.) to the evening (6 p.m.). A soft painless bandage was wrappedaround the subjects' hand and forearm, and a cotton support wasapplied to limit the arm movement and to keep the elbow joint at90° flexion (Avanzino et al., 2011; Bassolino et al., 2012). Duringimmobilization, the participants performed the same activities ofthe first day using only the left free limb. During the two testingdays, the left arm activity was monitored using an accelerometerset up in a multisensory actigraph (InnerView Professional, Sen-seWear PRO Armband), recording the cumulative amount of timespent (in minutes) during physical activity under a level of energyexpenditure set to the typical level of deskwork activity (Ains-worth et al., 2000).

At the end of the immobilization period, the experimenter re-moved the bandage, and the subjects were instructed not to usethe right arm until the end of the entire experimental session. The

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

overused and constrained arms were evaluated in a counter-balanced order.

2.3. Task selection

The exact number and functions of different body representa-tions are currently a matter of discussion (see Kammers et al.,2009). Thus, on account of this on-going debate, here we delib-erately decided to adopt the more neutral and generic term ofbody representation (BR), being well aware of the possibility toinclude in this definition rather different levels of body-relatedinformation processing in the brain. In particular, we refer to themetric features of BR. Accordingly, to assess a multisensory, high-level, mental representation of the body, processing several sen-sory cues to represent the size and position of body parts, we usedtwo different tasks previously employed to demonstrate plasticeffects induced by tool-use on body metric, that are the tactiledistance perception task and body-landmarks localization task(Canzoneri et al., 2013a). Likewise, also for PPS examination, weapplied the same task previously employed after tool-use, namelythe audio–tactile interaction (Canzoneri et al., 2012; 2013a).

In this way, to test for a possible dissociation between PPS andBR, we recurred to the same tasks previously used to show similardynamic effects on these spatial and bodily representations(Canzoneri et al., 2013a).

A between-subjects design was chosen to avoid the effects ofcarry-over and fatigue after 10 h of immobilization. Three experi-ments were performed in different groups of subjects to measurethe effects of non-use and overuse on PPS and BR representationsfor both arms. Specifically, we evaluated BR in Exp. 1 (Group 1),using the tactile distance perception task (Canzoneri et al., 2013a)and in Exp. 2 (Group 2) through the body-landmarks localizationtask (Canzoneri et al., 2013a). Moreover, in Exp. 2 a visual re-presentation of the whole body was also assessed using theDaurat-Hmeljiak test (Daurat-Hmeljiak et al., 1978). Finally, in Exp.3 (Group 3), we examined PPS by using the audio–tactile inter-action task (Canzoneri et al., 2012, 2013a).

2.4. Task description

2.4.1. Body representation (Experiments 1 and 2)In Exp. 1, to assess the representation of the metric properties

of the arm after non-use/overuse, we adopted the tactile distanceperception task (Canzoneri et al., 2013a). Two pairs of tactile sti-muli were administered, one on the forehead (as a reference bodypart) and the other on the forearm (the target body part). To setthe spatial distance between stimuli, we initially measured thetwo-point discrimination threshold (2pdt) on the forearm. The2pdt was defined as the shorter distance between two touches atwhich subjects clearly detect two different stimuli. The 2pdt wasdetermined for each arm in the two testing days (before and afterimmobilization) using a Staircase method before the beginning ofthe experiment. Blindfolded subjects were tactilely stimulatedwith tappers (diameter 5 mm) mounted on a calliper while theywere lying down with the arms resting in a prone position. Eitherdouble or single posts were randomly administered. Only doubleposts were used to compute the staircase. The starting double postseparation was 40 mm, clearly above the 2pdt. The separation wasprogressively reduced by 50% after each set of three successivecorrect responses. In case of errors by the subjects, the separationwas subsequently increased to midpoint of the current (erroneous)trial and the immediately preceding (correct) trial. This procedurewas terminated at the shortest separation at which two distinctposts were clearly perceived. We subsequently confirmed this2pdt determination applying five double posts at this separationrandomly intermixed with five single posts. If the subjects scored

per limb non-use and overuse on space and body representations.a.2014.11.028i

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

at least 7/10, then that threshold estimate was accepted for ex-perimental testing. Otherwise, the procedure was repeated. Theindividual 2pdt was chosen to set the distance between the pairsof posts used during the tactile distance perception task for eachsubject in every testing day. Three different interpoint distanceswere checked: 2pdt, 1.5�2pdt, and 2�2pdt. In each trial of thetactile distance perception task, subjects were touched with a pairof posts on the forehead, immediately followed by touching with apair of posts on the forearm. Subjects made un-timed two-alter-native forced-choice judgments of whether the two posts feltfarther apart on the forehead or on the forearm, respondingverbally. The task consisted of a total of 36 consecutive trials: for12 trials the interpoint distance for the couple of dots on theforehead and on the forearm was the same, for 12 trials the in-terpoint distance was longer for the couples of dots on the forearmand vise versa for the remaining 12 trials. An experimentermanually administered stimuli for approximately one second, withan inter-stimulus interval of one second between taps on theforehead and forearm. The probability of reporting the distancebetween the two stimuli as longer on the forearm than on theforehead (P-forearm) was adopted as an indirect measure of theperceived arm length (Canzoneri et al., 2013a). The rationale forusing this approach is that the perceived size of the tactile stimulidelivered to the body depends on the perceived dimension of thestimulated body part (De Vignemont, 2010; Longo and Haggard,2011; Medina and Coslett, 2010; Spitoni et al., 2010). Based on theresults of our previous study (Canzoneri et al., 2013a), if the dis-tance between the stimuli is overestimated (increased P-Forearm)after immobilization (i.e. POST), then the arm is perceived asshorter than before (baseline, PRE); conversely, if the distancebetween the stimuli is underestimated (decreased P-Forearm),then the arm is perceived as longer at the POST than at the PREcondition.

In Exp. 2, to assess the perceived position of the arm after non-use/overuse, we adopted body-landmarks localization task, as re-cently described by our group (Canzoneri et al., 2013a). The sub-jects were asked to verbally indicate when a moving marker, re-corded using a motion capture system, matched the felt position oftwo occluded body parts: the middle finger (specifically, the nail)and the tip of the elbow joint (namely the olecranon). Before thetask, the experimenter indicated these anatomical landmarks onher own body. Participants were blindfolded, and then the ex-perimenter passively placed their arm on a table in a prone posi-tion. The forearm was aligned with the shoulder joint and fixed onthe table with tape to avoid arm movement for the whole taskduration. To prevent participants from viewing their arm duringthe task, a rectangular black box (90 cm long�50 cmwide�20 cm high) was positioned over the limb. Immediatelyafter box placement, the eye-patch was removed. During eachtrial, the experimenter verbally cued the participant as to whichlandmark to judge. The experimenter manually moved a retro-reflective marker over the surface of the box, along the long-itudinal axis of the forearm. The retro-reflective marker (1.5 cm indiameter) was stuck on the tip of a 50 cm long black cane. Indifferent randomized trials, the marker was moved in two differ-ent directions, either approaching (from distal-to-proximal direc-tion) or receding (from proximal-to-distal direction) from thesubject's body. Participants were instructed to give a verbal signalof “Stop” when the retro-reflective marker was perceived justabove the felt position of the target anatomical landmark. At thatsignal, the experimenter stopped the movement, leaving themarker where indicated. The subjects were allowed to furtheradjust the final position of the marker, asking the experimenter tomove it backward or forward, and following definitive confirma-tion from each of them, the marker's location was then recordedthrough an optical motion capture system (Vicon). After the last

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

trial, to register the actual positions of anatomical landmarks, thebox was removed, and the participants were again blindfolded andtwo retro-reflective markers (1 cm in diameter) were placed onthe nail of the middle finger and on the elbow. The task comprised20 randomized trials: 10 trials for each body landmark, with anequal number of repetitions in which the marker was moved inthe proximal-to distal or distal-to-proximal direction. The meanestimated distance between the middle finger and the elbow wasconsidered as an indirect measure of the perceived arm length. Inaddition, we calculated the position error (namely the differencebetween the actual position and the mean estimated location) andits variability (expressed as standard deviations, S.D.) among thetrials. Data analysis was performed using custom MATLAB(Mathworks, Natick, MA) software.

Additionally in Exp. 2, a visual image of the whole body wasassessed using the Daurat-Hmeljiak test (Daurat-Hmeljiak et al.,1978). A series of tiles was presented sequentially to each subject,representing a human body part. Participants had to place thesetiles on an empty background to reconstruct a whole human fig-ure. The tiles included the neck, the arms, the hands, the legs, thetrunks of the left and right side for the frontal view and the trunk,the arm, the hand and the leg for the two profiles. Three differentviews were adopted: frontal, left and right profiles. For each view,a test table showing only the position of the head was provided toeach subject. Every trial, the participant received a single tile toposition on the test table to reconstruct an entire body re-presentation. The experimenter traced the chosen position on ananswer sheet and removed the used tile before presenting the newone. On both testing days, the task was performed using the lefthand. Following immobilization, the participants' right arm re-mained bandaged for the whole duration of the test. We con-sidered the mean distances between the head, fixed on the sheet,and each body part. Custom MATLAB software (Mathworks, Na-tick, MA) was used to calculate these distances.

2.4.2. Peripersonal space representationIn Exp. 3 to evaluate the PPS representation we employed an

audio–tactile interaction task, following a similar experimentalprocedure described in our previous studies (Canzoneri et al.,2013a, 2012). Blindfolded subjects sat down with their arm restingin a prone position on a table. The same arm position adoptedbefore immobilization was also used for the POST testing im-mediately after the bandage's removal. Participants were asked toverbally respond as fast as possible to a tactile stimulus deliveredon the forearm (using a constant-current electrical stimulatorDS7A, Digitimer, Hertfordshire, United Kingdom, via a pair ofneurological electrodes, Neuroline, Ambu, Ballerup, Denmark). Thetactile reaction times (RTs) were recorded using a voice-activatedrelay. The participants were required to ignore task-irrelevantconcurrent sounds approaching (IN sound) or receding (OUTsound) from their arms along a trajectory between their hand'sposition and 100 cm from their hands. To give the impression ofeither approaching or receding from the subject's body, the soundswere manipulated in intensity (see Canzoneri et al., 2012, 2013afor a detailed description of sounds manipulation). For each trial,the sound (lasting 3000 ms) was preceded and followed by1000 ms of silence. The critical manipulation was that the tactilestimulus was delivered at 5 different temporal delays from theonset of the trial (T1: 1300 ms; T2: 1800 ms; T3: 2500 ms; T4:3200 ms; and T5: 3700 ms). Given the symmetric shape of the twosounds (Canzoneri et al., 2012), there was a spatial correspondencebetween the perceived position of IN and OUT sound at T1-IN andT5-OUT (both corresponding to the farthest distance from thebody¼D1), T2-IN and T4-OUT (far distance¼D2), T3-IN and T3-OUT (intermediate distance¼D3), T4-IN and T2-OUT (close dis-tance¼D4), T5-IN and T1-OUT (the closest distance¼D5). We then

per limb non-use and overuse on space and body representations.a.2014.11.028i

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

analyzed tactile RTs as a function of the five possible perceiveddistances, from D1, the farthest distance, to D5, the closest dis-tance, both for IN and for OUT sound. Moreover, catch trials withonly auditory stimulation were provided. To measure the RTs inthe unimodal tactile condition (without any sound) in some trials,tactile stimulation was delivered during the 1000 ms silent peri-ods, preceding or following sound, namely at 300 ms (D0) and at4600 ms (D6) from the beginning of the trial. For each temporaldelay, 12 randomized target trials were administered for the INand for the OUT sound, resulting in a total of 120 trials with atactile target randomly intermingled with 36 catch trials and 12unimodal tactile stimulations, equally divided into three blocks,lasting 10 min each. Sounds boost RTs to tactile stimuli whenperceived close to the stimulated body part, that is within theboundaries of PPS (Bassolino et al., 2010; Canzoneri et al., 2013a,2012; Serino et al., 2007, 2011), compared with sounds perceivedin the far space. Therefore, to measure the extension of PPS re-presentation around the immobilized and the overused arm, wecompared RTs to tactile stimuli associated with sounds perceivedat different distances (D1, D2, D3, D4, D5) from the body.

Moreover, to confirm that the sound source was actually per-ceived at different locations according to different temporal delaysfor IN and OUT sound (Canzoneri et al., 2012), several weeks afterimmobilization a control experiment was performed in a subgroupof 7 subjects who had undergone Exp. 3. Participants wereblindfolded and received tactile stimulation on their right forearmat the 5 different temporal delays (T1, T2, T3, T4 and T5) during INand OUT sound (112 trials). At the end of each trial, participantswere asked to verbally indicate the perceived position of thesound in space when they had felt the tactile stimulation, on ascale from 1 (very close) to 100 (very far).

2.5. Statistical analysis

To confirm the increased use of the left free limb during rightarm immobilization, we compared the duration (in minutes) of leftarm physical activity during non-use on the previous day using arepeated measures ANOVA (rmANOVA), with Time (day before,during immobilization) as the within-subjects factor and Group (1,2 and 3) as the between-subjects factor.

2.5.1. Body representationIn Exp. 1, to evaluate modifications on tactile acuity and the

perceived arm length, we compared the 2pdt and the P-forearm,before (PRE) and after (POST) immobilization, using separatepaired t-tests.

In Exp. 2, the perceived length of the constrained and overusedarms was analyzed using two separate rmANOVAs on the meandistance between the perceived position of the middle finger and theelbow, with time (PRE and POST) and the marker movement di-rection (proximal-to distal or distal-to-proximal) as within-sub-jects factors. Moreover, for both testing days, we compared theposition errors (i.e., the difference between the mean estimatedlocation of each target landmark and its actual position) to de-termine the contribution of each landmark on the perceived armlength and calculated the standard deviation of error positionamong trials to assess the response variability. Both parameterswere statistically analyzed by separate rmANOVAs, with time,marker movement direction and anatomical landmarks (elbowand middle finger) as within-subjects factors.

The mean distances between the head and each body parts ob-tained in the Daurat-Hmeljiak test before and after immobilizationwere compared using rmANOVAs, with time (PRE and POST) andview (profile or frontal view) as within-subjects factors.

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

2.5.2. Peripersonal space representationFinally, in Exp. 3 a rmANOVA on subjects’ sound-localization

judgments with sound movement direction (IN and OUT) anddistance (D1, D2, D3, D4, D5) as within-subjects factors was run toverify that the IN and OUT sound source was actually perceived atdifferent spatial positions (Canzoneri et al., 2012). In order to ac-curately evaluate the interaction between the positions of theperceived sounds and the tactile responses of the subjects, weperformed two separate rmANOVAs on the mean RT, with time(PRE and POST), distance (D1, D2, D3, D4, D5) and sound move-ment direction (IN and OUT) as within-subjects factors for the twoarms. In addition, the RTs of the unimodal tactile condition werecompared by Student's paired t tests to assess differences beforeand after non-use, for both arms.

All analyzes were performed using STATISTICA software (StatSoft, version 10, Padova, Italy). The significance threshold was setat Po0.05, and the interactions were explored using Newman–Keuls post-hoc comparisons.

3. Results

3.1. Preliminary note

As we had recently reported (Avanzino et al., 2011), here weconfirmed the duration of physical activity of the free arm to besignificantly higher during immobilization than the day prior toimmobilization (Time: F1,36¼68.60, Po0.00001) with no sig-nificant difference between the groups (Time�Group: F2,36¼0.05,P¼0.949). Thus, we evaluated two opposite effects: the non-use ofthe right arm and the concurrent overuse of the left limb. In linewith previous studies, the two different effects were evaluatedseparately for both sides (Avanzino et al., 2011; Lissek et al., 2009;Weibull et al., 2011).

Using this approach, we did not consider in our analyzesa-priori differences between the representation of the right andthe left hand, which goes beyond the aims of the present study.

3.2. Body representation (Experiments 1 and 2)

In the tactile distance perception task (Exp. 1), we observed thatP-forearm was similar in the two sessions (P¼0.841) when con-sidering the immobilized limb, whereas this parameter was sig-nificantly reduced with respect to the overused limb (Po0.05).The tactile acuity, based on the 2pdt, was significantly lower(Po0.01) on the constrained arm, but improved on the overusedarm (Po0.01) (Fig. 1).

In the body-landmarks localization task (Exp. 2), the perceivedarm length for the overused side increased in the POST-test session(Fig. 2, upper panel) (main effect of Time: F1,12¼5.57, Po0.05).This result reflected a significant displacement of the perceivedlocation of the middle finger toward its real position (i.e., reduc-tion of middle finger position error, Fig. 2, lower panel, and Fig. 3)(Po0.05), while no change in the elbow position error was found(P¼0.758; Time�Anatomical landmarks: F1,12¼5.05, Po0.05).Moreover, in the POST-test session, the variability of the positionerror among trials was significantly reduced (Time: F1,12¼5.08,Po0.05; Time�Anatomical landmarks: F1,12¼1.73, P¼0.213),suggesting improved judgment. Conversely, for the immobilizedarm, no significant changes in the perceived arm length (Time:F1,12¼1.02, P¼0.333), position error (Time: F1,12¼1.16, P¼0.303;Time�Anatomical landmarks: F1,12¼0.54, P¼0.476) or positionerror variability (Time: F1,12¼1.88, P¼0.195) were detected afternon-use. The marker movement direction did not influence suchparameters, as no significant main effect and no interaction werefound to be associated with this factor (P40.05 for all analyzes).

per limb non-use and overuse on space and body representations.a.2014.11.028i

Fig. 1. Effect of immobilization on the tactile distance perception task and on thetactile threshold (Experiment 1). The upper panel shows the mean percentage oftrials (7S.E.) in which subjects perceived the distance between the two tactilestimuli administered on the forearm as longer than the distance between the twotactile stimuli presented on the forehead before (PRE, light gray) and after (POST,dark gray) the immobilization procedure, for the immobilized and the overusedarm. The probability to perceive a longer distance on the forearm did not change inthe POST-test session on the immobilized side, but significantly decreased on theoverused side. The lower panel represents the mean (7S.E.) tactile threshold(2pdt, cm) measured before (PRE, light gray) and after (POST, dark gray) non-use onthe immobilized and on the overused arm. In the POST-test session, the tactilethreshold significantly increased on the immobilized arm and decreased on theoverused limb.

Fig. 2. Effect of immobilization on the perceived arm length (Experiment 2, body-landmarks localization task). The upper panel represents the mean (7S.E.) per-ceived length of the arm before (PRE, light gray) and after (POST, dark gray) theimmobilization procedure, for the immobilized and the overused arm. The lowerpanel shows the mean (7S.E.) position error of the middle finger and the elbow onthe immobilized and on the overused limb measured before (PRE, light gray) andafter (POST, dark gray) immobilization. The perceived length of the immobilizedarm did not change after non-use. In contrast, the overused arm was perceivedsignificantly longer in the post than in the pre-test session. This effect is due to asignificant reduction of the middle finger position error on the overused arm (seeFig. 3).

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Furthermore, in Exp. 2, no significant effects were observed on themean distances between the head and each body parts measuredin the frontal or profile view of the Daurat-Hmeljiak test, con-cerning both the left and the right side (P40.05 for all conditions).

3.3. Peripersonal space representation (Experiment 3)

The perceived position of the IN and OUT sound at differentdistances was confirmed using the sound-localization experiment.The subjects perceived sounds at different positions according todifferent temporal delays (Distance: F4,24¼56.53, Po0.00001),with no significant difference between IN and OUT sound at spa-tially corresponding temporal delays (Distance� Sound movementdirection: F4,24¼1.29, P¼0.301, mean response7S.E.:D1¼84.33 cm710.72; D2¼71.12 cm78.14; D3¼49.62 cm74.27;D4¼25.07 cm72.62; D5¼8.69 cm73.07).

As shown in Fig. 4, the distance at which sounds were able toaffect the tactile RT was closer to the body after immobilizationthan before, suggesting a contraction of PPS around the con-strained limb (Time�Distance: F4,48¼3.28, Po0.05; regardless ofSound movement direction, Time�Distance� Sound movementdirection: F4,48¼1.02, P¼0.404). Indeed Newman–Keuls post-hoccomparisons revealed that before immobilization the RTs at D1(the farther from the arm) were significantly slower than those atother distances (D5: Po0.001; D4: Po0.001; D3: Po0.001; D2:Po0.001), indicating that the boundary of PPS was located be-tween D1 and D2. After immobilization, the RTs at D1 and D2 werecomparable (P¼0.105), but significantly different from those ob-served at D3, D4 and D5 (D1 versus D5: Po0.01; D1 versus D4:Po0.05; D1 versus D3: Po0.001), suggesting that the boundary of

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

PPS was located between D2 and D3, which was closer to the limbof the participant than before immobilization. Moreover, followingimmobilization, the subjects were significantly slower in re-sponding to tactile stimuli associated with sounds delivered closeto their bodies (PRE versus POST: D5: Po0.05; D4: Po0.05; D3:Po0.01; D2: Po0.05), except for the farthest position (D1:P¼0.343), pointing out a reduction of audio–tactile interactionwithin PPS.

In contrast, we did not find modifications of the RTs in responseto tactile stimuli delivered on the overused arm before and afterimmobilization (Time�Distance: F4,48¼1.18, P¼0.333; Time-�Distance� Sound movement direction: F4,48¼0.26, P¼0.904). Amajor effect of distance was observed (D: F4,48¼48.54, Po0.0001),revealing that the RTs at D1 were significantly slower than those atother distances (D1 versus D2: Po0.01; D1 versus D3: Po0.001;D1 versus D4: Po0.001; D1 versus D5: Po0.001), and confirmingthat PPS boundaries are typically located between D1 and D2 forboth limbs. Moreover, we noticed that the RTs to unimodal tactilestimuli (D0 and D6) were not affected by immobilization for bothconstrained (P¼0.377) and overused (P¼0.464) arms.

4. Discussion

We immobilized the right arm for 10 h to manipulate “thepossibility to act in space” without artificially changing the bodymetrics. While previous studies have shown symmetric effectsbetween the two hemispheres on motor and somatosensory do-mains after the non-use of one limb and the consequent overuse ofthe free one (Lissek et al., 2009; Avanzino et al., 2011; Weibullet al., 2011), with the present results an asymmetry emerges. Wefound that PPS boundaries contracted around the non-used arm,but did not change around the overused limb (Exp. 3). In contrast,while the perceived arm length was modified (i.e. subjects located

per limb non-use and overuse on space and body representations.a.2014.11.028i

Fig. 3. Subject-by-subject perceived position of the middle finger and of the elbow (Experiment 2). The subject-by-subject (7standard deviation, S.D.) position error wasnormalized with respect to each participant’s real arm length for the middle finger (above) and the elbow (below), as illustrated. The average position error (7S.E.) amongsubjects, normalized with respect to the mean actual length of all participants' arms, is also shown (S̄). Light and dark gray dots, respectively, refer to the values measuredbefore (PRE) and after (POST) the immobilization procedure, for the immobilized (upper panel) and overused (lower panel) arm. The two horizontal dashed lines indicate themean actual position (100%) of the middle finger and the elbow. The minimum and maximum Y-axis values correspond to 60% and 120% of these normalized values,respectively. After immobilization, the middle finger position of the overused arm was perceived significantly farther from the body than before. The subjects' variability inlocalizing both the middle finger and the elbow of the overused side was significantly reduced after immobilization. Conversely, no modifications were observed on theimmobilized arm.

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎6

their middle finger more distally, closer to its actual position) afteroveruse (Exp. 1 and 2), the perception of the immobilized limbremained unchanged (Exp. 1 and 2). Taken together, these findingssuggest a dissociation between PPS and BR plastic proprieties. Anadditional analysis between-experiments further support this re-sult (see Supplementary Data).

4.1. Non-use, but not overuse, affects PPS representation

Although previous studies demonstrated that immobilizationaffects motor performance (Bassolino et al., 2012) and corticalexcitability (Avanzino et al., 2011; Bassolino et al., 2013; Huberet al., 2006), this works is, at the best of our knowledge, the firstreport demonstrating the plastic properties of PPS representationafter non-use. Before immobilization, the farthest distance atwhich sounds speed up tactile RTs was between D1 and D2 (cor-responding approximately to a sound position perceived between85 and 70 cm from the arm), and after non-use, this distanceshifted to a position between D2 and D3 (corresponding ap-proximately to a location between 70 and 50 cm), suggesting thatPPS boundary around the constrained arm contracted. Im-portantly, no modifications were observed on the RT to unimodaltactile stimulation (D0, D6) or tactile stimuli associated withsounds perceived in far space (i.e., D1), ruling out any general at-tentional or somatosensory effects. Thus, arm immobilization re-duces the ability to integrate multisensory stimuli over an arearoughly corresponding to the arm working space. In line with thisfinding, previous works reported alterations of space representa-tion when the action is constrained. An immediate contraction ofthe near space representation have been induced in a bisection

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

line task when participants wore wrist weights (Lourenco andLongo, 2009). Similarly, the perception of geographical slant isaltered by wearing heavy backpacks or by fatigue (Bhalla andProffitt, 1999). Recently, a shift of PPS boundaries towards thestump was described in upper-limb amputees without prosthesis(Canzoneri et al., 2013b). Notably, in these works we cannot ex-clude that these results might depend on concurrent changes inBR since the body status was affected (amputees) or experimen-tally manipulated (weights or fatigue). In amputees, for instance,the shift of PPS was associated with a concomitant shrinkage ofthe perceived arm length (Canzoneri et al., 2013b). Differently,here the observed contraction of PPS was not associated to chan-ges of body dimension, but it is rather due to the reduced possi-bility to act in space during non-use. Such interpretation is con-sistent with the seminal definition of PPS associating multisensoryprocessing of stimuli near the body with the representation ofpotential motor acts within that space (Gallese and Sinigaglia,2010; Graziano and Cooke, 2006; Rizzolatti et al., 1997). Along thisline, other authors showed a continuous remapping of PPS due toaction execution (Brozzoli et al., 2009, 2010). However, our datademonstrated that the possibility to act within that space is es-sential to maintain an efficient PPS representation.

During immobilization, the remaining free arm primarily in-teracts with the environment. In consideration of our previousresults on complementary effects of immobilization on the sen-sorimotor representation of the non-used and overused arms(Avanzino et al., 2011), one might hypothesize that PPS around theoverused arm would extend to compensate for the contractionaround the immobilized arm. However, overuse did not affect PPSdimension around the free arm. At a first sight, this null effect

per limb non-use and overuse on space and body representations.a.2014.11.028i

Fig. 4. Effect of immobilization on PPS representation: the audio–tactile interactiontask (Experiment 3). Mean RTs (7S.E.) to tactile stimuli with sounds perceived in5 different spatial locations (D1, D2, D3, D4, D5), with respect to the participant’sbody (see Supplemematry Data). Data are shown before (PRE, light gray dots) andafter (POST, dark gray dots) the immobilization procedure, for the immobilized(upper panel) and overused (lower panel) arm. The sigmoidal functions describingthe relationship between the RTs and sound distance before (light gray line) andafter (dark gray line) the immobilization procedure are shown. This result wasobtained after fitting individual data of RTs at each distance to a sigmoidal function(least squares regression) (Canzoneri et al., 2013a, 2012). For the immobilized limb,the critical point in space where sounds speeded up the tactile RTs were locatedbetween D1 and D2 before immobilization, and between D2 and D3 after im-mobilization, i.e., closer to the body, suggesting a contraction of PPS boundaries. Inaddition, after immobilization, RTs to tactile stimuli associated with sounds deliv-ered close to participants’ bodies (D2, D3, D4, D5) become longer than before. Nochange was found for the overused limb between the PRE- and the POST-testsessions.

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

might appear in contradiction with recent studies reporting anexpansion of PPS after intensive training with tools (Canzoneriet al., 2013a; Sposito et al., 2012). Nevertheless, tools physicallyprolong the arm length allowing subjects to reach objects outsidethe normal reachable space, whereas here the action-space re-mained limited around the physical structure of the overused arm.

4.2. Overuse, but not non-use, modulates the perceived arm length(i.e. BR)

BR was resilient to action deprivation, as the perceived armlength remained unchanged after 10 h of immobilization. It isfeasible that the visual information related to the arm metric couldplay a major role in determining this null effect. In fact, we notedthat the bandage did not alter the visual perception of the armlength. The importance of vision in calibrating body metrics is inline with former evidence demonstrating that visual informationhas a crucial role in body illusions, e.g. in attenuating illusorymovement induced by tendon vibration (Lackner and Taublieb,1984) or in altering perceived tactile distance (Taylor-Clarke et al.,2004). In addition, during non-use, proprioceptive (static pro-prioceptive information) inputs may also contribute to this “con-servative” perception of the body length.

The robustness of BR is consistent with that found in previousclinical and experimental observations, showing that although BR

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

extension is recurrently reported, evidence of contraction is rare(Robinson and Podoll, 2000; Todd, 1955). Moreover, in healthyindividuals, vibratory (De Vignemont et al., 2005) or visuotactile(Pavani and Zampini, 2007) sensory illusions lead to a perceptionof body parts as bigger, rather than smaller than their actual di-mensions, suggesting an asymmetric tendency to recognize en-larged but not reduced body parts. An asymmetric plasticity of BRmay also reflect daily life activities where we frequently manip-ulate objects that elongate our body parts, while the reverse effectis uncommon.

Conversely, the perceived dimension of the overused armchanged. The data from the tactile distance perception task (Exp.1) proved that the subjects perceived their overused limb as longerthan at the PRE-test. Notably, such increased perceived length ofthe overused arm can be more correctly interpreted as bias re-duction (Exp. 2). In fact, at baseline the middle finger position wasunderestimated, since it was perceived nearer the trunk. Thisunderestimation of finger length is in agreement with previousreports (e.g. Longo and Haggard, 2010). After overuse, this bias wasreduced: the middle finger was more distally located, closer to itsactual position. Such bias reduction can be viewed as an im-provement in estimating the body part localization. This inter-pretation is supported by the decreased position error variabilityamong trials after overuse (Figs. 2 and 3). Particularly, we did notobserve a similar reduction of the position error or responsevariability on the immobilized limb. We propose that body metricestimation improves with repetitive sensorimotor activity, therebyenhancing the quantity of sensory feedback and consequently thecalibration of the effector dimensions.

We cannot determine if this effect on the overused limb is aconsequence of immobilization of the other arm or a direct resultof overuse, since we did not test the effect of a long intense use ofone limb without immobilization. However, given that no differ-ences were found on the immobilized arm, it is unlikely thatmodifications of the overused limb compensate for the unchangedBR on the opposite side.

Changes in the perceived dimension of the arms are in-dependent of low-level somatic sensations (tactile discriminationthreshold) or explicit whole body representations (as assessed byDaurat-Hmeljiak test). Indeed, regardless of variations in bodymetrics, the tactile discrimination threshold (2pdt) significantlyincreased on the constrained arm and decreased on the overusedarm (Exp. 1). Similar modifications in tactile acuity have beenpreviously reported after immobilization (Lissek et al., 2009;Weibull et al., 2011). Moreover, no changes in the explicit wholebody representation were found, as supported by the lack of sig-nificant effects in the Daurat-Hmeljiak test.

4.3. The plasticity of body and space representations can dissociate

Differently from our previous study (Canzoneri et al., 2013a), inwhich similar modifications have been described for PPS and BRafter tool-use, the present work highlights a dissociation in theplastic properties of PPS and BR. First, these findings indicate thatPPS is modified, even when the perceived body dimension is notaltered and vise versa. Second, while PPS is capable of shrinking(like following immobilization), a contraction of BR is unlikely.

Accordingly with these behavioral dissociations, distinct areasprimarily involved in the awareness of the near space (ventralpremotor cortex and middle frontal gyrus) and of the body surface(supramarginal gyrus, post-central gyrus and, particularly, thewhite matter medial to these regions) were recently identified inpatients with neglect syndrome (Committeri et al., 2007).

per limb non-use and overuse on space and body representations.a.2014.11.028i

M. Bassolino et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎8

4.4. Conclusions

In conclusion, the present study reveals that the plasticity in PPSand metric features of BR rely on different mechanisms. The im-mobilization effects on PPS might depend either on a decrease ofmotor activity or on a reduction of the space inwhich movements canbe performed. However, the null effect of overuse on PPS seems tosupport the latter hypothesis: PPS representation depends on thedimension of the space inwhich the body potentially acts. In contrast,the finding that limb overuse, but not immobilization, affected thearm perceived length suggests that the plasticity of BR is not directlydependent on modifications of PPS representation. Rather BR seemsforged on a complex interplay between visual and sensorimotor in-formation related to the body, that contributes to the maintenance(non-use) and the update (overuse) of the perceived effector di-mension. All together these findings provide a new insight on the roleof action in shaping body and space representations. Besides thetheoretical contribution to current knowledge on this topic, this studyopens the way to develop and implement new clinical interventionsfor patients with motor disabilities and limb disuse.

Acknowledgments

The authors thank Valentina Pippo for her help in data collec-tion and Marco Jacono for the technical support.

Appendix A. Supplementary information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.neuropsychologia.2014.11.028.

References

Ainsworth, B.E., Haskell, W.L., Whitt, M.C., Irwin, M.L., Swartz, A.M., Strath, S.J.,Leon, A.S., 2000. Compendium of physical activities: an update of activity codesand MET intensities. Med. Sci. Sports Exerc. 32 (9 Suppl.), S498–S504.

Avanzino, L., Bassolino, M., Pozzo, T., Bove, M., 2011. Use-dependent hemisphericbalance. J. Neurosci. 31 (9), 3423–3428. http://dx.doi.org/10.1523/JNEUR-OSCI.4893-10.2011.

Bassolino, M., Bove, M., Jacono, M., Fadiga, L., Pozzo, T., 2012. Functional effect ofshort-term immobilization: kinematic changes and recovery on reaching-to-grasp. Neuroscience 215, 127–134. http://dx.doi.org/10.1016/j.neuroscience.2012.04.019.

Bassolino, M., Campanella, M., Bove, M., Pozzo, T., Fadiga, L., 2013. Training theMotor Cortex by Observing the Actions of Others During Immobilization.doi: 10.1093/cercor/bht190.

Bassolino, M., Serino, A., Ubaldi, S., Làdavas, E., 2010. Everyday use of the computermouse extends peripersonal space representation. Neuropsychologia 48 (3),803–811. http://dx.doi.org/10.1016/j.neuropsychologia.2009.11.009.

Bhalla, M., Proffitt, D.R., 1999. Visual-motor recalibration in geographical slantperception. J. Exp. Psychol. 25 (4), 1076–1096.

Brozzoli, C., Pavani, F., Urquizar, C., Cardinali, L., Farnè, A., 2009. Grasping actionsremap peripersonal space. Neuroreport 20 (10), 913–917. http://dx.doi.org/10.1097/WNR.0b013e32832c0b9b.

Brozzoli, C., Cardinali, L., Pavani, F., Farnè, A., 2010. Action-specific remapping ofperipersonal space. Neuropsychologia 48 (3), 796–802. http://dx.doi.org/10.1016/j.neuropsychologia.2009.10.009.

Canzoneri, E., Magosso, E., Serino, A., 2012. Dynamic sounds capture the boundariesof peripersonal space representation in humans. PLoS One 7 (9), e44306. http://dx.doi.org/10.1371/journal.pone.0044306.

Canzoneri, E., Ubaldi, S., Rastelli, V., Finisguerra, A., Bassolino, M., Serino, A., 2013a.Tool-use reshapes the boundaries of body and peripersonal space representa-tions. Exp. Brain Res. (228), 25–42. http://dx.doi.org/10.1007/s00221-013-3532-2.

Canzoneri, E., Marzolla, M., Amoresano, A., Verni, G., Serino, A., 2013b. Amputationand prosthesis implantation shape body and peripersonal space representa-tions. Sci. Rep. 3, http://dx.doi.org/10.1038/srep02844.

Cardinali, L., Brozzoli, C., Farnè, A., 2009a. Peripersonal space and body schema: twolabels for the same concept? Brain Topogr. 21 (3–4), 252–260. http://dx.doi.org/10.1007/s10548-009-0092-7.

Cardinali, L., Frassinetti, F., Brozzoli, C., Urquizar, C., Roy, A.C., Farnè, A., 2009b. Tool-

Please cite this article as: Bassolino, M., et al., Dissociating effect of upNeuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologi

use induces morphological updating of the body schema. Curr. Biol. 19 (12),R478–R479. http://dx.doi.org/10.1016/j.cub.2009.05.009.

Committeri, G., Pitzalis, S., Galati, G., Patria, F., Pelle, G., Sabatini, U., Pizzamiglio, L.,2007. Neural bases of personal and extrapersonal neglect in humans. Brain 130(Pt 2), 431–441. http://dx.doi.org/10.1093/brain/awl265.

Daurat-Hmeljiak, C., Stambak, M., Berges, J., 1978. Il test dello schema corporeo. Unaprova di conoscenza e costruzione dell’immagine del corpo. OrganizzazioniSpeciali, Firenze.

De Vignemont, F., 2010. Body schema and body image–pros and cons. Neu-ropsychologia 48 (3), 669–680. http://dx.doi.org/10.1016/j.neuropsychologia.2009.09.022.

De Vignemont, F., Ehrsson, H.H., Haggard, P., 2005. Bodily illusions modulate tactileperception. Curr. Biol. 15 (14), 1286–1290. http://dx.doi.org/10.1016/j.cub.2005.06.067.

Farnè, A., Làdavas, E., 2000. Dynamic size-change of hand peripersonal space fol-lowing tool use. Neuroreport 11 (8), 1645–1649.

Gallese, V., Sinigaglia, C., 2010. The bodily self as power for action. Neuropsycho-logia 48 (3), 746–755. http://dx.doi.org/10.1016/j.neuropsychologia.2009.09.038.

Graziano, M.S.A., Cooke, D.F., 2006. Parieto-frontal interactions, personal space, anddefensive behavior. Neuropsychologia 44 (6), 845–859. http://dx.doi.org/10.1016/j.neuropsychologia.2005.09.009.

Holmes, N.P., Sanabria, D., Calvert, G.A., Spence, C., 2007. Tool-use: capturingmultisensory spatial attention or extending multisensory peripersonal space?Cortex 43 (3), 469–489.

Huber, R., Ghilardi, M.F., Massimini, M., Ferrarelli, F., Riedner, B.A., Peterson, M.J.,Tononi, G., 2006. Arm immobilization causes cortical plastic changes and locallydecreases sleep slow wave activity. Nat. Neurosci. 9 (9), 1169–1176. http://dx.doi.org/10.1038/nn1758.

Iriki, A., Tanaka, M., Iwamura, Y., 1996. Coding of modified body schema during tooluse by macaque postcentral neurones. Neuroreport 7 (14), 2325–2330.

Kammers, M.P.M., Mulder, J., Vignemont, F., Dijkerman, H.C., 2009. The weight ofrepresenting the body: addressing the potentially indefinite number of bodyrepresentations in healthy individuals. Exp. Brain Res. 204 (3), 333–342. http://dx.doi.org/10.1007/s00221-009-2009-9.

Lackner, J.R., Taublieb, A.B., 1984. Influence of vision on vibration-induced illusionsof limb movement. Exp. Neurol. 85 (1), 97–106.

Lissek, S., Wilimzig, C., Stude, P., Pleger, B., Kalisch, T., Maier, C., Peters, S.A., Nicolas,V., Tegenthoff, M., Dinse, H.R., 2009. Immobilization impairs tactile perceptionand shrinks somatosensory cortical maps. Curr Biol 19 (10), 837–842. http://dx.doi.org/10.1016/j.cub.2009.03.065.

Longo, M.R., Haggard, P., 2010. An implicit body representation underlying humanposition sense. Proc. Natl. Acad. Sci. USA 107 (26), 11727–11732. http://dx.doi.org/10.1073/pnas.1003483107.

Longo, M.R., Haggard, P., 2011. Weber's illusion and body shape: anisotropy oftactile size perception on the hand. J. Exp. Psychol. 37 (3), 720–726.

Lourenco, S.F., Longo, M.R., 2009. The plasticity of near space: evidence for con-traction. Cognition 112 (3), 451–456. http://dx.doi.org/10.1016/j.cognition.2009.05.011.

Maravita, A., Iriki, A., 2004. Tools for the body (schema). Trends Cogn. Sci. 8 (2),79–86. http://dx.doi.org/10.1016/j.tics.2003.12.008.

Maravita, A., Spence, C., Kennett, S., Driver, J., 2002. Tool-use changes multimodalspatial interactions between vision and touch in normal humans. Cognition 83(2), B25–B34.

Medina, J., Coslett, H.B., 2010. From maps to form to space: touch and the bodyschema. Neuropsychologia 48 (3), 645–654. http://dx.doi.org/10.1016/j.neuropsychologia.2009.08.017.

Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburghinventory. Neuropsychologia 9 (1), 97–113.

Pavani, F., Zampini, M., 2007. The role of hand size in the fake-hand illusionparadigm. Perception 36 (10), 1547–1554.

Rizzolatti, G., Fadiga, L., Fogassi, L., Gallese, V., 1997. The space around us. Science277 (5323), 190–191.

Robinson, D., Podoll, K., 2000. Macrosomatognosia and microsomatognosia in mi-graine art. Acta Neurol. Scand. 101 (6), 413–416.

Serino, A., Bassolino, M., Farnè, A., Làdavas, E., 2007. Extended multisensory spacein blind cane users. Psychol. Sci. 18 (7), 642–648. http://dx.doi.org/10.1111/j.1467-9280.2007.01952.x.

Serino, A., Canzoneri, E., Avenanti, A., 2011. Fronto-parietal areas necessary for amultisensory representation of peripersonal space in humans: an rTMS study. J.Cogn. Neurosci. 23 (10), 2956–2967. http://dx.doi.org/10.1162/jocn_a_00006.

Serino, A., Haggard, P., 2010. Touch and the body. Neurosci. Biobehav. Rev. 34 (2),224–236. http://dx.doi.org/10.1016/j.neubiorev.2009.04.004.

Spitoni, G.F., Galati, G., Antonucci, G., Haggard, P., Pizzamiglio, L., 2010. Two forms oftouch perception in the human brain. Exp. Brain Res. 207 (3–4), 185–195. http://dx.doi.org/10.1007/s00221-010-2446-5.

Sposito, A., Bolognini, N., Vallar, G., Maravita, A., 2012. Extension of perceived armlength following tool-use: clues to plasticity of body metrics. Neuropsychologia50 (9), 2187–2194. http://dx.doi.org/10.1016/j.neuropsychologia.2012.05.022.

Taylor-Clarke, M., Jacobsen, P., Haggard, P., 2004. Keeping the world a constant size:object constancy in human touch. Nat. Neurosci. 7 (3), 219–220. http://dx.doi.org/10.1038/nn1199.

Todd, J., 1955. The syndrome of Alice inWonderland. Can. Med. Assoc. J. 73 (9), 701–704.Weibull, A., Flondell, M., Rosén, B., Björkman, A., 2011. Cerebral and clinical effects

of short-term hand immobilisation. Eur. J. Neurosci. 33 (4), 699–704. http://dx.doi.org/10.1111/j.1460-9568.2010.07551.x.

per limb non-use and overuse on space and body representations.a.2014.11.028i