American Journal of Primatology 73:281–290 (2011)

RESEARCH ARTICLE

Population-Level Right-Handedness for a Coordinated Bimanual Taskin Naturalistic Housed Chimpanzees: Replication and Extensionin 114 Animals From Zambia and Spain

MIQUEL LLORENTE1,2�, DAVID RIBA1,2, LAIA PALOU3, LARA CARRASCO4, MARINA MOSQUERA2,MONTSERRAT COLELL4, AND OLGA FELIU1

1Unitat de Recerca i Laboratori d’Etologia, Fundacio Mona, Riudellots de la Selva, Girona, Spain2Institut Catala de Paleoecologia Humana i Evolucio Social (IPHES), Universitat Rovira i Virgili (URV), Spain3Departament de Metodologia de les Ciencies del Comportament, Universitat de Barcelona, Barcelona, Spain4Departament de Psiquiatria i Psicobiologia Clınica, Universitat de Barcelona, Barcelona, Spain

Recently, many studies have been conducted on manual laterality in chimpanzees. Nevertheless,whether nonhuman primates exhibit population-level handedness remains a topic of considerabledebate. One of the behaviors studied has been bimanual coordinated actions. Although recent studieshave highlighted that captive chimpanzees show handedness at population level for these tasks, someauthors have questioned the validity and consistency of these results. The first reason has been thehumanization of the samples. The second one has been that the results refer to animals in Americanbiomedical centers and the studies were conducted by the same team [WD Hopkins et al.]. This articleaims to assess the laterality in bimanual coordination (tube task) activities in animals housed in anintermediate environment (Chimfunshi, Zambia). This has been conducted by replicating previousstudies on similar samples (Mona Foundation, Spain), and then by extending the results to chimpanzeeshoused in intermediate settings. Individuals were evaluated through four experimental sessions (tests).Results indicated that 86% of the Chimfunshi sample was lateralized (48% RH, 38% LH). Furthermore,the sample showed population-level right-handedness in the mean handedness index, in Test 1, Test 2,and the first half of the study (Test 112). Rearing experience did not have an influence on handpreference. Taken together, the two sample (intermediate settings: Chimfunshi and Mona) resultsindicate a clear right-handedness. In conclusion, this replication and extension shows that (1) the Monaand Chimfunshi chimpanzees are right-handed in certain conditions, (2) the results are consistent withthose obtained by Hopkins in captive settings, (3) the humanization of the samples does not affectmanual laterality, (4) females are right-handed at population-level, but not males, and (5) these resultsreinforce the fact that the complexity of the task plays a dominant role in the expression of handlaterality among chimpanzees. Am. J. Primatol. 73:281–290, 2011. r 2010 Wiley-Liss, Inc.

Key words: chimpanzee handedness; bimanual coordination; hand preferences; replication;intermediate environments

INTRODUCTION

The study of hand laterality in nonhumanprimates offers a major background to understandhuman evolution for brain hemispheric specializa-tion, asymmetries of the motor cortex, language, andhand dominance at population level [Bradshaw &Rogers, 1993].

Unlike humans, nonhuman primates have evi-denced to be a very heterogeneous biological orderregarding manual asymmetries, as studies during thelast 25 years have shown. Globally, these studies yieldwide variability concerning the methodology used,the tasks observed, and the environments in whichthe work was carried out. This has caused stronglimitations for making reliable comparisons both atinter- and intraspecific level [McGrew & Marchant,

1997]: at the individual level, hand preferences seemto be consistent in all taxa from prosimians to greatapes. At population level, evidence focuses on

Published online 15 October 2010 in Wiley Online Library (wileyonlinelibrary.com).

DOI 10.1002/ajp.20895

Received 16 April 2010; revised 15 September 2010; revisionaccepted 22 September 2010

Contract grant sponsors: Fundacion Atapuerca, Generalitat deCatalunya (Ajut de mobilitat d’estudiants de Master); SpanishMinisterio de Ciencia e Innovacion; Contract grant number:MICIIN-HAR2009-07223/HIST; Contract grant sponsor: Univer-sitat Rovira i Virgili; Contract grant number: URV-2009AIRE-05.

�Correspondence to: Miquel Llorente, Unitat de Recerca iLaboratori d’Etologia, Centre de Recuperacio de Primats,Fundacio Mona. Carretera de Cassa 1 km, 17457, Riudellots dela Selva, Girona, Spain. E-mail: mllorente@fundacionmona.org

rr 2010 Wiley-Liss, Inc.

complex tasks; that is, those requiring coordinatedbimanual and sequential actions, and also on someother behaviours required for bipedal posture andpostural readjustment, among others [Blois-Heulinet al., 2006; Braccini et al., 2010; Colell et al., 1995a,b;Hopkins, 1993; Hopkins & Pearson, 2000; Hopkins &Rabinowitz, 1997; Hopkins et al., 1993, 2003b;Vauclair et al., 2005; Wesley et al., 2002; Westergaard& Suomi, 1996; Zhao et al., 2010]. This heterogeneouspattern is even more stressed on comparing wild andcaptive chimpanzees. Explanations of this heteroge-neity are still in debate, and they represent one of themajor keys around which the subject of manuallaterality in these species turns.

Several researchers have suggested major fac-tors, such as the human environment. In this sense,some authors [McGrew & Marchant, 1997; Palmer,2003; Warren, 1980] claim that evidence of right-handedness at population level may be related to thehumanized or encultured environments where sub-jects were raised. However, other researchers, suchas Hopkins and Cantalupo [2005], support the use ofdifferent methodologies (experimental vs. observa-tional), and protocols of recording and analyzingamong the studies may also cause this disparity ofresults. Finally, other authors [Fagot & Vauclair,1991; Hopkins & Cantalupo, 2005; Schweitzer et al.,2007] support the importance of variables inherentto the task, such as the grip morphology, posture,bimanual coordination vs. unimanual tasks, etc.

Most of the evidence on manual laterality inchimpanzees comes from captive animals, whichdemonstrates clear right-hand asymmetry in simplemotor tasks (i.e. bipedal reaching) [Hopkins &Pearson, 2000], drinking water [Colell et al.,1995a], eating–feeding [Hopkins & Fernandez-Carriba,2000], hand-to-mouth [Bard et al., 1990; Hopkins &Bard, 1995], throwing [Hopkins et al., 1993], andcomplex tasks, such as ball task [Hopkins & Pearson,2000], dipping task bimanual [Hopkins & Rabinowitz,1997], fixed tube task [Hopkins et al., 2007; Wesleyet al., 2002], horizontal panel [Colell et al., 1995b],joystick manipulation [Hopkins et al., 1989], andtube task [Hopkins & Cantalupo, 2004; Hopkins &Pearson, 2000; Hopkins et al., 2001, 2004, 2005;Phillips & Hopkins, 2007; Wesley et al., 2002].Although there are fewer studies with wild chim-panzees, right-hand asymmetries have been alsofound for certain complex tools behaviors, such asnut cracking, wadge dipping, and ant dipping[Boesch, 1991; Humle & Matsuzawa, 2002; Matsuzawaet al., 2001]. However, till date, there is only oneexperimental study focused on intermediate samples(naturalistic housed chimpanzees [NCh]), which alsoevidenced right-handedness [Llorente et al., 2009].

The tube task is the task more used in experi-mental studies with nonwild animals. It wasproposed by Hopkins [1995], and it consists ofextracting the food from a tube through bimanual

coordinated actions, where one hand acquires thedominant role and the other subordinate. For thistask, most of the studies done over the years haveshown clear evidence of right asymmetries at thepopulation level [Hopkins & Bard, 2000; Hopkins &Cantalupo, 2003; Hopkins & Pearson, 2000; Hopkinset al., 2004; Wesley et al., 2002]. However, someauthors question the validity and consistency ofthese results [Annett, 2006; Crow, 2005; McGrew &Marchant, 1997; Palmer, 2002, 2003], particularlybecause of the possible humanization of the samplesand their origin, because all the evidence wasobtained from animals belonging to Americanresearch centers and also because it was obtainedby the same research team [Hopkins et al.] (Fig. 1).

Therefore, in our view, it is important toreplicate the same experiments with other types ofchimpanzees, those sheltered in naturalistic settings,in order to check the validity and consistency of theright-handedness in captive samples. Despite the factthat NCh are also held in captivity, they aresheltered in environments that try to emulate theconditions of chimpanzees in the wild and allowthe animals to develop species-specific behaviors.Therefore, studies with NCh increase the ecologicalvalidity of previous results.

This study aims to assess the laterality inbimanual coordination (tube task) in animals housedin an intermediate environment (Chimfunshi Sanc-tuary, Zambia), by replicating previous studies doneon similar samples (Mona Foundation [MF], Spain),in order to extend the results to chimpanzees housedin these settings.

METHODS

Project Structure

This study replicates the first experimental testof the tube task carried out at MF in 2007 [Llorenteet al., 2009]. This replica has later been performed atChimfunshi Wildlife Orphanage (CWO), followingthe same methodology. This research adhered to theAmerican Society of Primatologists principles for theethical treatment of primates.

Subjects and Housing

Chimfunshi wildlife orphanageOne hundred and twenty Chimpanzees (Pan

troglodytes) distributed in seven captive groups werestudied at the CWO. Their ages ranged between 0 andto more than 33 years old. Table I shows additionalinformation about age classes, sex, and the rearinghistory of each group. Most of the chimpanzees wereused for trading (smuggling, circuses, and shows)before being sheltered in this Center.

The groups were living in different enclosures,which were distributed between two main areas: theProject Area and the Orphanage, which were

Am. J. Primatol.

282 / Llorente et al.

physically separated from one another. In the ProjectArea, groups 1, 2, 3, and 4 were housed in largeoutdoors compounds (enclosures 1, 2, 3, and 4,respectively) with attached indoor quarters in whichthe animals were placed daily during feeding time(from 11:30 a.m. to 13:30 p.m.). These indoorshandling facilities consisted of different rooms ofsimilar sizes and layouts. Specifically, the average sizeof the rooms was 6� 4 m. Outdoor enclosures werecarved out of the forest and floodplains along theUpper Kafue River, with enough thick jungle and fruitgroves and open grasslands to allow the chimpanzeesto roam as if in the wild. Enclosure 1, 2, and 3measured 150 ha, whereas enclosure 4 measured80 ha. In all the studied groups, access to the outsideenclosure was blocked during data collection.

The Orphanage was home to the other threegroups of chimpanzees (groups 5, 6, and 7), whichcould not be relocated. In those cases, each grouplived in different conditions. Some of the individualsin group 5 remained all day long inside theindoor enclosures for security reasons and the samehappened with the enclosed chimpanzees in group 6.In contrast, group 7 spent the night and feeding timein the indoor facilities, but during the rest of the day,they enjoyed a 5 ha enclosure.

Mona Foundation SanctuaryThree captive groups of chimpanzees (P. troglo-

dytes) were studied at MF Primate Rescue Center(Girona, Spain) [Llorente et al., 2009; Mosquera et al.,

0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500

Llorente et al. (2009)-Mona (Tool)

Llorente et al. (2009)-Mona

Llorente et al. (2009)-Mona (Finger)

Hopkins & Cantalulo (2003)-Yerkes

*Llorente et al (this study)-Chimfunshi+Mona (Tool)

Hopkins, Wesley et al. (2004)-Yerkes

Hopkins, Cantalupo et al. (2005)-Yerkes

*Llorente et al (this study)-Chimfunshi+Mona

*Llorente et al (this study)-Chimfunshi+Mona (Finger)

Hopkins, Hook et al. (2003)-Bastrop

Hopkins, Wesley et al. (2004)-Bastrop

Hopkins & Cantalulo (2004)-Yerkes

Hopkins, Fernández-Carribat et al. (2001)-Yerkes

Hopkins & Pearson (2000)-Yerkes

Phillips & Hopkins (2007)-Yerkes

Hopkins, Wesley et al. (2004)-Alamogordo

*Llorente et al (this study)-Chimfunshi

*Llorente et al (this study)-Chimfunshi (Finger)

Wesley, Fernández-Carriba et al. (2002)-Yerkes

*Llorente et al (this study)-Chimfunshi (Tool)

Mean HI

Fig. 1. Mean HI scores for different independent samples of chimpanzees that have been tested on the tube task. Black bar: significantpopulation preferences. Grey bar: nonsignificant population preferences. �Data from this study.

TABLE I. Social Groups at Chimfunshi Wildlife Orphanage. Composition Expresses Both the Range of Ages inYears and the Sex: M: Male; F: Female

Composition (range of years old) Rearing experience

Group N (0–4) (5–7) (8–12) (11–14) (16–33) (15–33) Human Mother

1 23 5M11F 2M11F 1M 3F 5M 5F 8 152 46 7M16F 5M16F 1M11F 3F 3M113F 1F433 17 293 13 1M11F 1M 4M14F 1M11F 10 34 13 2M 1M 2M13F 2M12F 1M 11 25 8 8M 7 16 5 1F 1F 1M 1M11F 2 37 12 1M11F 4M12F 3F 1M 9 3

Am. J. Primatol.

Handedness in Naturalistic Housed Chimpanzees / 283

2007]. They were between 4 and 51 years old. Animalswere in their social groups during the tests sessions.The first group (A) was composed of four males; thesecond group (B) was composed of four males andthree females; and third group (C) was composed ofthree males. Table II shows individual information,characteristics, and their background before arrivingat MF. The subjects were housed in a naturalisticoutdoor enclosure of 5,640 m2, with wooden struc-tures and natural ground with Mediterranean andriverside vegetation. Furthermore, the Center has twosocialization enclosures of each measuring 25 m2,connected to a 140 m2 primate pavilion.

Procedure

We used the same procedure to evaluate handlaterality in complex bimanual tasks as that appliedin the previous study in MF [Llorente et al., 2009].Evaluation was based on the tube task, proposed byHopkins [1995], as a bimanual coordinated tasksensitive to determine hand motor bias. For the tubetask, we used cylindrical rubber hoses of 25 to25.5 cm in length and with a diameter 2.5 cm, insteadof rigid tubes. Therefore, we call this experiment‘‘hose task.’’ In MF, author (M.L.) carried out all theobservations and the experiments. This study wasconducted from January 2007 to April 2007. In CWO,LP and LC carried out all the observations andexperiments. That study was conducted betweenOctober and November 2008.

The usefulness of the tube task as a manuallaterality indicator has been proven by Hopkins inseveral studies with captive chimpanzees [Hopkins,2006; Hopkins et al., 2004], and it has also been usedwith other primate species, such as Cebus apella[Lilak & Phillips, 2008], Cebus capucinus [Meunier &Vauclair, 2007] Cercopithecus neglectus, [Schweitzeret al., 2007], Papio anubis [Vauclair et al., 2005], Pan

paniscus [Chapelain & Hogervorst, 2009], and otherape species [Hopkins et al., 2003a]. Hoses were filledwith food, thus preventing its extraction with thetongue or by hitting it. In MF, the food was honey,peanuts, muesli, and seeds. In CWO, the food waspeanut butter, honey, peanuts, banana, apple, chow,and nshima (cornmeal product, a staple food inZambia). Animals had to remove the food with theirfingers or by using tools, such as sticks, branches, orcanes to facilitate extraction. We used a focal animalsampling [Altmann, 1974]; the session continueduntil the individual lost interest or left the hose forone 1 min. Each individual was tested four times andtest sessions were separated by a minimum of 2 daysbetween sessions. Each animal needed to obtain aminimum of 50 responses. Each test was valid if theanimal obtained a minimum of six responses. Thehand used to extract the food was recorded eachtime the subject inserted its finger/tool, removed thefood, and placed its finger/tool in its mouth. We alsorecorded the finger used to remove the food (thumb,index, middle, ring, and little).

Procedural considerationsSome authors have criticized recording beha-

viors as events because it may cause problems in thedata independence [McGrew & Marchant, 1997].According to this criticism, recording repetitivemotor actions of the subjects could increase thesample size and would thus increase the probabilityof detecting significant asymmetries in the hand useat individual level. To date, some authors have foundeffects of the methodology used [Chapelain &Hogervorst, 2009]. However, other authors havefailed to detect significant differences in recordingmanual behaviors as events or bouts, both beingsensitive to detect lateral biases [Hopkins, 2006;Hopkins et al., 2004; Meguerditchian et al., 2010].We have chosen to record the data as events, that is,

TABLE II. Individual Information, Characteristics, and Background of Mona Chimpanzee Sample

Name Sex Birth date Age class Birth place

Year arrivedat Mona

Foundation Background

Previoushumancontact Group

Toto Male 1956 Adult Wild, Nigeria 2003 Pet, Zoo Half BRomie Female 1979 Adult Wild 2001 Reproduction, Circus Very high BJulio Male 1979 Adult Wild? 2005 Zoo Half CVıctor Male 1982 Adult Wild, Cote d’Ivoire 2006 Pet Very high CToni Male 1983 Adult Wild, Guinea Bissau 2001 Zoo, Advertising Very high AMarco Male June 04, 1984 Adult Captivity 2001 Advertising, Circus Very high ATico Male 1985 Adult Captivity? 2005 Pet, Zoo Half CCharly Male July 02, 1989 Adult Captivity 2001 Advertising, Circus Very high APancho Male May 27, 1990 Adult Captivity 2001 Advertising, Circus Very high AWaty Female 1996 Adolescent Captivity 2002 Circus Very high BSara Female 1998 Adolescent Captivity 2004 Pet, TV Very high BBongo Male 2000 Juvenile Captivity 2002 Circus Very high BNico Male 2001 Juvenile Captivity 2004 Pet Very high BJuanito Male 2003 Juvenile Captivity 2003 Pet Very high B

Am. J. Primatol.

284 / Llorente et al.

considering the total frequency of manual use. Inthis way, we can replicate and compare our presentresults and our previous results [Llorente et al.,2009] with most other experimental studies as theyfollow this same procedure.

Data analysisWe used a similar methodology to that carried

out by other authors for similar experiments[Hopkins et al., 2004]. Data analyses were performedat the individual and group level. At the individuallevel, we first calculated binomial tests on data foreach individual to evaluate if they were significantlylateralized. Second, a handedness index (HI) sug-gested by Hopkins [1995] was calculated for eachsubject to quantify individual laterality bias. Thismeasure was estimated using the formula R�L/R1L(R 5 number of times right hand was used; L 5 numberof times left hand was used). Positive values show aright-hand bias and negative values show a left-handbias. For the hose task, we calculated an individualHI, adding the total frequencies obtained throughoutfour tests sessions mixed in HOSE (SUM-HI),following the same formula for HI. We used absolutevalues of HI (ABS-HI) as a reflection of the strengthof hand preference, independent of the direction. Wecalculated a mean HI averaging the HI values (MHI)of four hose task tests. To evaluate population levelhandedness, we used a one sample t-test based onindividual HI values and Z-scores. We used nonpara-metric tests (w2) to estimate the proportion ofright, left, or nonpreferent individuals. We used theSpearman correlation test to evaluate the consis-tency of hand preference throughout test sessions inthe hose task. Wilcoxon’s and Friedman’s ANOVAtests assessed differences in the direction andstrength of hand preferences among tests. We usedMann–Whitney U and Kruskal–Wallis to test samplevariables (sex, place of birth, group, or age class).

All the field protocols, data collection proce-dures, and data analyses were conducted in accor-dance with all the principles of ethical treatmentestablished by ASAB, the Spanish and Cataloniangovernments, the Zambian government, and theinternal rules and guidelines of the MF and CWO.

RESULTS

Replication: Chimfunshi Wildlife Orphanage

Of the 120 individuals in the sample, 100obtained the minimum of responses required(n 5 50). Globally, there were 14,854 manual actions,of which 55.48% (n 5 8,241) were performed by theright hand and 44.52% (n 5 6,613) by the left hand.Based on binomial tests, 14 individuals were non-preferent, and 86 are lateralized for this task: 48 areright-handed and 38 left-handed. The average forSUMHI (M-SUMHI) is 0.107 (SE 5 0.064); therefore,

subjects expressed a right-hand bias. The MHI valueis 0.122 (SE 5 0.063). SUMHI and MHI valuescorrelate highly (Rs 5 0.991; P 5 0.000), thus showingthat both kinds of handedness characterization wereconsistent among subjects. Hence, we used SUMHIvalues to minimize Type I error. The average forABSHI for all subjects was 0.573 (SE 5 0.030).The ABSHI values among right- (0.676) and left-handed (0.594) subjects were similar (U 5 714.500;P 5 0.086). There were differences between thenumber of lateralized and nonlateralized subjects(w2(1, n 5 100) 5 51.840, P 5 0.000), but there were nodifferences between the number of right- and left-handed subjects (w2(1, n 5 86) 5 1.163, P 5 0.281).

Regarding the variables associated with thesample, SUMHI showed differences for sex, becausefemales were more right-handed than males(U 5 955.000; P 5 0.047). However, no influence ofrearing experience (U 5 1,198.000; P 5 0.889), group(Kruskal 5 2.582; P 5 0.859), or age class (Kruskal 5

0.162; P 5 0.922) were found in the direction oflateralization (SUMHI). Also, we did not finddifferences in the strength of lateralization in sex(U 5 1,194.500; P 5 0.743), rearing (U 5 1,114.500;P 5 0.470), group (Kruskal 5 2.014; P 5 0.918), or ageclass (Kruskal 5 2.425; P 5 0.298).

According to the extraction technique, the sub-jects removed the food 95.66% using their fingers and4.34% using tools. They performed most of the actionswith the index finger (D2; 89.02%), followed by themiddle finger (D3; 3.77%), thumb (D1; 1.50%), littlefinger (D5; 1.31%), and ring finger (D4; 0.07%). Wefound an effect of extraction technique on the hand(D1; D2; D3; D4; D5; TOOL; w2 5 217.022, P 5 0.000).On one hand, the analyses of adjusted residualsshowed certain correlations between subjects per-forming extractions with the index finger (AR 5 9.3)and preferentially with the right hand. On the otherhand, extractions with the little finger (AR 5 13.2) andtools (AR 5 5.7) were correlated with the left hand.There was no difference between digital and toolextraction techniques in direction (HI Finger 5 0.103;HI Tool 5 0.066; Z 5�0.048, P 5 0.654) or strength(ABSHI Finger 5 0.577; ABSHI Tool 5 0.588; Z 5

�0.131, P 5 0.896) of hand preference.Four tests showed a high correlation, indicating

that preferences were stable and consistent through-out sessions at the individual level (Table III). Thecorrelation between the values of HI for the first halfof the experiment (Test 112) and the second half ofthe experiment (Test 314) was also significant(Rs 5 0.770; P 5 0.000). There is a trend for the HIvalues (direction) to decrease with the experimentaltests (Table IV), but these differences are notsignificant (Friedman’s ANOVA 5 4.887; P 5 0.180).Strength (ABSHI) tends to increase from Test 1 to 3,decreasing the value at Test 4. However, againthe differences are not significant (Friedman’sANOVA 5 6.486; P 5 0.090).

Am. J. Primatol.

Handedness in Naturalistic Housed Chimpanzees / 285

Regarding hand preferences at population level,one sample t-test does not point to handedness forthe SUMHI value (t(99) 5 1.678: P 5 0.097), and wefound a borderline right-hand preference for theMHI value (t(99) 5 1.943: P 5 0.055). Additionally,we carried out another one sample t-test with theZ-score values, and results were consistent with thatobtained for HI (t(99) 5 1.621: P 5 0.108). Theevaluation of the population preferences in bothparts of the study yielded right asymmetry for thefirst part (Test 11Test 2) (t(99) 5 2.480: P 5 0.015),but not for the second part (Test 31Test 4) (t(99) 5

1.104: P 5 0.272) of the experiment. The estimationof the population preferences for each test indicatesthat Test 1 (t(99) 5 2.455: P 5 0.016) and Test 2(t (99) 5 2.270: P 5 0.025) show clear right popula-tion asymmetry; Test 3 (t(99) 5 1.795: P 5 0.076)shows borderline significant right asymmetry; andTest 4 (t(99) 5 0.292: P 5 0.771) do not show anykind of significant asymmetries at population level.

Regarding the variables associated to the sampleat population level, there is a right asymmetry forfemales (t(53) 5 2.582: P 5 0.013), but not for males(t(45) 5�0.255: P 5 0.800). The rest of the variables,human reared (t(57) 5 1.251: P 5 0.216), motherreared (t(41) 5 1.104: P 5 0.276), juveniles (t(35) 5

0.689: P 5 0.495), adolescents (t(27) 5 1.582: P 5

0.125), adults (t(35) 5 0.786: P 5 0.437), and Group1–7 have not shown any population asymmetry, asoccur with the actions performed by fingers (t(99) 5

1.606: P 5 0.111) and tools (t(19) 5 0.413: P 5 0.684).

Extension: Chimfunshi Wildlife Orphanage1

Mona Foundation

In order to evaluate the global pattern of handlateralization of chimpanzees housed in naturalistic-intermediate environments, we estimate both HI andABSHI from the MF and CWO as a whole. We

followed the same methodology used by Hopkinset al. [Hopkins, 2006; Hopkins et al., 2004] in theUnited States.

The total sample of MF and CWO includes 114NCh. The mean value for HI is 0.137, so the trend istoward right preferences. There are no significantdifferences between both colonies (U 5 532.000;P 5 0.147). The mean value for the absolute HI(ABS-HI) is 0.603. Here, the MF sample showedhigher strength than the Zambian sample(U 5 359.000; P 5 0.003).

Results for the one sample t-test showed right-handedness at population level in certain conditions(Table V). Both SUMHI (t (113) 5 2.220: P 5 0.028)and MHI (t(113) 5 2.611: P 5 0.010) showed clearright asymmetry. Results of the one sample t-testwith the Z-score values also indicated a rightpreference at population level (t(113) 5 2.246: P 5

0.027). Based on binomial tests, 14 individuals werenonpreferent and 100 are lateralized for this task: 58are right-handed and 42 left-handed. There weredifferences between the number of lateralizedand nonlateralized subjects (w2(1,n 5 114) 5 64,877,P 5 0.000), but there were no differences betweenthe number of right- and left-handed subjects(w2(1,n 5 100) 5 2,560, P 5 0.110).

SUMHI also revealed right population asymme-try for the first part of the study (Test 112;[t(113) 5 3.059: P 5 0.003]) and for the finger use(t(112) 5 2.070: P 5 0.041).

DISCUSSION

At CWO, subjects expressed a right-hand bias.Right- and left-handedness was consistent amongsubjects and the strength of hand preference didnot vary between groups. There were differencesbetween the number of lateralized and nonlatera-lized subjects, but there were no differences betweenthe numbers of right- and left-handed subjects.

When taking into consideration data from CWOand MF, the samples are right lateralized atindividual and population level, without differencesbetween both colonies. However, the MF sampleshowed higher strength than the Zambian sample,which may be an effect of the bigger sample of thelatter.

A similar effect may occur at CWO where sexseems to be the only associated variable that showsdifferences both at individual and population level,because females were more right-handed than males.At MF, this difference has not been observed. CWOhas a sample of 54 females out of 100 individuals,whereas MF has 3 females out of 14 individuals. Thisfact makes it difficult to validate the possible effect ofthe sex variable in MF.

However, the fact that CWO females are moreright-handed than males is an outstanding result,because most studies did not find any difference

TABLE III. Spearman Correlations Between HIValues Across Test Sessions From CWO

HI Test 1 HI Test 2 HI Test 3 HI Test 4

HI Test 1 0.664�� 0.623�� 0.587��

HI Test 2 0.707�� 0.712��

HI Test 3 0.791��

��Significant at Po0.01.

TABLE IV. HI and ABSHI Values ThroughoutExperimental Sessions From CWO

HI ABSHI

Test 1 0.170 0.632Test 2 0.160 0.642Test 3 0.136 0.701Test 4 0.021 0.646

Am. J. Primatol.

286 / Llorente et al.

when evaluating the sex variable. Only the work ofCorp and Byrne [2004] on a wild population ofchimpanzees at Mahale got similar results studyingbimanual coordinated conducts linked to fruit pro-cessing. In that study, males also tended to be moreleft-handed, whereas females were clearly moreright-handed. In our study at CWO, this result iseven at population level. Although few studiesobserve the same pattern, our result is verysuggestive for its similarity to humans, becausewomen show higher preference in using the righthand than men [Annett, 1985, 2002; Medland et al.,2004]. Medland et al. [2004] suggest that this patternin humans may be conditioned by biological (parti-cularly genetic and hormonal) factors. Therefore, thepattern found in our study may reflect the existenceof one common ancestor between chimpanzees andhumans regarding sexual differences in hand later-ality. Furthermore, some authors support an evolu-tive continuity for the relation between bothvariables: sex and hand preference. According tothem, this relation may have a very old phylogenybecause some species of prosimians show similarpatterns [Ward et al., 1990].

None of the features involved in the extractiontechnique were significant at population level at bothCWO and MF. Despite this, subjects removed the

food 95.66% with their fingers (mostly the indexfinger) and 4.34% with tools. According to ourresults, subjects performing extractions with theindex finger preferentially did it with the right hand,and extractions with the little finger and tools did itwith the left hand. Concerning the former, ourresults are consistent with other studies on chim-panzees [Hopkins, 1995] and other primates[Schweitzer et al., 2007]. It looks like the use of theindex finger as an extracting technique seems toencourage the use of the right hand.

Also, there is a relation between the use of thelittle finger, tools, and the left hand. No explanationhas been proposed yet. However, it seems that handlaterality is affected by the distal motions of fingersand hands when performing bimanual complemen-tary tasks—where hands differ in their roles.According to Brinkman and Kuypers [1973], distalmovements require a frequent use of the contral-ateral brain hemisphere, which may explain ourresults. Regarding humans, McManus et al. [1986]found that differences in the extensor muscles andconnections of the tendons of the fingers may explainwhy, in a tapping task, the index finger was the mostefficient and the ring finger the worst. These resultsregarding the use of fingers are coincident with thisstudy: index finger, pinky, and thumb are the most

TABLE V. Results for the One Sample t-Test in Naturalistic Housed Chimpanzees From CWO1MF

MHI score Sample N MHI SE t df P

SUMHI CWO 100 0.107 0.064 1.678 99 0.097MHI CWO 100 0.122 0.063 1.943 99 0.055Test 112 CWO 100 0.159 0.064 2.480 99 0.015Test 314 CWO 100 0.077 0.070 1.104 99 0.272Test 1 CWO 100 0.170 0.069 2.455 99 0.016Test 2 CWO 100 0.160 0.070 2.270 99 0.025Test 3 CWO 100 0.136 0.076 1.795 99 0.076Test 4 CWO 100 0.021 0.072 0.292 99 0.771Finger use CWO 100 0.103 0.064 1.606 99 0.111Tool use CWO 20 0.066 0.159 0.413 19 0.684Males CWO 46 �0.024 0.094 �0.255 45 0.800Females CWO 54 0.219 0.085 2.582 53 0.013Human rear CWO 58 0.103 0.082 1.251 57 0.216Mother rear CWO 42 0.113 0.102 1.104 41 0.276Juveniles CWO 36 0.080 0.117 0.689 35 0.495Adolescents CWO 28 0.172 0.109 1.582 27 0.125Adults CWO 36 0.083 0.106 0.786 35 0.437SUMHI MF 14 0.352 0.211 1.670 13,000 0.119MHI MF 14 0.416 0.198 2.101 13,000 0.056Test 1 MF 14 0.388 0.200 1.939 13,000 0.075Test 2 MF 14 0.510 0.194 2.636 13,000 0.021Test 3 MF 14 0.453 0.203 2.234 13,000 0.044Test 4 MF 14 0.312 0.221 1.413 13,000 0.181Finger use MF 13 0.333 0.228 1.459 12,000 0.170Tool use MF 7 0.468 0.292 1.602 6,000 0.160SUMHI CWO1MF 114 0.137 0.062 2.220 113,000 0.028MHI CWO1MF 114 0.158 0.060 2.611 113,000 0.010Finger use CWO1MF 113 0.130 0.063 2.070 112,000 0.041Tool use CWO1MF 27 0.170 0.141 1.203 26,000 0.240

Am. J. Primatol.

Handedness in Naturalistic Housed Chimpanzees / 287

used fingers and ring finger the least used. In fact, inhumans, similar to chimpanzees [Ogihara et al.,2005], the fingers have two independent extensors(extensor indicis propius) and five (extensor digitiminimi) and one extensor (extensor pollicis longus),which could explain the improved implementation ofcertain actions and agrees with most used fingers inthe tube task. Furthermore, the index finger is themost sensitive because it has the largest neuronalrepresentation in motor cortex [Sutherling et al.,1992]. Therefore, this could be an additional factorcausing it to be used most.

The statistical test used to observe differentconducts between human- and mother-reared chim-panzees have not revealed significant differencesboth in direction and degree of preference. Thus, theoriginal environment and context where theseindividuals came from had no specific effect on theirhand preference patterns. This conclusion has beenalso reached in other studies where the sample waslarge and varied enough to test this variable[Hopkins, 2006]. These results were already shownin an earlier study [Mosquera et al., 2007], where theobservation of hand laterality in the MF highlyhumanized chimpanzees at spontaneous unimanualtasks offered similar results as the wild samples.Unlike that study, here it was possible to evaluatedirectly the original environment into the MF group.So, results seem to point that the environmentcannot explain the disparity of results regardingthe current pattern of hand preference in nonhumanprimates. If neither the environmental hypothesisnor humanization can explain the results, maybe thecomplexity of the task does explain it, as suggestedby other authors [Fagot & Vauclair, 1991; Lilak &Phillips, 2008; Riba et al., 2009; Schweitzer et al.,2007; Vauclair et al., 2005]. Actually, studies focusedon tasks requiring bimanual coordination, use oftools, and other psychomotor demands have revealedhigher degrees and intensities of manual prefer-ences, besides the environment in which they werevalued and the humanization degree of the samples.

Finally, some authors suggested that the func-tional and neuroanatomical asymmetries are derivedfeatures exclusive of Homo [Crow, 2004; McManus,2002]. However, our results and other evidence rejectthis hypothesis. On the first hand, several species ofnonhuman primates show neuroanatomical asym-metries of the planum temporale [Cantalupo et al.,2003], the sylvian fissure [Hopkins et al., 2000], thecerebral torque [Holloway & De la Coste-Lareymondie,1982; Hopkins & Marino, 2000], and the motor handarea (knob) [Sherwood et al., 2003], among others.On the second hand, we think that the trend topopulation level right-handedness may have evolu-tionary roots older than the Pan and Homo diver-gence (6–7 million years ago), as suggested by others[Hopkins, 2006]. This way would allow setting up ahomology at least between humans and apes regard-

ing hand laterality, particularly in those tasksrequiring strong cognitive demands and bimanualcoordination.

ACKNOWLEDGMENTS

D.R. has a grant from the Fundacion Atapuercaand L.C. received a grant from the Generalitatde Catalunya (Ajut de mobilitat d’estudiants deMaster). M.M. is a member of the AtapuercaResearch Team. This study has been supported byboth the Spanish Ministerio de Ciencia e Innovacion(MICIIN-HAR2009-07223/HIST) and the Universi-tat Rovira i Virgili (URV-2009AIRE-05). Our thanksto Charles Duke for revision of this article and wealso thank Sheila Siddle, Innocent Mulenga, TonyRauch, and all the managers at Chimfunshi.

REFERENCES

Altmann J. 1974. Observational study of behavior: samplingmethods. Behaviour 49:227–265.

Annett M. 1985. Left, right, hand and brain: the right shifttheory. Hove, UK: Lawrence Erlbaum Associates Ltd.

Annett M. 2002. Handedness and brain asymmetry: the rightshift theory. Hove, UK: Psychology Press.

Annett M. 2006. The distribution of handedness in chimpan-zees: estimating right shift in Hopkins’ sample. Laterality11:101–109.

Bard KA, Hopkins WD, Fort CL. 1990. Lateral bias in infantchimpanzees (Pan troglodytes). Journal of ComparativePsychology 104:309–321.

Blois-Heulin C, Guitton JS, Nedellec-Bienvenue D, Roparts L,Vallet E. 2006. Hand preference in unimanual and bimanualtasks and postural effect on manual laterality in captivered-capped mangabeys (Cercocebus torquatus torquatus).American Journal of Primatology 68:429–444.

Boesch C. 1991. Handedness in wild chimpanzees. InternationalJournal of Primatology 12:541–558.

Braccini S, Lambeth SP, Schapiro SJ, Tecumseh Fitch W.2010. Bipedal tool use strengthens chimpanzee handpreferences. Journal of Human Evolution 58:234–241.

Bradshaw JL, Rogers LJ. 1993. The evolution of lateralasymmetries, language, tool-use and intellect. San Diego:Academic Press.

Brinkman J, Kuypers HG. 1973. Cerebral control of contral-ateral and ipsilateral arm, hand and finger movements inthe split-brain rhesus monkey. Brain 96:653–674.

Cantalupo C, Pilcher DL, Hopkins WD. 2003. Are planumtemporale and sylvian fissure asymmetries directly related?A MRI study in great apes. Neuropsychologia 41:1975–1981.

Chapelain A, Hogervorst E. 2009. Hand preferences forbimanual coordination in 29 bonobos (Pan paniscus).Behavioural Brain Research 196:15–29.

Colell M, Segarra MD, Sabater Pi J. 1995a. Hand preferencesin chimpanzees (Pan troglodytes), bonobos (Pan paniscus),and orangutans (Pongo pygmaeus) in food reaching andother daily activities. International Journal of Primatology16:413–434.

Colell M, Segarra MD, Sabater Pi J. 1995b. Manual lateralityin chimpanzees (Pan troglodytes) in complex tasks. Journalof Comparative Psychology 109:298–307.

Corp N, Byrne RW. 2004. Sex difference in chimpanzeehandedness. American Journal of Physical Anthropology123:62–68.

Crow T. 2004. Directional asymmetry is the key to the origin ofmodern Homo sapiens (the Broca-Annett axiom): a reply to

Am. J. Primatol.

288 / Llorente et al.

Roger’s review of The Specialization of Modern HomoSapiens. Laterality 9:233–242.

Crow TJ. 2005. The cerebral torque and directional asymmetryfor hand use are correlates of the capacity for language inHomo sapiens. Behavioral and Brain Sciences 28:595–596.

Fagot J, Vauclair J. 1991. Manual laterality in nonhumanprimates: a distinction between handedness and manualspecialization. Psychological Bulletin 109:76–89.

Holloway RL, De la Coste-Lareymondie MC. 1982. Brainendocast asymmetry in pongids and hominids: somepreliminary findings on the paleontology of cerebraldominance. American Journal of Physical Anthropology58:101–110.

Hopkins WD. 1993. Posture and reaching in chimpanzees (Pantroglodytes) and orangurans (Pongo pygmaeus). Journal ofComparative Psychology 107:162–168.

Hopkins WD. 1995. Hand preferences for a co-ordinatedbimanual task in 110 chimpanzees (Pan troglodytes):cross-sectional analysis. Journal of Comparative Psychology109:291–297.

Hopkins WD. 2006. Comparative and familial analysis ofhandedness in great apes. Psychological Bulletin 132:538–599.

Hopkins WD, Bard KA. 1995. Asymmetries in spontaneoushead orientation in infant chimpanzees (Pan troglodytes).Behavioral Neuroscience 109:808–812.

Hopkins WD, Bard KA. 2000. A longitudinal study of handpreference in chimpanzees (Pan troglodytes). DevelopmentalPsychobiology 36:292–300.

Hopkins WD, Cantalupo C. 2003. Does variation in sample sizeexplain individual differences in hand preferences ofchimpanzees (Pan troglodytes)? An empirical study andreply to Palmer (2002). American Journal of PhysicalAnthropology 121:878–881.

Hopkins WD, Cantalupo C. 2004. Handedness in chimpanzees(Pan troglodytes) is associated with asymmetries of theprimary motor cortex but not with homologous languageareas. Behavioral Neuroscience 118:1176–1183.

Hopkins WD, Cantalupo C. 2005. Individual and settingdifferences in the hand preferences of chimpanzees(Pan troglodytes): a critical analysis and some alternativeexplanations. Laterality 10:65–80.

Hopkins WD, Fernandez-Carriba S. 2000. The effect ofsituational factors on hand preferences for feeding in 177captive chimpanzees (Pan troglodytes). Neuropsychologia38:403–409.

Hopkins WD, Marino L. 2000. Asymmetries in cerebral widthin nonhuman primate brains as revealed by magneticresonance imaging (MRI). Neuropsychologia 38:493–499.

Hopkins WD, Pearson K. 2000. Chimpanzee (Pan troglodytes)handedness: variability across multiple measures of handuse. Journal of Comparative Psychology 114:126.

Hopkins WD, Rabinowitz DM. 1997. Manual specialisation andtool use in captive chimpanzees (Pan troglodytes): the effectof unimanual and bimanual strategies on hand preference.Laterality 2:267–277.

Hopkins WD, Washburn DA, Rumbaugh DM. 1989. Note onhand use in the manipulation of joysticks by rhesusmonkeys (Macaca mulatta) and chimpanzees (Pan troglo-dytes). Journal of Comparative Psychology 103:91–94.

Hopkins WD, Pilcher DL, MacGregor L. 2000. Sylvian fissureasymmetries in nonhuman primates revisited: a compara-tive MRI study. Brain, Behavior and Evolution 56:293–299.

Hopkins WD, Bard KA, Jones A, Bales SL. 1993. Chimpanzeehand preference in throwing and infant cradling: implica-tions for the origin of human handedness. Current Anthro-pology 34:786–790.

Hopkins WD, Fernandez-Carriba S, Wesley MJ, Hostetter A,Pilcher D, Poss S. 2001. The use of bouts and frequencies inthe evaluation of hand preferences for a coordinatedbinamual tasks in chimpanzees (Pan troglodytes): an

empirical study comparing two different indices of later-ality. Journal of Comparative Psychology 115:294–299.

Hopkins W, Stoinski T, Lukas K, Ross S, Wesley M. 2003a.Comparative assessment of handedness for a coordinatedbimanual task in chimpanzees (Pan troglodytes), gorillas(Gorilla gorilla), and orangutans (Pongo pygmaeus). Journalof Comparative Psychology 117:302–308.

Hopkins WD, Hook M, Braccini S, Schapiro SJ. 2003b.Population-level right handedness for a coordinated biman-ual task in chimpanzees: replication and extension in asecond colony of apes. International Journal of Primatology24:677–689.

Hopkins WD, Wesley MJ, Izard MK, Hook M, Schapiro SJ.2004. Chimpanzees (Pan troglodytes) are predominantlyright-handed: replication in three populations of apes.Behavioral Neuroscience 118:659–663.

Hopkins WD, Cantalupo C, Freeman H, Russell J, Kachin M,Nelson E. 2005. Chimpanzees are right-handed whenrecording bouts of hand use. Laterality 10:121–130.

Hopkins WD, Russell JL, Cantalupo C. 2007. Neuroanatomicalcorrelates of handedness for tool use in chimpanzees (Pantroglodytes) implication for theories on the evolution oflanguage. Psychological Science 18:971–977.

Humle T, Matsuzawa T. 2002. Ant-dipping among thechimpanzees of Bossou, Guinea, and some comparisonswith other sites. American Journal of Primatology 58:133–148.

Lilak AL, Phillips KA. 2008. Consistency of hand preferenceacross low-level and high-level tasks in capuchin monkeys(Cebus apella). American Journal of Primatology 70:254–260.

Llorente M, Mosquera M, Fabre M. 2009. Manual laterality forsimple reaching and bimanual coordinated task in natur-alistic housed chimpanzees (Pan troglodytes). InternationalJournal of Primatology 30:183–197.

Matsuzawa T, Biro D, Inoue-Nakamura N, Tonooka R,Yamakoshi G. 2001. Emergence of culture in wild chimpan-zees: education by master-apprenticeship. In: Matsuzawa T,editor. Primate origins of human cognition and behavior.Tokyo: Springer.

McGrew WC, Marchant LF. 1997. On the other hand: currentissues in and meta-analysis of the behavioral laterality ofhand function in nonhuman primates. Yearbook of PhysicalAnthropology 40:201–232.

McManus IC. 2002. Right hand, left hand: the origins ofasymmetry in brains, bodies, atoms, and cultures. London:Weidenfeld & Nicolson.

McManus IC, Kemp RI, Grant J. 1986. Differences betweenfingers and hands in tapping ability: dissociation betweenspeed and regularity. Cortex 22:461–473.

Medland SE, Perelle I, Monte VD, Ehrman L. 2004. Effects ofculture, sex, and age on the distribution of handedness: anevalutation of the sensitivity of three measures of handed-ness. Laterality 9:287–297.

Meguerditchian A, Calcutt SE, Lonsdorf EV, Ross SR,Hopkins WD. 2010. Captive gorillas are right-handed forbimanual feeding. American Journal of Physical Anthro-pology 141:638–645.

Meunier H, Vauclair J. 2007. Hand preferences on unimanualand bimanual tasks in white-faced capuchins (Cebuscapucinus). American Journal of Primatology 69:1064–1069.

Mosquera M, Llorente M, Riba D, Estebaranz F, Gonzalez M,Lorenzo C, Sanmartı N, Toll M, Carbonell E, Feliu O. 2007.Ethological study of manual laterality in naturalistic housedchimpanzees (Pan troglodytes) from the Mona FoundationSanctuary (Girona, Spain). Laterality 12:19–30.

Ogihara N, Kunai T, Nakatsukasa M. 2005. Muscle dimen-sions in the chimpanzee hand. Primates 46:275–280.

Palmer AR. 2002. Chimpanzee right-handedness reconsidered:evaluating the evidence with funnel plots. American Journalof Physical Anthropology 118:191–199.

Am. J. Primatol.

Handedness in Naturalistic Housed Chimpanzees / 289

Palmer AR. 2003. Reply to Hopkins and Cantalupo: chimpan-zee right-handedness reconsidered. Sampling issues anddata presentation. American Journal of Physical Anthro-pology 121:382–384.

Phillips KA, Hopkins WD. 2007. Exploring the relationshipbetween cerebellar asymmetry and handedness in chimpan-zees (Pan troglodytes) and capuchins (Cebus apella).Neuropsychologia 45:2333–2339.

Riba D, Llorente M, Mosquera M, Feliu O. 2009. Handpreference in simple and complex tasks in naturalisticallyhoused chimpanzees at the Mona Foundation (Spain). FoliaPrimatologica 80:406.

Schweitzer C, Bec P, Blois-Heulin C. 2007. Does the complex-ity of the task influence manual laterality in De Brazza’smonkeys (Cercopithecus neglectus)? Ethology 113:983–994.

Sherwood CC, Broadfield DC, Holloway RL, Gannon PJ,Hof PR. 2003. Variability of Broca’s area homologue inAfrican great apes: implications for language evolution. TheAnatomical Record: Part A 271A:276–285.

Sutherling WW, Levesque MF, Baumgartner C. 1992. Corticalsensory representation of the human hand: size of fingerregions and nonoverlapping digit somatotopy. Neurology 42:1020–1028.

Vauclair J, Meguerditchian A, Hopkins WD. 2005. Handpreferences for unimanual and coordinated bimanual tasksin baboons (Papio anubis). Cognitive Brain Research 25:210–216.

Ward JP, Milliken GW, Dodson DL, Stafford DK, Wallace M.1990. Handedness as a function of sex and age in a largepopulation of Lemur. Journal of Comparative Psychology104:167–173.

Warren JM. 1980. Handedness and laterality in humans andother animals. Physiological Psychology 8:351–359.

Wesley MJ, Fernandez-Carriba S, Hostetter A, Pilcher D,Poss S, Hopkins WD. 2002. Factor analysis of multiplemeasures of hand use in captive chimpanzees: an alternativeapproach to the assessment of handedness in nonhumanprimates. International Journal of Primatology 23:1155–1168.

Westergaard GC, Suomi SJ. 1996. Hand preference for abimanual task in tufted capuchins (Cebus apella) and rhesusmacaques (Macaca mulatta). Journal of ComparativePsychology 110:406–411.

Zhao D, Gao X, Li B. 2010. Hand preference for spontaneouslyunimanual and bimanual coordinated tasks in wild Sichuansnub-nosed monkeys: implication for hemispheric speciali-zation. Behavioural Brain Research 208:85–89.

Am. J. Primatol.

290 / Llorente et al.