Behavioural Brain Research 173 (2006) 237–245

Research report

Manual laterality in Campbell’s Monkeys (Cercopithecus c. campbelli) in spontaneous and experimental actions Amandine Chapelain a,b,∗ , Philippe Bec b,1 , Catherine Blois-Heulin b,2 a

b

Department of Human Sciences, Loughborough University, Loughborough, Leicestershire LE11 3TU, UK Laboratoire e´ thologie, e´ volution, e´ cologie, CNRS UMR 6552, Station biologique Paimpont, 35380 Paimpont, France Received 18 July 2005; received in revised form 15 June 2006; accepted 23 June 2006

Abstract Behavioural asymmetries, once thought to be exclusively human, appear to be widespread in vertebrates. A population-level bias should stem from natural selection and reflect a cerebral dominance, while individual preferences might be linked to individual characteristics. Manual laterality has been extensively investigated in non-human primates. However, despite a strong data base, no general patterns have emerged, resulting in a few explanatory theories and little consensus. This study was interested in manual laterality in 12 Campbell’s monkeys (Cercopithecus c. campbelli). Several theories were examined, using both direct behavioural observations during feeding behaviour and six controlled experimental conditions, in which we varied task demands to investigate the effect of two factors. We systematically varied the individual posture (sat, tripedal, bipedal, clung) and the complexity of the task (box with or without a lid to open). Concerning the direction of preference, we found individual and action-specific preferences for experimental actions, which match previous reports. No population bias emerged and each subject appeared to react differently to the factors, hereby contradicting the theories. However, concerning the strength of preference, all individuals tended to be affected in similar ways. Spontaneous actions were less lateralized than experimental ones, and the simplest task and spontaneous category tended to show the weakest laterality. The relative complexity and novelty of these actions may account for these results. © 2006 Elsevier B.V. All rights reserved. Keywords: Hand preference, Cercopithecine, Posture, Complexity, Novelty

1. Introduction Laterality has long been considered a human characteristic, bound to language and handedness, but in the last decades, it has been shown in other species such as monkeys, birds, amphibians and fishes [45,75]. Why do animals prefer one side over the other? Regarding manual preferences, since the asymmetries affect only the function but not the structure of the hands, the preference would reflect a functional lateralization of the cerebral hemispheres [9]. The dominance of one hemisphere must actually provide many computational advantages (e.g. saving neural space by avoiding replication of functions, allow∗

Corresponding author. Tel.: +44 1509 223048; fax: +44 1509 223940. E-mail addresses: [email protected], chapelain [email protected] (A. Chapelain), [email protected] (P. Bec), [email protected] (C. Blois-Heulin). 1 Tel.: +33 2 99 61 81 67; fax: +33 2 23 23 69 27. 2 Tel.: +33 2 99 61 81 65; fax: +33 2 99 61 81 88/23 23 69 27. 0166-4328/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2006.06.028

ing simultaneous processing of different processes, avoiding hemispheric competition [9,60,61,73]) and as expected, brain lateralization may improve cognitive abilities and behavioural efficiency (i.e. [10,19,22,43,47,60,63]). The commonly held opinion is that only a population-level laterality may stem from evolutionary processes, because individual-level laterality might derive from individual factors (ontogeny, sex, age) [76] that have been shown to influence laterality (i.e. [25,53,54,77,81]). Alternatively, recent articles [23,74] suggest that laterality may evolve in two steps. Firstly, laterality may appear on an individual basis to provide computational advantages [60,61,73]. Indeed the advantages of lateralization in terms of brain efficiency are very valid on an individual basis. The second step would be an alignment of the direction of the asymmetries in most individuals of the population, as a result of social pressures [23,74]. When asymmetrical organisms have to interact together, it may be essential for an individual to adapt its laterality according to the asymmetries of the others (i.e. for school cohesion) [23]. How this hypothesis could apply to hand preference is not

238

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

clear [32], but the crucial point here is that individual-level laterality may stem from natural selection, and may be maintained if there are no requirements for social alignment. Most humans exhibit a right-hand preference, though the strength of laterality may vary according to the task [3,13,26,49,59,70]. To understand the origins and adaptive value of human handedness, studies have focussed on non-human primates (NHP). The study of hand preferences in NHP is a non-invasive means to investigate cerebral lateralization, since hand preference is an indicator of brain lateralization in humans [50]. Despite a strong data base on manual laterality in NHP, no general patterns have emerged, resulting in few explanatory theories [17,52,78] and little consensus. Studies on NHP have provided results that seem remote from the human pattern. They have reported the existence of individual preferences but rarely population-level biases [45,57]. Moreover, the strength and direction of preferences appeared to vary according to the task [66,69,78]. This raises a methodological issue because it makes results dependent on the method of assessment. Unfortunately, research is plagued by methodological inconsistencies, which prevent reliable between studies comparisons. However, research has shown that NHP exhibit functional asymmetries, which may be affected by different factors, bound to the task (complexity, sensory mode, etc.) or to the subject (posture, experience, sex, age, etc.). Three main theories have been proposed regarding hand preference in NHP. Along with the “postural origins theory” [52], laterality could stem from an adaptation to unimanual predation in primitive primates. The left hand could have become specialized for visually guided reaches, the right hand becoming specialized for postural support [51]. Then, as primates became terrestrial, the right hand became specialized for manipulations. This theory has given rise to a great number of new studies that mostly refute it [45,57]. The “theory of bipedalism” is also based on the influence of posture and proposes that the appearance of right-handedness could be bound to the adoption of the bipedal posture. Along with this, a bipedal stance would enhance the use of the right hand to reach in NHP, which has been reported in apes [29,33,56] and monkeys [79,80]. According to the “theory of task complexity” [17] laterality (population-level bias and strong preferences) would appear only for complex tasks (i.e. novel, precise, sequential or bimanual action) [17]. This theory has been confirmed in NHP [6,16,18,34,78] and humans [3,13,26,70]. Our study investigated the effects of these possible influential factors on manual preference of an arboreal primate that has not been studied before: the Campbell’s monkey. Two kinds of approaches have been developed. Firstly, spontaneous use of the hands during feeding behaviour has been recorded. Secondly, the individuals have been subjected to six experimental tasks to establish the effect of each factor. We assessed the influence of posture (sat, tripedal, bipedal, clung) and task complexity (box with or without a lid to open) on the expression of manual preference. The aim was to examine how each subject was affected by these factors, since the sample size (12 subjects) enable us to assess only general trends concerning group-level effects.

2. Materials and methods 2.1. Subjects Campbell’s monkeys (Cercopithecus c. campbelli) are arboreal guenons of Ivory Coast. The subjects were 12 Campbell’s monkeys from a social group housed at the Biological Station, Paimpont (France): six adult females (B, C, Lo, Pl, Sh, Ti), two subadult males (Pi, To) and four juvenile males (A, E, St, Y) aged from 1 to 11.5 years. All were captive-born and none has ever been involved in experimental tasks. The cage had an outdoor part (3 m× 6 m× 4.2 m) and an indoor part (4 m× 5.6 m× 2.9 m). The observations were conducted indoors. Both cages were provided with perches. The ground of the indoor cage was covered in litter. Animals received fresh fruit and vegetables in the morning and chow in the afternoon. Water was provided ad libitum.

2.2. Spontaneous feeding actions We collected data for familiar actions that belong to the natural repertoire of the subjects. Use of the hands was studied in spontaneous feeding during usual meal (apples, bananas, oranges, carrots) in the morning. Food items scattered on the ground were cut in pieces (2 cm) so that monkeys could pick them up with one hand. 2.2.1. Observational procedure A 5-min Focal-Animal Sampling [1] was recorded for each monkey on each day, for 57 days. Each subject was observed between 210 and 380 min. Sixtyeight observation hours were conducted from February to April 2004. The order in which the monkeys were observed was changed daily to avoid any influence of depletion of food or satiety of the subjects. 2.2.2. Behaviours A feeding behavioural list was made during a 1 week preliminary study. For each behaviour we recorded: which hand was used (right or left), what the other hand action was (inactive, postural support, holding a food item, holding a food item against the ground) and what the posture of the subject was (sat, bipedal, tripedal, quadrupedal) [45]. An individual had to exhibit a behaviour at least six times to be included in analysis (to perform binomial tests). For comparative analyses, we retained only the behaviours with at least four subjects available. Only the behaviours for which we could distinguish between a main action and a secondary action were retained. Comparative analyses were conducted on 26 behaviours (Table 1). We made two breakdowns of the behaviours. The first gathers behaviours that share the same main action. The second classifies behaviours into six categories of increasing complexity according to the secondary action (action of the other hand). Category 1 corresponds to the secondary action “inactive”, 2 to “postural support”, 3 to “hold a food item”, 4 to another action and the last gathers the displacements (D). We also recorded the teat used when young individuals suckle from their mothers. 2.2.3. Data independency To ensure data independency we used bouts rather than frequency. Only the first pattern of a sequence of identical patterns was recorded. Two identical actions were considered independent only if they were separated by a different action [44,45,48].

2.3. Experimental tasks The experimental tasks were conducted in the afternoon. Each subject was isolated in one of the three indoor cage compartments so that the presence of competitive congeners could not influence the use of the hands. Three days of habituation preceded each task to allow subjects to accustom to the apparatus. 2.3.1. Simple reach tasks with variation of postural demands 2.3.1.1. Biped and sat. The subject had to pass his arm through the wire-net to take a seed of sunflower from the observer’s hand. The seed was presented in the symmetry plan of the subject body to eliminate artefacts due to the position of the object relative to the hand (NHP: Ogle 1871 in [24], [15,50,76], humans: [68]).

Table 1 Details for spontaneous behaviours (values) No.

Main action

C

Secondary action

N

Lat

NLat

RH

LH

p

Mean HI

HI S.E.

Mean ABSHI

ABSHI S.E.

1 2

Sat Sat

Take a food item Take a food item

1 2

12 5

2 1

10 4

1 0

1 1

0.931 0.721

0.004 −0.133

0.015 0.156

0.149 0.667

0.007 0.054

3 4 5 6

Sat Sat Triped Sat

Take a food item Take a food item Take a food item Take a food item through the mesh Give to the other hand Take from the other hand Hold to mouth Hold to mouth

3 2 2 2

Inactive Hold a food item against the ground Hold a food item Postural support Postural support Postural support

12 12 12 4

6 9 4 0

6 3 8 4

3 3 2 0

3 6 2 0

0.351 0.320 0.904 0.741

0.137 −0.128 0.013 0.068

0.041 0.035 0.03 0.094

0.421 0.39 0.25 0.282

0.021 0.015 0.02 0.051

12 4 12 8

9 2 6 2

3 2 6 6

4 1 5 1

5 1 1 1

0.771 0.773 0.104 0.612

−0.043 0.143 0.131 0.093

0.042 0.226 0.021 0.062

0.425 0.714 0.221 0.404

0.019 0.101 0.015 0.033

12 12 12 4 7 10 11 8 7 12 11 12 8 12 5 8

8 5 2 0 0 3 6 2 1 7 2 9 1 0 0 1

4 7 10 4 6 7 5 6 6 5 9 3 7 11 5 7

7 3 1 0 0 2 2 1 0 6 2 2 1 1 0 1

1 2 1 0 0 1 4 1 1 1 0 7 0 0 0 0

0.111 0.486 0.698 0.048 0.453 0.306 0.382 0.971 0.608 0.022 0.019 0.089 0.452 0.062 0.575 0.218

0.162 0.068 0.047 −0.265 0.143 0.166 −0.109 −0.006 0.088 0.244 0.247 −0.265 0.125 0.076 −0.134 −0.202

0.027 0.027 0.034 0.041 0.068 0.048 0.036 0.052 0.061 0.026 0.027 0.041 0.055 0.011 0.098 0.053

0.305 0.248 0.348 0.265 0.333 0.463 0.354 0.356 0.386 0.338 0.296 0.484 0.405 0.126 0.4 0.364

0.015 0.017 0.016 0.041 0.049 0.016 0.017 0.021 0.02 0.017 0.022 0.021 0.02 0.006 0.05 0.034

7 8 9 10

Sat Sat Sat Sat

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Sat Sat Triped Triped Sat Sat Sat Triped Sat Sat Sat Triped Quadru Sat Sat Sat

Hold to mouth Hold to mouth Hold to mouth Hold to mouth Move litter aside Move litter aside Move litter aside Move litter aside Move litter aside Take food out from mouth Take food out from mouth Hold a food item Hold a food item Drop a food item Put food item back Reposition itself on the food item

1 3 1 2 3 2 2 D 1 3 2 2 4 3 2 D D 1 1

Inactive Hold a food item Inactive Hold a food item against the ground Hold a food item Postural support Postural support Postural support Inactive Hold a food item Postural support Postural support Keep litter apart Hold a food item Postural support Postural support Postural support Inactive Inactive

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

Posture

Number of subjects right-handed (RH), left-handed (LH), lateralized (Lat) and not lateralized (NLat). Values of HI (mean + S.E.) and ABSHI (mean + S.E.) for each spontaneous behaviour. No.: number attributed to the behaviour. C—category of behaviours according to the action of the other hand: 1 for inactive, 2 for postural support, 3 for hold a food item, 4 for other action, D for displacements. N: number of subjects taken into account. p: p of t-test. Bold: significant result.

239

240

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

To compel the subject to adopt a bipedal posture, the seed was given 50 cm above ground (30 cm for juveniles), which corresponds to the shoulder height of an upright subject [29]. The seed was given 25 cm above ground (15 cm for juveniles) for “sat”. The hand used to take the seed was noted for each reach. For “biped” the monkey had to be stood upright facing the observer and maintaining both hind limbs on the floor while reaching. For “sat”, the monkey had to be sat on the floor facing the observer. The other hand had to lie on the wire-net for standardization [29]. To avoid any influence of the observer, the hand presenting the seed was randomly alternated. 2.3.1.2. Clung. The subject had to take a seed on a tray fixed onto the wirenet outside of the cage 130 cm above ground. To solve this task, the monkey had to be clinging with one hand and two feet while putting his arm through a hole made up in the wire-net. The seed was placed right in front of the hole to eliminate artefacts due to the position of the object [15,50,76]. The hand used to take the seed was noted only when the monkey had climbed right up under the tray and was directly in front of it while reaching, to avoid artefacts due to the position of the monkey relative to the tray. For the three previous tasks, the seed was placed at an identical distance from the subject to obtain a comparable extension of the arm for the three tasks. 2.3.1.3. Triped. The subject had to pick up a seed from the floor (first cleared of its litter) [80]. The observer threw a seed at the background of the cage in straight line behind the subject sat in front of her. The hand used to take the seed was noted only when the reach was tripedal and direct. That is, when both feet and one hand touched the ground, limbs were stretched and the subject did not move the litter aside before reaching. Only contralateral and central reaches were noted in order to avoid artefacts due to the position of the object [15]. These four postural tasks were classified according to increasing postural demands (relative instability and number of points of support): “sat”, “triped”, “biped” and “clung” from the least to the most demanding task. 2.3.2. Tasks with variation of complexity 2.3.2.1. Box. The subject had to take a seed out of a metal box with a lid kept closed by a spring. Both hands had to be used in a sequential action: the monkey had to open the lid and keep it open with one hand, while taking the seed with the other hand [58,72]. The box was attached onto the wire-net inside the cage 5 cm above ground. The subjects adopted a bipedal posture to look inside the box while reaching. The hand used to take the seed was noted when the subject was upright in front of the box while reaching, to eliminate artefacts due to the relative position of the box to the monkey. 2.3.2.2. Box without lid. Here the lid of the previous box was kept open against the wire-net. The hand used to take the seed out of the open box was noted only when the monkey maintained both hind limbs on the floor and one hand leaning on the edge of the box while reaching, so that the posture was similar to the one adopted for “box”. Requiring a simple unimanual action, “box without lid” was considered less complex than “box” that demanded a sequential bimanual action. 2.3.3. Number of trials per subject Each subject made 200 trials for each task because fewer trials may not be appropriate for assessing individual preference properly [44,45,48]. No more than 20 trials were made per day to limit possible effect of action repetition [66]. 2.3.4. Data independency To ensure independency of reaches, a seed was thrown in the background of the cage after each trial [44,45,48]. The subject had to move off and pick up the seed, before coming back and repositioning itself in front of the apparatus for another trial.

2.4. Statistical analysis We calculated a handedness index (HI) score for each subject using the formula: (R − L)/(R + L) [31,79]. R and L represent the number of times the right

and left hands were used, respectively. HI reveals the direction of preference: positive values indicate a bias towards the use of the right hand, and negative values a bias towards the left hand. The absolute value of HI (ABSHI) was used to characterise the strength of preference [31,79]. The binomial test (B) [67] evaluated the significance of differences between the use of the left and right hands for each subject. Group-level biases were evaluated by one sample t-test (T) [67] performed on HI values. The effects of the factors investigated were tested by Friedman two-way analysis of variance by ranks (F), Wilcoxon signed ranks (W), Kruskall–Wallis one-way analysis of variance (KW), Mann–Whitney (MW) and Spearman rank-order correlation coefficient (S) [67]. Biases were considered significant if p ≤ 0.05. The Bonferroni’s correction was used to correct for multiple comparison effects [67]. Females and adults categories being confounded, age and sex effects could not be tested.

3. Results 3.1. Spontaneous actions Among the 26 behaviours kept in analysis, only five behaviours (6, 14, 15, 24, 25) did not induce any significant individual preference (Table 1). 3.1.1. Effect of the behaviour (Table 1) The behaviours “take out/hold” (20) and “take out/support” (21) presented a group-level bias toward the right hand to take food out from the mouth (Tp = 0.022 and 0.019 respectively). To suckle, all three subjects were strongly lateralized (mean ABSHI = 0.857). A and E preferred the right teat and St the left one (Bp = 0.003, 0.004 and 1 × 10−3 respectively). 3.1.2. Consistency between behaviours The direction and strength of laterality varied depending on the behaviour (F, d.f. = 10, p = 1 × 10−4 ). Four subjects (B, E, St and To) were always lateralized, whatever the behaviour was (Bp = 0.053, 0.017, 0.041 and 0.006 respectively). Three subjects showed a manual specialization (same preference across actions): A, B and Sh were right-handed (Bp = 0.016, 0.031 and 2 × 10−3 respectively). 3.1.3. Effect of the main action (Table 2) For the main action “take out from mouth” and “hold to mouth”, the right hand was more used than the left one at the group-level (Tp = 0.001 and 0.035 respectively). The strength (KW, d.f. = 7, p = 0.003) but not the direction (KW, d.f. = 7, p = 0.332) of laterality was affected by the main action. Table 2 Direction (HI mean + S.E.) and strength (ABSHI mean + S.E.) of manual preferences for each main action whatever the secondary action was Main action

HI

S.E.

ABSHI

S.E.

p

Take out from mouth Take from other hand Hold to mouth Drop Move aside Take a food item Change hand Put back

0.246 0.143 0.101 0.076 0.047 −0.001 −0.043 −0.134

0.013 0.226 0.006 0.011 0.010 0.007 0.042 0.098

0.318 0.714 0.298 0.126 0.381 0.333 0.425 0.4

0.009 0.101 0.004 0.006 0.005 0.004 0.019 0.05

0.001 0.773 0.035 0.062 0.479 0.981 0.771 0.575

p: p values of t-test. Bold: significant result.

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

241

Table 3 Direction (HI mean + S.E.) and strength (ABSHI mean + S.E.) of manual preferences for each category based on secondary action, whatever the main action was Categories

HI

S.E.

p

ABSHI

S.E.

p

1 2 3 D

0.039 (ab) 0.019 (ab) 0.175 (a) −0.135 (b)

0.006 0.004 0.009 0.019

0.578 0.652 0.007 0.166

0.256 (a) 0.342 (b) 0.405 (b) 0.421 (b)

0.004 0.002 0.005 0.009

0.578 0.652 0.007 0.166

Categories of secondary actions—1: other hand inactive, 2: other hand used for postural support, 3: other hand used to hold a food item, 4: other hand used for another action, D: displacements. p: values of t-test, Bold: significant result. Letters—Mann–Whitney results: same letters, no significant differences; different letters, significant difference.

3.1.4. Effect of the secondary action (action of the other hand) (Table 3) Category 3 (hold a food item) showed a group-level bias for the right hand to do a main action while the left holds a food item (Tp = 0.007). The direction and strength of laterality varied according to the secondary action (KW, d.f. = 2, p = 0.012 and 0.007 respectively).

Fig. 1. Laterality values for experimental tasks. Histogram: HI mean (+S.E.); point: ABSHI mean (+S.E.).

eralized, whatever the task was (Bp = 0.031). No subject showed a manual specialization (preference consistent between tasks) (Bp > 0.219).

3.2. Experimental tasks (Table 4) For each task the number of lateralized subject was greater than the number of not lateralized subjects (Bp ≤ 0.006), except “triped”. No group-level bias was found whatever the task was (Tp > 0.11). 3.2.1. Effect of the task and consistency between tasks The direction and strength of preference did not vary according to the task (d.f. = 5, F = 4.810, p = 0.44 and F = 8.286, p = 0.141). Five subjects (A, C, Lo, Pi, Sh) were always lat-

3.2.2. Effect of the posture (Fig. 1) The strength (d.f. = 5, F = 10.227, p = 0.017) but not the direction (d.f. = 5, F = 2.455, p = 0.484) of preference varied depending on the posture. The strength was weaker for “triped” compared to other tasks (except “sat” when Bonferroni’s correction was applied). “Sat” and “biped” were positively correlated for the direction and strength of preference (S for HI “sat“/”biped”: p = 0.002, r = 0.795; for ABSHI: p = 0.023, r = 0.646). The classification based on the observed strength of preference did not correspond to the one based on postural demands.

Table 4 Manual preferences (number of right-hand uses) for the experimental tasks Subjects

Sat

Biped

Clung

Triped

Box without lid

Box

L

R

Lat

A B C E Lo Pi Pl Sh St Ti To Y

18 194 198 195 5 36 54 16 46 95 11 192

24 172 193 199 11 187 184 10 187 144 8 194

188 190 20 197 195 28 14 3 4 27 121 121

177 106 133 89 0 46 89 185 146 90 88

170 50 15 200 4 2 46 130

145 198 12

3 42 30

1 2

2 1 3 0 4 5 3 4 2 3 4 1

4 4 3 4 2 1 2 2 2 1 1 3

6 5 6 4 6 6 5 6 4 4 5 4

11 0 0.001 3 8 0.112

9 0 0.004 4 5 0.575

Lat NLat p of binomial test Lat/NLat RH LH p of t-test on HI

11 1 0.006 4 7 0.632

12 0 0.0001 8 4 0.308

12 0 0.0001 6 6 0.757

6 5 1 4 2 0.791

178 0 197 2

Lat: number of lateralized subjects, NLat: number of not lateralized subjects, RH: number of right-handed subjects, LH: number of left-handed subjects. p: values of p for tests. Bold: significant results. L, R, Lat: number of tasks for which the subject is left-handed, right-handed and lateralized respectively.

242

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

3.2.3. Effect of the complexity (Fig. 1) There was no significant difference between “Box” and “Box without lid” concerning the direction and strength of preference (Wp = 0.859 and 0.11 respectively). 3.3. Comparison between experimental and spontaneous actions 3.3.1. Direction of preference The direction did not differ between experimental and spontaneous actions (MWp = 0.222). The subjects that showed a manual specialization for spontaneous actions (A, B, Sh) did not display it for experimental ones. None of the subjects was coherent between experimental and spontaneous actions for the direction of preference. 3.3.2. Strength of preference The strength was greater in experimental actions compared to spontaneous ones (MWp < 1 × 10−3 ). However, the strength for “triped” was not different from any spontaneous category (MWp > 0.05 and >0.424 respectively). Ten out of 12 subjects expressed stronger preferences for experimental actions in comparison with spontaneous ones (MWp < 0.043). 3.3.3. Experimental versus spontaneous tripedal reach Laterality for the spontaneous action “take a food item/postural support” in tripedal posture (5) did not differ from the corresponding experimental task “triped” (direction: Wp = 0.594, strength: Wp = 0.075). 4. Discussion 4.1. General results We reported evidence of individual-level preferences in Campbell’s monkeys for both spontaneous and experimental actions. No group-level bias was found (except for a few spontaneous behaviours). There are several possible explanations for the absence of group-level bias, the first being the small sample size. According to Hopkins [32] a minimum sample of 59 subjects would be required to detect a group-level bias, because the effect to be shown is small. However, our data match studies showing a weak laterality for spontaneous actions [25,27,43,62,65,71,76]. This could be related to the customs and diet of the species because ambilaterality seems to be the norm in arboreal species [25], and monkeys with a generalist diet (vast behavioural repertory) seem to show weaker preferences than monkeys with a specialized diet (stereotyped behaviours) [4]. For spontaneous and experimental actions, most of the subjects showed action-specific preferences. This inter-task inconsistency is in keeping with the existing literature in NHP [6,25,27,28,35,71,78]. In humans, the direction of preference seems to be mostly consistent between tasks, and this would be an evidence of brain lateralization [41]. Therefore some people think that the inconsistency observed in NHP testify to an absence of cerebral dominance. According to Warren [78], taskspecificity proves that preference is learned throughout the prac-

tice of the task, but for Corballis [9], it just reflects a weaker form of laterality that does not justify the human/NHP discontinuity. However, one must note that tasks used with humans are different from those used with NHP [68]. Tasks used with humans usually assess cognitive asymmetries, whereas those used with NHP assess motor asymmetries [68]. Generally, tasks used with humans involve tool-use or complex precise actions, whereas tasks used with NHP are much simpler (but see [42,46] for tooluse, [55] for humanlike tasks). Importantly, when humans are tested with tasks similar to those used with NHP, their preferences are less pronounced [3,19,68,70]. If the preferences of NHP are related to cerebral asymmetry, the observed taskspecificity shows that its expression may be affected by different factors (i.e. complexity, posture). The effect of some possible influential factors will now be considered. 4.2. Effect of the experimental or spontaneous character of the action The character of the action did not influence the direction of preference. Individuals showed no consistency between the two kinds of actions for the direction of preference, which is in keeping with previous studies [21,28]. The number of lateralized subjects was large for experimental actions (except “triped”) but weak for spontaneous ones. This result corresponds to the literature showing that experimental actions are more likely to reveal hand preference [45,72]. Moreover, experimental actions (except “triped”) induced stronger preferences than spontaneous ones. This result matches data for humans and NHP [19,21,27,45,65,72] and may be explained by some characteristics of the experimental actions. The greater complexity of experimental tasks could account for the stronger laterality. Indeed, experimental actions (except “triped”) possessed both attributes of complex actions according to the “theory of task complexity” [17]: they were new and had spatiotemporal or cognitive demands. The fact that most of the subjects were not lateralized for the familiar spontaneous actions fits in with this explanation. “Triped” also supports this because it was nor demanding nor new (similar to the spontaneous action (5)), and its strength of laterality was similar to the one observed for spontaneous actions. The repetitive character of experimental actions was mentioned as a preferences creator [78], but data regarding this factor are contradictory (stronger preferences: [39,40,41], no change: [65,66]). Here, a possible influence of repetition was minimized by the compelled displacement of the subjects between trials and the small number of trials per session [66]. 4.3. Effect of the complexity of the task The “theory of task complexity” [17] proposes that: (a) only complex tasks could induce a group-level bias, which would reflect a cerebral dominance; (b) preferences would be stronger for complex compared to simple actions, which are more easily affected by random factors. Our results would go along with the second prediction because the simplest task (“triped”) tended to show the weakest laterality compared to other tasks, which is

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

in keeping with previous studies [8,29,38,66]. All other tasks, as new tasks with postural or motor demands, may possess sufficient complexity to induce a strong laterality. The strong laterality observed for these tasks matches the theory, and the data from humans and NHP [6,16,18,21,70]. 4.3.1. Box tasks “Box” required a complex bimanual action so it was expected to induce a group-level bias similar to the one previously reported for bimanual actions [18,34,42,69]. We found no grouplevel bias in “box”. Quiatt and Derr [58] showed a right grouplevel bias, but no group level-bias was found in other studies for a box task [5,8,14,50,56,65,72]. Regarding the strength of preference, “Box” tended to induce the strongest preference compared to the other tasks, which would match the theory [17]. The strength did not significantly differ between the sequential and non-sequential box tasks, which matches the data of Marchant and Steklis [50], but differs from the one of Quiatt and Derr [58] who found a greater laterality for the sequential task. The effect of complexity was not significant here, while it was in some other studies [2,8,58,69,72]. Some researchers think that only a precise sequential bimanual manipulation on one object, with a different role of each hand (handle/hold) could induce a strong laterality [7,62]. Our task did not demand such a coordinated bimanual action on one object, so it might not be complex enough. 4.4. Effect of the posture of the subject According to the “postural origins theory” [52]: (a) manual laterality may be an adaptation required for the control of an unstable posture; (b) the right hand would be specialized for postural support. Laterality would be enhanced when postural demands are important, according to the need to maintain balance [11,66]. For spontaneous and experimental actions, the right hand was not preferred for postural support. However, the strength tended to increase with the raising of postural demands. 4.4.1. Bipedal posture Along with the “theory of bipedalism”, the bipedal posture would induce a group-level bias towards the right hand to reach. In “biped”, we found a trend towards the right hand but no significant group-level bias. This result does not match studies showing a significant right group-level bias in bipedal posture [29,33,56,69,79]. However, the use of the right hand tended to increase from “sat” to “biped” (three left-handed and one not lateralized subjects became right-handed for “biped”), which tends to match previous studies showing an increase of right hand use in bipedal posture [8,29,33,56,69] (see [80] for a review). The absence of right group-level bias in “biped” could be related to the small sample size. Alternatively, several explanations could account for this result [56]. The first explanation could be the help provided by the other hand clung onto the wire-net. Indeed, the preference for the right hand seems to be stronger when both hands are free, in comparison with when one is used for postural support [12]. Secondly, Campbell’s monkeys are arboreal and

243

the right bias in bipedal posture seems to be weaker in arboreal compared to terrestrial species [33,51,56]. This being in line with the theory that right-handedness could have appeared with the transition from the arboreal to the terrestrial habitat [51]. The strength of preference was high for “biped”, which corresponds to previous studies showing that bipedal posture increases the use of the preferred hand [29,38,56,66]. 4.4.2. Clung posture The “postural origins theory” [52] predicts a right-hand preference for support in the demanding clung posture. No group-level bias appeared for “clung”, which corresponds to the results of several studies [2,28,64], but differs from other studies [15,37]. The relative weak strength of laterality in “clung” goes against the hypothesis but matches data from several studies [2,15,28]. This posture may not be complex enough, in terms of novelty, to induce laterality since it is close to a usual posture for arboreal species. This may explain the weaker preference in “clung” compared to “biped” that is an atypical posture. 4.4.3. Tripedal posture “Triped” induced no group-level bias. This result does not fit in with “postural origins theory” [52] (right hand for support), but corresponds to the “theory of task complexity” [17] (no group-level bias) and matches the literature [2,29,33,37,56,66,69] (see [30] for a review). This absence of group-level bias could be accounted for by the influence of several factors. One factor could be related to the last hand placed on the ground during the displacement preceding the reach [15]. Another artefact could be due to a preference to turn over one side to come back to the observer. As the monkey reached almost at the same time as it turned, the hand used to reach could have been affected by the side of turning [64]. Such lateral biases have been reported in other NHP [38]. Regarding the strength of preference, “triped” tended to show the weakest strength compared to the other tasks and few subjects were lateralized, which corresponds to the literature [2,14,15,18,20,27,29,33,56,66,81]. According to the postural demands, “triped” was expected to be more demanding than “sat”. However, “triped” may be less demanding in terms of visuospatial demands since it did not require visuospatial guidance through the wire-net as other postural tasks did [15]. 5. Conclusion We examined the effect of several possible influential factors on manual preferences of Campbell’s monkeys. Individual-level preferences have been shown for both spontaneous and experimental actions, and they appeared to be affected by the investigated factors (complexity, posture). The absence of group-level bias could be related to the low number of subjects. Alternatively, this could be accounted for by an insensibility of the measures to reveal hemispheric dominance, or an absence of endogenous asymmetry [6]. The presence of individual preferences suggests that endogenous laterality exists in these monkeys, but its expression seemed to be influenced by different factors resulting in task-specificity. This study confirms the theory of a multifac-

244

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245

torial origin of manual laterality. The choice of the hand could depend on the combination of several variables, which could be (a) extrinsic factors such as complexity [2,69], sensory mode or object position; (b) intrinsic individual factors such as posture, arousal, emotional state [36], temperament, age or sex. A greater number of subjects is required to examine the effect of such variables that could influence each subject individually. Finally, the standardization of assessment protocols between future studies appears to be essential, since inter-study inconsistencies may stem from methodological differences. Acknowledgements We would like to thank all the persons of the Biological Station and anonymous reviewers for their advice and assistance. References [1] Altmann J. Observational study of behavior: sampling methods. Behaviour 1974;49:227–67. [2] Anderson JR, Degiorgio C, Lamarque C, Fagot J. A multi-task assessment of hand lateralization in capuchin monkeys (Cebus apella). Primates 1996;37:97–103. [3] Annett M. Left, right, hand and brain: the right shift theory. Hillsdale, NJ: Erlbaum; 1985. [4] Bennett AJ, Ward JP, Milliken GW, Stafford DK. Analysis of lateralized components of feeding behavior in the ring-tailed lemur (Lemur catta). J Comp Psychol 1995;109:27–33. [5] Blois-Heulin C, Guitton JS, Nedellec-Bienvenue D, Ropars L, Vallet E. Hand preference in unimanual and bimanual task and postural effect on manual laterality in captive red-capped mangabey (Cercocebus torquatus torquatus). Am J Primatol; in press. [6] Byrne RW, Byrne JM. Hand preferences in the skilled gathering tasks of mountain gorillas (Gorilla g. berengei). Cortex 1991;27:521–46. [7] Byrne RW, Corp N, Byrne JME. Manual dexterity in the gorilla: bimanual and digit role differentiation in a natural task. Anim Cogn 2001;4:347–61. [8] Colell M, Segarra D, Sabater-Pi J. Manual laterality in chimpanzees in complex tasks. J Comp Psychol 1995;3:298–307. [9] Corballis MC. Laterality and human evolution. Psychol Rev 1989;96: 492–505. [10] Crow TJ, Crow LR, Done DJ, Leask S. Relative hand skill predicts academic ability: global deficits at the point of hemispheric indecision. Neuropsychologia 1998;36:1275–82. [11] De Vleeschouwer K, Van Elsacker L, Verheyen RF. Effect of posture on hand preferences during experimental food reaching in bonobos. J Comp Psychol 1995;109:203–7. [12] Diamond AC, Mac Grew WC. True handedness in the cotton-top tamarin (Saguinus oedipus). Primates 1994;35:69–77. [13] Fagart J, editor. Droitiers/Gauchers. Des asym´etries dans tous les sens. Solal; 2004. [14] Fagot J. Behavioural and hemispheric lateralization in non-human primates. Doctoral Dissertation. Universit´e Aix-Marseille I, France; 1988. [15] Fagot J, Drea CM, Wallen K. Asymmetrical hand use in rhesus monkeys (Macaca mulatta) in tactually and visually regulated tasks. J Comp Psychol 1991;105:260–8. [16] Fagot J, Vauclair J. Handedness and bimanual coordination in the lowland gorilla. Brain Behav Evol 1988;32:89–95. [17] Fagot J, Vauclair J. Manual laterality in nonhuman primates: a distinction between handedness and manual specialization. Psychol Bull 1991;109:76–89. [18] Fagot J, Vauclair J. La lat´eralisation chez les singes. Recherche 1993;252:299–304. [19] Faurie C. L’´evolution du polymorphisme de lat´eralit´e dans les populations humaines. Doctoral Dissertation. Universit´e de Montpellier, France; 2004.

[20] Forsythe C, Milliken GW, Stafford DK, Ward JP. Posturally related variations in the hand preferences of the ruffed lemur (Varecia variegata variegata). J Comp Psychol 1988;102:248–50. [21] Fragaszy DM, Adams-Curtis LE. An exploration of manual preference and performance in crabeating macaques. In: Ward JP, Hopkins WD, editors. Primate laterality: current behavioral evidence of primate asymmetries. Springer-Verlag; 1993. p. 75–105. [22] Fragaszy DM, Mitchell SR. Hand preference and performance on unimanual and bimanual tasks in capuchin monkeys. J Comp Psychol 1990;104:275–82. [23] Ghirlanda S, Vallortigara G. The evolution of brain lateralization: a game theoretical analysis of population structure. Proc Roy Soc Lond 2004;271:853–7. [24] Harris L. On the evolution of handedness: a speculative analysis of Darwin’s views and a review of early studies of handedness in “the nearest allies of man”. Brain Lang 2000;73:132–88. [25] Harrison KE, Byrne RW. Hand preferences in unimanual and bimanual feeding by wild vervet monkeys (Cercopithecus aethiops). J Comp Psychol 2000;114:13–21. [26] Healey JM, Liederman J, Geschwind N. Handedness is not a unidimensional trait. Cortex 1986;22:33–53. [27] Heestand JE. Behavioral lateralization in four species of apes? Doctoral Dissertation. University of Washington; 1986. [28] Hook-Costigan MA, Rogers LJ. Hand, mouth and eye preferences in the common marmoset (Callithrix jacchus). Folia Primatol 1995;64:180– 91. [29] Hopkins WD. Posture and reaching in chimpanzees (Pan troglodytes) and orangutans (Pongo pygmaeus). J Comp Psychol 1993;107:162–8. [30] Hopkins WD. Chimpanzee handedness revisited: 55 years since Finch. Psychon Bull Rev 1996;3:449–57. [31] Hopkins WD. On the other hand: statistical issues in the assessment and interpretation of hand preference data in nonhuman primates. Int J Primatol 1999;20:851–66. [32] Hopkins WD. Comparative and familial analysis of handedness in great apes. Psychol Bull; in press. [33] Hopkins WD, Bennett AJ, Bales SL, Lee J, Ward JP. Behavioral laterality in captive bonobos (Pan paniscus). J Comp Psychol 1993;107:403– 10. [34] Hopkins WD, Cantalupo C. Individual and setting differences in the hand preferences of chimpanzees (Pan troglodytes): a critical analysis and some alternative explanations. Laterality 2005;10:65–80. [35] Hopkins WD, Pearson K. Chimpanzee (Pan troglodytes) handedness: variability across multiple measures of hand use. J Comp Psychol 2000;114:126–35. [36] King JE. Laterality in hand preferences and reaching accuracy of cotton top tamarins (Saguinus Oedipus). J Comp Psychol 1995;109:34–41. [37] King JE, Landau VI. Manual preference in varieties of reaching in squirrel monkeys. In: Ward JP, Hopkins WD, editors. Primate laterality: current behavioural evidence of primate asymmetries. New York: Springer-Verlag; 1993. p. 107–24. [38] Larson CF, Dodson DL, Ward JP. Hand preferences and whole (Galago senegalenses). Brain Behav Evol 1989;33:261–7. [39] Lehman RAW. Persistence of primate hand preference despite initial training to the contrary. Behav Brain Res 1980;1:547–51. [40] Lehman RAW. Distribution and changes in strength of hand preference in cynomolgus monkeys. Brain Behav Evol 1980;17:209–77. [41] Lehman RAW. Manual preference in prosimians, monkeys and apes. In: Ward JP, Hopkins WD, editors. Primate laterality: current behavioural evidence of primate asymmetries. New York: Springer-Verlag; 1993. p. 149–81. [42] Lonsdorf EV, Hopkins WD. Wild chimpanzees show populationlevel handedness for tool use. Proc Natl Acad Sci USA 2005;102: 12634–8. [43] Mac Grew WC, Marchant LF. Chimpanzees, tools and termites: hand preference or handedness? Curr Anthropol 1992;33:114–9. [44] Mac Grew WC, Marchant LF. On which side of the apes? Ethological study of hand use. In: Mac Grew WC, Marchant L, Nishida, editors. Great ape societies. Cambridge University Press; 1996. p. 255–72.

A. Chapelain et al. / Behavioural Brain Research 173 (2006) 237–245 [45] Mac Grew WC, Marchant LF. On the other hand: current issues in and meta-analysis of the behavioral laterality of hand function in non human primates. Yearb Phys Anthropol 1997;40:201–32. [46] Mac Grew WC, Marchant LF. Using the tool at hand: manual laterality and elementary technology in Cebus spp. and Pan spp. Int J Primatol 1997;18:787–810. [47] Mac Grew WC, Marchant LF. Laterality of hand use pays off in foraging success for wild chimpanzees. Primates 1999;40:509–13. [48] Marchant LF, Mac Grew WC. Laterality of function in apes: a meta-analysis of methods. J Hum Evol 1991;21:425–38. [49] Marchant LF, Mac Grew WC, Eibl-Eibesfeldt I. Is human handedness universal? Ethological analyses from three traditional cultures. Ethology 1995;101:239–58. [50] Marchant LF, Steklis HD. Hand preference in a captive island group of chimpanzees. Am J Primatol 1986;10:301–13. [51] Mac Neilage PF. Implications of primate functional asymmetries for the evolution of cerebral hemispheric specialisations. In: Ward JP, Hopkins WD, editors. Primate laterality: current behavioural evidence of primate asymmetries. New York: Springer-Verlag; 1993. p. 319–41. [52] Mac Neilage PF, Studdert-Kennedy MG, Lindblom B. Primate handedness reconsidered. Behav Brain Sci 1987;10:247–303. [53] Milliken GW, Forsythe C, Ward JP. Multiple measures of hand-use lateralization in the ring-tailed lemur. J Comp Psychol 1989;103:262–8. [54] Milliken GW, Stafford DK, Dodson DL, Pinger CD, Ward JP. Analyses of feeding lateralization in the small-eared bushbaby (Otolemur garnettii): a comparison with the ring-tailed lemur (Lemur catta). J Comp Psychol 1991;105:274–85. [55] Morris RD, Hopkins WD, Bolser-Gilmore L. Assessment of hand preference in two-language trained chimpanzees (Pan troglodytes): a multimethod analysis. J Clin Exp Neuropsychol 1993;15:487–502. [56] Olson DA, Ellis JE, Nadler RD. Hand preferences in captive gorillas, orangutans and gibbons. Am J Primatol 1990;20:83–94. [57] Papademetriou E, Sheu CF, Michel G. A meta-analysis of primate hand preferences, particularly for reaching. J Comp Psychol 2005;119:33–48. [58] Quiatt D, Derr S. Rhesus monkey hand preferences in lifting lids and reaching for food. In: Proceedings of the XIV Congress International Primatological Society. 1994. p. 37. [59] Raymond M, Pontier D. Is there geographical variation in human handedness? Laterality 2004;9:35–51. [60] Rogers LJ. Evolution of hemispheric specialization: advantages and disadvantages. Brain Lang 2000;73:236–53. [61] Rogers LJ. Advantages and disadvantages of lateralization. In: Rogers LJ, Andrew RJ, editors. Comparative vertebrate lateralization. Cambridge University Press; 2002. p. 126–53. [62] Rogers LJ, Kaplan G. Hand preferences and other lateral biases in rehabilitated orang-utans (Pongo pygmaeus pygmaeus). Anim Behav 1996;51:13–25.

245

[63] Rogers LJ, Zucca P, Vallortigara G. Advantages of having a lateralized brain. Proc Biol Sci 2004;271:S420–2. [64] Roney LS, King JE. Postural effects on manual reaching laterality in squirrel monkeys and cotton-top tamarins. J Comp Psychol 1993;107: 380–5. [65] Rothe H. Handedness in the common marmoset (Callithrix jacchus). Am J Phys Anthropol 1973;38:561–6. [66] Sanford C, Guin K, Ward JP. Posture and laterality in the bushbaby (Galago senegalensis). Brain Behav Evol 1984;25:217–24. [67] Seagel S, Castellan NJ. Nonparametric statistics for the behavioural sciences. Singapore: McGraw-Hill; 1988. [68] Seltzer C, Forsythe C, Ward JP. Multiple measures of motor lateralization in human primates (Homo sapiens). J Comp Psychol 1990; 104:159–66. [69] Spinozzi G, Grazia Castorina M, Truppa V. Hand preferences in unimanual and coordinated-bimanual tasks by tufted capuchin monkeys. J Comp Psychol 1998;112:183–91. [70] Steenhuis RE, Bryden MP. Different dimensions of hand preference that relate to skilled and unskilled activities. Cortex 1989;25:289–304. [71] Teichroeb JA. A review of hand preferences in nonhuman primates: some data from patas monkeys. Lab Prim Newsletter 1999;38:1–5. [72] Trouillard E, Blois-Heulin C. Manual laterality and task complexity in De Brazza’s monkey (Cercopithecus neglectus). Laterality 2005;10: 7–27. [73] Vallortigara G. Comparative neuropsychology of the dual brain: a stroll through animals’ left and right perceptual worlds. Brain Lang 2000;73:189–219. [74] Vallortigara G, Rogers LJ. Survival with an asymmetrical brain: advantages and disadvantages of cerebral lateralization. Behav Brain Sci 2005;28:575–89. [75] Vallortigara G, Rogers LJ, Bisazza A. Possible evolutionary origins of cognitive brain lateralization. Brain Res Rev 1999;30:164–75. [76] Vauclair J, Fagot J. Spontaneous hand usage and handedness in a troop of baboons. Cortex 1987;23:265–74. [77] Ward JP, Milliken GW, Dodson DL, Stafford DK, Wallace M. Handedness as a function of sex and age in a large population of Lemur. J Comp Psychol 1990;104:167–73. [78] Warren JM. Handedness and laterality in humans and other animals. Physiol Psychol 1980;8:351–9. [79] Westergaard GC, Kuhn HE, Lundquis TAL, Suomi SJ. Posture and reaching in tufted capuchins (Cebus apella). Laterality 1997;2:65–74. [80] Westergaard GC, Kuhn HE, Suomi SJ. Bipedal posture and hand preference in humans and other primates. J Comp Psychol 1998;112:55–64. [81] Westergaard GC, Lussier ID, Higley JD. Between species variation in the development of hand preference among macaques. Neuropsychologia 2001;39:1373–8.