- Email: [email protected]
Unconstrained 3D-kinematics of prehension in five primates: Lemur, capuchin, gorilla, chimpanzee, human
Journal of Human Evolution 65 (2013) 303e312
Contents lists available at ScienceDirect
Journal of Human Evolution journal homepage: www.elsevier.com/locate/jhevol
Unconstrained 3D-kinematics of prehension in five primates: Lemur, capuchin, gorilla, chimpanzee, human Elodie Reghem a, b, Laurence Chèze c, Yves Coppens d, Emmanuelle Pouydebat a, * a
UMR 7179, Département Ecologie et Gestion de la Biodiversité, MNHN, 55 rue Buffon, case postale 55, FR-75231 Paris, France EA 4322 Handibio, Université du Sud Toulon Var, La Garde, France c IFSTTAR, LBMC, UMR_T9406, Université Lyon 1, 43 Boulevard du 11 novembre, 69622 Villeurbanne, France d Chaire de Paléoanthropologie et Préhistoire, Collège de France, 11 Place Marcelin Berthelot, 75005 Paris, France b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 February 2013 Accepted 26 June 2013 Available online 29 July 2013
Primates are known for their use of the hand in many activities including food grasping. Yet, most studies concentrate on the type of grip used. Moreover, kinematic studies remain limited to a few investigations of the distal elements in constrained conditions in humans and macaques. In order to improve our understanding of the prehension movement in primates, we analyse here the behavioural strategies (e.g., types of grip, body postures) as well as the 3D kinematics of the whole forelimb and the trunk during the prehension of small static food items in five primate species in unconstrained conditions. All species preferred the quadrupedal posture except lemurs, which used a typical crouched posture. Grasp type differed among species, with smaller animals (capuchins and lemurs) using a whole-hand grip and larger animals (humans, gorillas, chimpanzees) using predominantly a precision grip. Larger animals had lower relative wrist velocities and spent a larger proportion of the movement decelerating. Humans grasped food items with planar motions involving small joint rotations, more similar to the smaller animals than to gorillas and chimpanzees, which used greater rotations of both the shoulder and forearm. In conclusion, the features characterising human food prehension are present in other primates, yet differences exist in joint motions. These results provide a good basis to suggest hypotheses concerning the factors involved in driving the evolution of grasping abilities in primates. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Grasping evolution Kinematics Forelimb
Introduction All of the members of the Order Primates share the characteristic ability to grasp food or objects with one or both hands (Napier, 1960; Bishop, 1964; Costello and Fragaszy, 1988; Christel, 1993; MacFarlane and Graziano, 2009; Pouydebat et al., 2009; Reghem et al., 2011). This ability is a key behaviour implied in daily activities in both humans and non-human primates and its evolution remains unresolved (Pouydebat et al., 2008). Although all primates are able to grasp, different food prehension strategies have been identified with respect to both manual grip postures (Bishop, 1964; Christel, 1993; Spinozzi et al., 2004; Pouydebat et al., 2011) and the kinematics of the forelimb (Scott and Kalaska, 1997; Christel and Billard, 2002). In kinematic studies, prehension is often described as consisting of two phases. The first phase is the transport phase
* Corresponding author. E-mail address: [email protected] (E. Pouydebat). 0047-2484/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jhevol.2013.06.011
where the hand is transported to the object to be grasped. The second one is the grasping phase where the grip aperture between thumb and index opens and ultimately closes around the object (Jeannerod, 1981, 1984). The kinematics of prehension in primates have been investigated only in humans and macaques. In addition, most studies focus on the kinematical similarities; that is to say the kinematic parameters common between humans and macaques (Scott and Kalaska, 1997; Roy et al., 2000, 2002, 2006; Sartori et al., 2012). Both species show a similar bell-shaped pattern of the wrist velocity with a longer deceleration than acceleration phase, and a grip aperture varying according to the object size. However, most of these results were obtained for macaques in constrained conditions (e.g., intensive learning of the movement, grasping through a slot, parts of the body constrained, the head immobilised) that induce stereotyped movements, a bias in elbow motion, and which restrict shoulder motion despite the importance of the proximal joints in forelimb movement. In part because of these constrained conditions, many studies have focused only upon the distal components, i.e., the wrist kinematics and grip aperture. One study (Christel and
304
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Billard, 2002) investigated untrained macaques in unconstrained condition (i.e., in their group within their habitual enclosure) and quantified the ranges of motion of the trunk and the forelimb joints in comparison with humans. Despite the fact that similar results were obtained in comparison with prior studies on wrist velocity and grip aperture, Christel and Billard (2002) emphasised differences in kinematic strategies of both species. Indeed, macaques were faster than humans and showed more variability. Moreover, humans and macaques performed opposite movements. During the reaching phase, macaques adopted a larger elbow, wrist and trunk motion and a smaller shoulder motion, contrary to what is observed in humans. Therefore, the trunk of macaques contributes in a greater proportion to the prehension movement, compensating for the lower excursion at the shoulder. The authors suggested that variation in motor control, posture, and morphology could explain the differences. Indeed, macaques use their forelimb in locomotion, which may imply a different shoulder joint morphology, muscular strength, body posture, and a change in the motor control driving the observed differences (Christel and Billard, 2002). Whether these kinematic parameters are specific to particular primates or general features of non-human primates remains, however, unclear. Only very few kinematic studies have been conducted under unconstrained conditions involving spontaneous and free gestures. These conditions appear, however, fundamental to understanding both the source of variation in movement and the prehension mechanism. Thus, to address questions pertaining to the evolution of prehension and the human specificities such studies are needed. In this context, the following research questions are addressed in the present study: How do several primate species of different phylogenetically distinct taxa perform their prehension movement? Are the similarities in motor control observed in humans and macaques shared by all non-human primates? Here, we hypothesise that the velocity profiles and the range of motion will differ across the species, depending of their size and morphology. To address these questions, we examine here unconstrained prehension in five species representative of different groups of primates: the ring-tailed lemur (Lemur catta, prosimian), the yellow-breasted capuchin (Sapajus xanthosternos, New World monkey), gorillas (Gorilla gorilla), chimpanzees (Pan troglodytes) and humans (Homo sapiens). We quantify the 3D-kinematics of the reaching and the type of grip used during a prehension task in unconstrained conditions. Reaching is described by the ranges of motion of the forelimb joints and the trunk, and by the wrist velocity profile. Grasping is here described by the grip postures and not by the movement of the fingers. The behavioural strategies during prehension are quantified and include body posture, the position of the subjects relative to the food, and their distance to the food. Material and methods Species Seven adults humans (two women and five men, mean age: 33.0 7.3 years) were investigated. The study was carried out in accordance with ‘The ethical codes of the World Medical Association’ (Declaration of Helsinki). For the non-human primates, the experimental protocol used adhered to the legal requirements of the European Union and the American Association of Physical Anthropologists Code of Ethics. Three untrained adults were recorded for each species. Chimpanzees (P. troglodytes, three females, mean age: 19.7 3.8 years), gorillas (Gorilla gorilla, two females and one male, mean age: 17.6 4.1 years), and capuchins (Sapajus xanthosternos, two females and one male, mean age: 7.6 5.6 years)
were filmed in the zoological park of La Palmyre, France. Lemurs (Lemur catta, three males, mean age: 3.0 years) were filmed in the zoological park of Jardin Zoologique Tropical, France. Food selection and food size We selected raisins as our target food because all subjects consume fruit as part of their diet. The size of the food was standardised according to the length of the hand calculated based on the landmarks of the wrist and those on the metacarpal heads. The size of the raisin was one-quarter of this length for each species (Table 1). The lemurs were different when grasping raisins compared with the other species as they systematically grasped the raisins with the mouth. The size of food for which the lemurs systematically used their hand corresponded to a small apple morsel. We chose to keep the small food item as a basis of our study because it was comparable with the only study on macaques recorded in unconstrained conditions, thus allowing our data to be compared with that available for macaques (Christel and Billard, 2002). Furthermore, the small food size was chosen to test the type of grip (power or precision) used in different species. Indeed, not all of the non-human primates are able to use precision grips to grasp small food items (e.g., lemurs) and small food sizes can reveal more behavioural versatility (many types of precision grips) than large food sizes (mainly involving a power grip). Experimental set up for humans Reaching and grasping movements were recorded at 100 frames/second using a Motion AnalysisÒ system (LBMC, Lyon; Motion Analysis Corporation, Santa Rosa, USA) with eight EagleÒ cameras surrounding the subject. Subjects received no particular instruction except to reach, grasp, and move the raisin to the mouth. When in a sitting posture, participants sat on a stool with the hands rested on the table. In the quadrupedal posture, body weight was supported by the knees and the palms of the hands. Raisins were scattered on a surface of 50 cm2 in front of the subject. Twelve infrared reflective markers were positioned on anatomical landmarks of the forelimb and the trunk (Fig. 1A) as recommended by the International Society of Biomechanics (ISB) (Wu et al., 2005). Experimental set up for non-human primates Data were acquired using five video cameras (SanyoÒ Full HD) at 60 frames/second. The food was scattered on the ground on a surface of 50 50 cm. All of the non-human primates were in unconstrained conditions, meaning that they were not trained prior the recordings. They were in their habitual enclosure, in a group, and totally free to choose their body posture and their distance relative to the food. As no markers could be placed on the animals in zoos for reasons of security (and the animals would have taken off any external markers immediately), ten landmarks were manually digitised at the same body location as those in humans (see kinematic analysis for more details).
Table 1 Food size according to the species. Species Humans Chimpanzees Gorillas Capuchins Lemurs
Small food size (cm)
Large food size (cm3)
1.5 1.7 1.7 0.5 0.5
_ _ _ _ 2
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
305
Figure 1. Position of the landmarks (A) and the segment coordinate systems (B) on the forelimb. Two technical markers placed on the arm and one on the forearm helped to reconstruct the motion of forelimb in humans. For the five species (humans, chimpanzees, gorillas, capuchins, lemurs), only the trunk landmarks [1] and [2] were used for reasons of visibility. A segment coordinate system (SCS) was computed for each segment (trunk, arm, forearm) using a minimum of three landmarks per segment. Then, the orthogonal rotation matrix from proximal SCS to distal SCS was computed for each joint and interpreted so that the first rotation (flexion/extension) was around the z axis embedded on the proximal SCS, the last rotation (internal/external rotation) was around the y axis embedded on the distal SCS and the abduction/adduction took place around the x floating axis. Trunk angles were obtained using ground and trunk SCS, shoulder angles using trunk and arm SCS, and elbow angles using arm and forearm SCS. Landmark legend (A): [1] processus spinosus of the seventh cervical vertebra, [2] eighth thoracic vertebra, [3] suprasternal notch, [4] xiphoid process, [5, 6] left and right dorsal point on the acromioclavicular joint, [7, 8] medial and lateral epicondyles of humerus, [9, 10] radial and ulnar styloid processes, [11, 12] head of second and fifth metacarpus. Segment coordinate system legend (B): G: Ground, T: Trunk, A: Arm, F: Forearm.
Behavioural analysis of non-human primates Four behavioural variables were collected for 100 grasps per subject (total ¼ 1200 grasps) from video recordings (sessions of 20 minutes/day): body posture (sitting, squatting, quadrupedal), the position of the subject relative to the food (in front, at left or at right), hand preference, and preferred grip type. Precision grips are defined here as a grasp between the thumb and index tips on the lateral, pulp or medial side of the distal part of the fingers (Christel, 1993; Pouydebat et al., 2009). The power grips describe a grasp with the palm and all of the fingers, and the scissor grip describes a grasp between the index and the third finger. As the existence of kinematic differences between the use of the dominant and nondominant hand remains debated in the literature (Grosskpof and Kuhtz-Buschbeck, 2006), we selected the dominant hand of each subject in our kinematic analysis. Hand preferences were considered for subjects who had a significant z-score (Table 2). As only one chimpanzee subject did not show a dominant hand, we arbitrarily analysed its right hand. These behavioural variables helped us to select comparable trials among all of the species for the movement analysis. The preferred grip type and dominant hand used by each subject were retained for subsequent kinematic analysis. Kinematic analysis Kinematic data was analysed for five trials for each subject (15 grasp sequences per species for the non-humans and 35 grasp sequences for humans). First, as the non-humans preferentially grasped food in front of them (see results), we retained prehension movements in front of the subjects for all of the species to standardise the comparison. Second, we selected comparable food distances by calculating the ratio of each species’ forelimb length (arm þ forearm) to the food distance. The most used food distances were retained in which the food was no further than an arm’s
length away. Third, for all of the sequences selected, the hand was on the ground in a stationary position before being moved forward to reach. For human subjects, the software EvaRTÒ 5.0 reconstructed the marker trajectories and extracted their coordinates in 3D. For nonhumans, video calibration, manual digitisation (landmarks were manually digitised frame by frame) of the anatomical landmarks on each 2D video frame and the reconstruction of the 3D coordinates were performed using a custom-written MatlabÒ routine (Loco 3.3). For all of the species in our sample, kinematic data were processed using MatlabÒ (The MathWorks, Inc., Natick, Massachusetts). Data were low-pass filtered at a frequency of 6 Hz with a second order dual-pass Butterworth filter. Movement onset of the grasp (reaching phase) was identified based on the wrist velocity profile and was defined as the time when wrist velocity reached 5% of its peak (Alstermark et al., 1993; Santello et al., 2002; Graham et al., 2003). All variables were obtained in a trunk-centered frame of reference. The X axis was oriented toward the food, the Y axis was vertical and the Z axis was directed laterally (Fig. 1B). We quantified several variables for the wrist kinematics analysis. The wrist velocity was calculated from the filtered three dimensional Cartesian coordinates (x, y, z) of the mid-point of the two wrist landmarks. The wrist velocities were converted to dimensionless units to reduce the effect of the size of the species and allow inter-specific comparisons (Hof, 1996; Vereecke et al., 2006). Dimensionless velocities were obtained from velocities divided by the mean of the arm plus forearm length of all of the subjects for each species. The mean velocity, the maximal amplitude, and the time to the velocity peak were measured. The mean velocity and the maximal amplitude are reported both in absolute values and in dimensionless values. The dimensionless wrist velocity curves were then interpolated for each species allowing resampling on 100 points in order to superimpose the curves, to compute the mean patterns and represent them relative to the
306
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 2 List of the subjects of all of the species and scores of their manual preference (right-handers ¼ R, left-handers ¼ L or ambidextrous ¼ A). Species
Subjects
Sex
Age in years
Forelimb length (cm)
Mass (kg)
HI
Z-score
Manual preference
Humans
1 2 3 4 5 6 7 1 2 3 1 2 3 1 2 3 1 2 3
Male Female Female Male Male Female Male Female Female Female Male Female Female Male Female Female Male Male Male
28 29 38 22 33 44 37 18 17 24 21 13 19 14 6 3 3 3 3
57.5 58.0 59.0 57.0 57.0 57.0 60.0 65.4 48.3 67.6 87.5 72.1 81.5 24.8 19.1 18.3 22.9 20.7 20.4
68.0 57.0 73.0 62.0 62.0 53.0 72.0 44.0 35.0 55.0 150.0 145.5 65.5 3.3 3.1 2.0 2.3 2.5 3.0
1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.07 0.20 0.06 0.46 0.20 0.60 0.24 0.24 0.34 0.93 1.00 1.00
10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 0.7 NS 2.7 1.96** 0.6 NS 4.6 1.96** 2.0 1.96* 6.0 1.96** 2.4 1.96* 2.4 1.96* 3.1 1.96** 8.9 1.96** 8.5 1.96** 9.4 1.96**
R R R R R R R A R A R L L L L L L R L
Chimpanzees
Gorillas
Capuchins
Lemurs
All of the non-human species were in quadrupedal postures and humans in sitting and quadrupedal postures. The score of humans was the same in both postures. NS: not significant; *: significant at 0.05; **: significant at 0.01; ***: significant at 0.001.
overall movement duration (100%). The movement duration was the time measured from wrist movement onset to the actual grasp. Next, the angles of the trunk (relative to the ground) and the forelimb joints (shoulder, elbow) were computed using the ISB recommendations (Wu et al., 2005) (Fig. 1B). Eight degrees of freedom were quantified: (a) trunk flexion-extension (forward/ backward motion in sitting; downward/upward motion in quadrupedal), (b) trunk inclination (lateral motion of the trunk on the left or right side), (c) trunk rotation, (d) shoulder flexionextension, (e) shoulder abductioneadduction, (f) shoulder rotation, (g) elbow flexion-extension, (h) elbow rotation (i.e., pronation and supination). The ranges of motion were calculated from the absolute differences between minimal and maximal angles for the trunk, the shoulder, and the elbow joints. The curves of the forelimb joints and trunk motion were interpolated over 100 points and averaged for the subjects of each species to represent the patterns relative to the overall movement duration. Statistical analyses All analyses were performed using the R graphical and statistical package v.2.9.0 (R Development Core Team, 2009). The degree of manual asymmetry was calculated for each subject via the handedness index (HI) using the formula (R-L)/(R þ L), where R and L represent the total number of right and left responses (Hopkins, 1999). The HI values ranging from þ1.0 to 1.0 indicate a right hand preference for positive values and a left-hand preference for negative values (Hopkins, 1999). Then, the binomial zscore determined if the manual preference is significant. The subjects were classified with z score, z 1.96 as right-handed, z 1.96 as left-handed, and 1.96 < z < 1.96 as ambipreferent (Hopkins, 1999). Analyses of variance (ANOVA) with one factor (species) and Tukey’s HSD post-hoc tests were used to test for differences in kinematics among species. The absolute values of angular displacements, the time to wrist velocity peak (in % and ms), the duration of the deceleration phase (in % and ms), the movement duration, and the dimensionless values of wrist velocity were tested. A Shapiroe Wilks’ test was used to test the normality of the data and the Bartlett-test was used to check for homogeneity of variance.
Results Behaviour All species showed a hand preference except for two chimpanzees (Table 2). Moreover, all of the species, except lemurs, used precision grips and humans used only this type of grip. All other species used two or three different grip types (Table 3). Gorillas used few scissor grips compared with chimpanzees, which used both in the same proportion. Capuchins and lemurs were the only ones to use power grips. Capuchins mainly used the power grip compared with precision and mouth grips. Lemurs mainly used the power grip and showed the greatest use of the mouth. Considering the body posture adopted during grasping, the quadrupedal posture was preferred in most of the non-human species with lemurs using a typical crouched posture (98% of the cases). Capuchins were always in a quadrupedal posture (99.7%), whereas gorillas and chimpanzees were more diverse in their positions, including sitting (respectively, 47.4% and 34.4% of the time) and quadrupedal postures (respectively, 52.6% and 65.6% of the time). Concerning the position of the subject relative to the food location, all of the subjects preferentially placed themselves in front of the food (lemurs: 90.7%, capuchins: 61.3%, gorillas: 70.3%, chimpanzees: 44.0%) rather than at the left (lemurs: 4.0%, capuchins: 19.0%, gorillas: 17.0%, chimpanzees: 24.3%) or at the right (lemurs: 5.3%, capuchins: 19.7%, gorillas: 12.7%, chimpanzees: 31.7%). Food prehension sequences with the preferred hand and grip type (i.e., precision grip for chimpanzees, gorillas and humans; power grip for lemurs and capuchins) in a quadrupedal posture with the food located in front were retained for the kinematic study. Wrist velocity kinematics The statistics and the means of the values of the wrist velocity variables for all of the species are reported in the Tables 4 and 5. Fig. 2 shows the dimensionless curves of the wrist velocity for each species. Two groups were identified and showed no significant differences with respect to the mean and peak amplitudes of the wrist velocity: the first being humans and apes (humans, gorillas
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312 Table 3 Types of grasp used by the five species expressed in percentages.
Precision grasps Scissor grasps Whole hand grasps Mouth grasps
Humans (N ¼ 7)
Chimpanzees (N ¼ 3)
Gorillas (N ¼ 3)
Capuchins (N ¼ 3)
Lemurs (N ¼ 3)
100 _ _ _
56.66 42.33 _ 1.00
95.66 4.33 _ _
30.00 _ 65.33 4.66
_ _ 87.00 13.00
Precision grasp ¼ between the thumb and index, scissor grasp ¼ between the index and the third finger, whole hand grasp ¼ between all the fingers and the palm, mouth grasp ¼ the mouth alone. All of the non-human species were in quadrupedal postures and humans in sitting and quadrupedal postures. The score for humans is reported for both postures combined.
and chimpanzees), the second being the smaller species (capuchins and lemurs) (Table 4). Humans in both body postures displayed a similar time to peak wrist velocity as chimpanzees and gorillas (approximately 39e42%) yet an earlier peak compared with capuchins and lemurs (approximately 48%; Fig. 2 and Table 5). Consequently, apes, including humans, exhibited a longer deceleration phase. Note, however, that all species display a longer deceleration than acceleration phase. Considering the dimensionless velocities (means and peak amplitudes), capuchins and lemurs presented the highest ones compared with the non-human apes (Fig. 2 and Table 5). Consequently, they exhibited a shorter movement duration compared with apes. Humans showed the lowest velocities and exhibited the longest movement duration (Table 5). Ranges of motion of trunk and forelimb joints Concerning the general quadrupedal posture, the position of the head of lemurs near the food on the ground influenced the general posture of their trunk and forelimbs compared with the other species. The trunk and forelimb segments of lemurs were in a compact Z-like position (Fig. 3). The other species showed a more distant position of the head relative to the food, involving a more upright trunk, flexed shoulder, and reduced elbow flexion. However, a major difference appeared for the hand posture at the onset of motion, which changed according to the species. Indeed, nonhuman apes showed a knuckle walking posture, meaning that
307
they supported body weight on the back of their phalanges whereas humans, capuchins and lemurs used palmigrade postures with the palm and the finger tips in contact with the substrate (Fig. 3). The statistics and absolute values of the range of motion of the trunk and forelimb joints are presented in Tables 6 and 7. Each value represents the mean across all subjects. When comparing humans in both postures with the non-human species in a quadrupedal posture, humans showed the most significant differences independent of body postures (Table 6). Their shoulder and elbow rotation were significantly lower compared with the non-human species. Although no significant differences were detected for the other ranges of motion, humans often presented low ranges of motion especially in a quadrupedal posture (Table 7). Moreover, in a sitting posture, the human trunk flexion was as pronounced as that of non-human species and showed no significant differences. Yet, the human shoulder abduction was significantly lower than that of non-human species. In contrast, the range of trunk flexion of humans in a quadrupedal posture was significantly lower than that of the non-human species, but the shoulder abduction was not significantly different from that in non-human taxa (Tables 6 and 7). When comparing the non-human species, no significant differences in the ranges of motion with the exception of shoulder flexion (lemurs versus gorillas and chimpanzees) and elbow rotation (lemurs versus gorillas) were observed (Table 6). Indeed, the range of shoulder flexion in lemurs (65.3 ) was significantly greater than in chimpanzees (41.8 ) and gorillas (36.1 ). Only lemurs and gorillas were significantly different in terms of elbow rotation with wider ranges observed in gorillas (74.1 ) compared with lemurs (41.6 ). More generally, gorillas and lemurs presented the more extreme ranges of motion for most of the forelimb joints (Table 7). Despite the fact that the non-human species exhibited no significant differences, they did exhibit two different strategies during prehension. The ranges of rotation at the shoulder and elbow of the non-human apes (gorillas and chimpanzees) were greater compared with the smallest species (capuchins and lemurs) (Table 7). At the shoulder, the smallest species compensated by greater ranges of flexion-extension compared with the non-human apes, but at the elbow, the ranges were similar for all of the nonhuman species. When we compared humans in both postures
Table 4 Results of ANOVA’s and Tukey’s HSD post-hoc tests of the variables of wrist velocity and movement duration during prehension comparing the five species. Mean of the wrist velocity (s1)
Amplitude of the wrist velocity peak (s1)
Time to the wrist velocity peak (ms)
Time to the wrist velocity peak (%)
Duration of the deceleration phase (ms)
Duration of the deceleration phase (%)
F5,474.4 ¼ 55.5 p < 0.001***
F5,1202861 ¼ 37.88 p < 0.001***
F5,1468 ¼ 6.80 p < 0.001***
F5,3989124 ¼ 36.25 p < 0.001***
F5,1468 ¼ 6.80 p < 0.001***
F5,9527451 ¼ 47.30 p < 0.001***
0.93 NS 0.02* 0.013* 0.001*** 0.001*** 0.13 NS 0.09 NS 0.001*** 0.001*** 0.99 NS 0.001*** 0.001*** 0.001*** 0.001*** 0.41 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.22 NS 0.78 NS 0.08 NS 0.49 NS 0.93 NS
0.94 NS 0.98 NS 0.99 NS 0.007** 0.002** 0.69 NS 0.94 NS 0.001*** 0.001*** 0.99 NS 0.17 NS 0.10 NS 0.057 NS 0.03* 0.99 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.07 NS 0.17 NS 0.03* 0.07 NS 0.99 NS
0.94 NS 0.98 NS 0.99 NS 0.001*** 0.001*** 0.69 NS 0.94 NS 0.001*** 0.001*** 0.99 NS 0.17 NS 0.1 NS 0.057 NS 0.03* 0.99 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.04* 0.21 NS 0.011* 0.07 NS 0.98 NS
Movement duration (ms)
ANOVA results F5,21.51 ¼ 31.05 p < 0.001*** Results of Tukey’s HSD post-hoc tests Humans(S)ehumans(Q) 0.99 NS Humans(S)echimpanzees 0.051 NS Humans(S)egorillas 0.021* Humans(S)ecapuchins 0.001*** Humans(S)elemurs 0.001*** Humans(Q)echimpanzees 0.14 NS Humans(Q)egorillas 0.06 NS Humans(Q)ecapuchins 0.001*** Humans(Q)elemurs 0.001*** Chimpanzeesegorillas 0.99 NS Chimpanzeesecapuchins 0.001*** Chimpanzeeselemurs 0.001*** Gorillasesapajus 0.001*** Gorillaselemurs 0.001*** Capuchinselemurs 0.79 NS
All of the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures. NS: not significant; p < 0.1*; p < 0.01**; p < 0.001***.
308
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 5 Means and standard deviations (SD) of the velocity of the wrist and the movement duration in the five species during prehension. Humans (S) (N ¼ 7) Humans (Q) (N ¼ 7) Chimpanzees (N ¼ 3) Gorillas (N ¼ 3) Capuchins (N ¼ 3) Lemurs (N ¼ 3) Mean wrist velocity (mms1) Dimensionless values (s1) Amplitude of the wrist velocity peak (mms1) Dimensionless values (s1) Time to the wrist velocity peak (ms) Time to the wrist velocity peak (%) Duration of the deceleration phase (ms) Duration of the deceleration phase (%) Movement duration (ms)
486 0.83 957 1.65 439 40.7 659 59.3 1098
143 0.2 248 0.4 107 6 227 6 304
537 0.92 1125 1.94 343 39.2 537 60.8 880
112 0.2 214 0.4 76 5 133 5 182
866 1.43 1780 2.94 239 42.2 339 57.8 578
329 0.5 613 1 49 9 107 9 117
1202 1.49 2415 3.00 252 41.0 359 59.0 611
465 0.5 916 1.1 91 9 111 9 165
705 3.36 1459 6.95 173 47.8 189 52.2 362
234 1.1 506 2.4 37 2 42 2 78
652 3.05 1293 6.05 201 48.4 211 51.6 412
287 1.3 531 2.5 51 7 35 7 67
All the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures. Absolute values (mms1 and ms) dimensionless values (s1) and percentages (%) are reported.
with the non-human taxa, humans appeared more similar to lemurs and capuchins as they display lower rotation at the shoulder and elbow (Table 7). Concerning the abduction of the shoulder, humans presented the lowest values in both postures with all nonhuman species being similar and showing larger values. Concerning the contribution of the trunk in flexion relative to the ground, the lemurs exhibited the widest range of all of the nonhuman species (22.3 ) and capuchins the lowest (14.5 ). Apes, including humans, shared a similar trunk flexion (approximately 16 , Table 7), except humans in quadrupedal posture, which displayed very small ranges of flexion (3.7 ). Discussion The aim of this study was to explore how several phylogenetically distinct primate species performed prehension movements. We also wanted to test whether the similarities in motor control observed in humans and macaques in previous studies (Scott and Kalaska, 1997; Roy et al., 2000, 2002, 2006; Christel and Billard, 2002; Sartori et al., 2012) were shared by all primates. We hypothesised that the velocity profiles and the ranges of motion would differ among species due to differences in size and morphology. Our study reveals three major results: the behavioural data show similarities among the species, but also differences with the excursions of the forelimb joints and the trunk show different patterns according to the species. The velocity profiles and ranges of motion also differ across species and depend on size and/or morphology as predicted. Whereas most species preferred the quadrupedal posture, lemurs used a typical crouched posture. The grasp type also differed among species, with smaller animals (capuchins and lemurs) using whole-hand grips and larger animals (humans, gorillas, chimpanzees) using more precision grips. Larger animals had lower relative wrist velocities and spent a greater
Figure 2. Dimensionless wrist velocity curves during the entire movement for the five species. The movement is initiated at 0% and grasping occurs at 100%.
proportion of the movement in deceleration. Humans grasped food items with planar motions involving small joint rotations, more similar to the smaller animals in contrast to gorillas and chimpanzees, which used higher rotations of both the shoulder and forearm. Behaviour The behavioural data reveal that all species choose similar positions prior to reaching for and grasping food: animals position themselves in front of the food at a distance, which can be described as ‘comfortable’. Moreover, all of the subjects prefer to use a quadrupedal posture to grasp in contrast to the findings of Christel and Billard (2002), where macaques in unconstrained condition mainly used a sitting posture. This suggests that body stability may differ according to the species and influence their choice of posture. Concerning the grip types used, apes always used a grip between two fingers (precision and scissor grips) as was observed in previous studies (Christel, 1993; Pouydebat et al., 2009) even if the variability of their type of grip in this study is low compared with that quantified in previous studies. Wrist velocity kinematics All species showed a bell-shaped, single-peaked profile of wrist velocity as is observed for humans (Jeannerod, 1981, 1984; Marteniuk et al., 1987; Paulignan et al., 1997), macaques (Roy et al., 2000; Christel and Billard, 2002), cats (Alstermark et al., 1993), rats and opossums (Ivanco et al., 1996). Our results confirm that the bell-shaped profile of the wrist velocity is a basic component of the object prehension mechanism. Moreover, the wrist velocity profile in humans (Jeannerod, 1981), macaques (Roy et al., 2000) and the species in our study is asymmetric with a deceleration phase that is always longer than the acceleration phase. Differences appeared in both movement duration and in the asymmetry of the wrist velocity profile curve. Indeed, the smallest species (capuchins and lemurs) showed shorter movement durations, a later peak in the velocity profile, and consequently a shorter deceleration phase compared with larger species (chimpanzees, gorillas and humans). The later occurrence of peak velocity is likely due to the difference in preferred grip type. Indeed, capuchins and lemurs mainly grasped with their whole hand. In humans and macaques, the use of a power grip has been reported to affect the wrist kinematics and is known to extend the timing of velocity peak and to shorten the deceleration phase (Gentilucci et al., 1991; Castiello et al., 1992; Roy et al., 2002). As this grip needs less accuracy (e.g., more contact surface for the fingers) than the precision grip, it does not need to decelerate as much as the precision grip does in order to adjust the grasp (Fitts, 1954). Investigating the
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
309
Figure 3. Posture of the different species prior to prehension movements in unconstrained conditions. Note the palmigrade hand posture of humans (A, B), capuchins (E) and lemurs (F), and the knuckle walking posture of chimpanzees (C) and gorillas (D).
grasping kinematics of capuchins when using a precision grip is a question that requires further attention, yet, is essential to improve our understanding of the mechanism and evolution of prehension in primates. The latest occurrence of velocity peak and the shortest deceleration phase exhibited by lemurs is likely due to the difference in the size of the object grasped. Indeed, lemurs grasped larger food items than capuchins and the other species. In humans and macaques, reaching also depends on the intrinsic properties of the object such as its size (Bootsma et al., 1994; Roy et al., 2002). An increase of the size of the object decreases the need for an accurate grasp and induces a later occurrence of velocity peak, and a shorter deceleration phase (Fitts, 1954; Gentilucci et al., 1991; Castiello
et al., 1992). Gorillas and chimpanzees exhibited similar precision grip types and wrist velocity profiles when grasping small food items. Their peak wrist velocity occurred at the same time and they were faster than humans probably due to the pressure exerted on the subjects for access to food. These similarities are shared with macaques, suggesting that monkeys and apes respond in the same way to the same stimulus. Although we did not compare different conditions (e.g., different food sizes and manual postures) in each species, the kinematic characteristics of these conditions found within species (macaques or humans) are also present across species (chimpanzees, gorillas, capuchins and lemurs). Indeed, the kinematic
310
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 6 Results of ANOVA’s and Tukey’s HSD post-hoc tests on the ranges of motion during prehension comparing the five species. Trunk flexion
Trunk inclination
Trunk rotation
Shoulder flexion
Shoulder abduction
Shoulder -rotation
Elbow flexion
Elbow rotation
Wrist extension
Wrist deviation
ANOVA results F5,5073 ¼ 13.9 F5,410 ¼ 1.2 F5,1568 ¼ 4.7 F5,8942 ¼ 5.4 F5,3211 ¼ 6.2 F5,33769 ¼ 13.8 F5,7197 ¼ 10.2 F5,60046 ¼ 26.8 F5,51944 ¼ 32.1 F5,23706 ¼ 33.3 p < 0.001*** p < 0.3 ns p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** Results of Tukey Humans(S)e humans(Q) Humans(S)e chimpanzees Humans(S)e gorillas Humans(S)e capuchins Humans(S)e lemurs Humans(Q)e chimpanzees Humans(Q)e gorillas Humans(Q)e capuchins Humans(Q)e lemurs Chimpanzeese gorillas Chimpanzeese capuchins Chimpanzeese lemurs Gorillase capuchins Gorillase lemurs Capuchinse lemurs
HSD post-hoc tests 0.001*** 0.80 NS
0.0016**
0.47 NS
0.02*
0.99 NS
0.001***
0.95 NS
0.01**
0.99 NS
0.99 NS
0.69 NS
0.08 NS
0.09 NS
0.02*
0.001***
0.001***
0.001***
0.001***
0.001***
0.99 NS
0.99 NS
0.99 NS
0.004**
0.0012**
0.001***
0.16 NS
0.001***
0.001***
0.001***
0.96 NS
0.35 NS
0.011*
0.97 NS
0.004**
0.03*
0.7 NS
0.001***
0.001***
0.001***
0.25 NS
0.96 NS
0.87 NS
0.63 NS
0.001***
0.02*
0.64 NS
0.003**
0.001***
0.001***
0.001***
0.99 NS
0.99 NS
0.81 NS
0.96 NS
0.001***
0.99 NS
0.001***
0.001***
0.001***
0.001***
0.87 NS
0.14 NS
0.21 NS
0.56 NS
0.001***
0.09 NS
0.001***
0.001***
0.001***
0.001***
0.91 NS
0.99 NS
0.99 NS
0.78 NS
0.056 NS
0.006**
0.001***
0.001***
0.001***
0.001***
0.99 NS
0.37 NS
0.044*
0.43 NS
0.04*
0.008**
0.03*
0.001***
0.001***
0.99 NS
0.76 NS
0.48 NS
0.95 NS
0.98 NS
0.95 NS
0.44 NS
0.71 NS
0.58 NS
0.84 NS
0.99 NS
0.99 NS
0.99 NS
0.62 NS
0.99 NS
0.43 NS
0.09 NS
0.99 NS
0.99 NS
0.99 NS
0.35 NS
0.99 NS
0.76 NS
0.007**
0.95 NS
0.5 NS
0.11 NS
0.06 NS
0.99 NS
0.98 NS
0.99 NS
0.47 NS
0.18 NS
0.15 NS
0.99 NS
0.11 NS
0.96 NS
0.37 NS
0.75 NS
0.97 NS
0.39 NS
0.97 NS
0.99 NS
0.001***
0.99 NS
0.15 NS
0.98 NS
0.001***
0.37 NS
0.47 NS
0.13 NS
0.91 NS
0.39 NS
0.36 NS
0.99 NS
0.99 NS
0.99 NS
0.22 NS
0.99 NS
0.91 NS
All of the non-human species were in quadrupedal posture and humans in sitting (S) and quadrupedal (Q) postures. NS: not significant; p < 0.1*; p < 0.01**; p < 0.001***.
features of lemurs and capuchins are consistent with those of the prehension of large objects with the whole hand in humans and macaques. The kinematics of apes is consistent with those of humans and macaques grasping small objects between the thumb and index finger. Finally, this study and the previous ones revealed similarities in kinematics and grip types for macaques, gorillas, chimpanzees and humans (kinematics: Roy et al., 2000, 2002; Christel and Billard, 2002; Sartori et al., 2012; grip types: Christel, 1993; MacFarlane and Graziano, 2009; Pouydebat et al., 2009). This suggests that prehension skills have evolved early on in Old World monkeys but further analyses are needed to shed light on the development of grasping abilities in New World monkeys.
capuchins and lemurs preferred movements in a parasagittal plane (flexion-extension) to reach the food whereas gorillas and chimpanzees favour shoulder rotation. At the elbow, the same motion strategies were observed for these species, except for capuchins, which preferred rotational movements similar to those of the nonhuman apes. The two strategies seem to be correlated with the structure of the forelimb joints, especially that of the shoulder. Indeed, the overall mobility of the forelimb in primates is affected by scapular and glenohumeral orientation (Jenkins, 1973; Larson, 1993; Whitehead and Larson, 1994; Schmidt et al., 2002; Chan, 2007a, b; Schmidt and Krause, 2011). Ranges of motion of the shoulder of capuchins and lemurs are not restricted to the parasagittal plane as are those of the macaques (Christel and Billard, 2002; Jindrich et al., 2011; Schmidt and Krause, 2011). Lemurs seem to be more constrained by their shoulder morphology than capuchins, who display a glenohumeral joint that is more independent from the scapula (Schmidt and Krause, 2011). During prehension, lemurs use wider ranges of shoulder flexion than
Kinematics of trunk and forelimb joints When we compared the quadrupedal species, including humans in this posture, the forelimb joints contributed to the prehension movement following two strategies. At the shoulder, humans,
Table 7 Means and standard deviations of angular values (in degrees) of the ranges of motion during prehension in the five species. Trunk flexion Humans (S) Humans (Q) Chimpanzees Gorillas Capuchins Lemurs
16.6 3.7 16.1 16.3 14.5 22.3
8 2.1 7.8 10.6 13.5 11.0
Trunk inclination 15.6 18.1 19.3 15.3 20.6 17.6
4.5 5.2 7.5 8.8 14.5 11.7
Trunk rotation 20.2 12.5 13.4 18.7 11.7 17.4
6.5 6.7 5.2 5.2 11.7 10.8
Shoulder flexion 56.6 48.8 41.8 36.1 52.2 65.3
10.7 12.8 23.8 17.4 25.3 26.9
Shoulder abduction 12.3 19.8 22.4 25.0 23.9 25.6
6.6 10.5 9.8 11.2 15.1 8.4
Shoulder rotation 23.2 22.1 58.2 64.7 42.6 43.7
11.7 11.8 35.3 37.0 19.6 25.6
Elbow flexion 51.0 32.6 34.3 42.2 45.8 45.4
All of the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures.
10.9 9.6 16.8 12.0 14.8 9.2
Elbow rotation 17.3 21.7 63.1 74.1 58.9 41.6
6.5 10.0 29.5 34.2 35.2 18.2
Wrist extension 19.0 33.7 61.9 72.5 63.6 59.7
8.0 5.3 27.6 25.4 29.2 19.5
Wrist deviation 10.1 9.1 35.3 40.5 37.0 32.7
6.4 3.8 17.9 20.6 17.2 9.0
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
capuchins but their trunk is more inclined downward, which may force the execution of greater shoulder movements to advance the arm. However, in our data, both species presented similar ranges of motion overall with the exception of elbow rotation being higher in capuchins. In contrast, apes have the most mobile forelimbs in primates with a dorsal position of the scapula and high degree of motion of the glenohumeral joint in relation to climbing (Lewis, 1969; Tuttle, 1969; O’Connor and Rarey, 1979; Rose, 1989; Swartz, 1989; Thorpe et al., 1999; Schmidt and Krause, 2011; Zilhman et al., 2011). While humans display shoulder morphology similar to that of non-human apes (Corruccini, 1975; Chan, 2007a, b), they do not use the same joint motion strategy. Humans, which are strictly terrestrial bipeds, have a habitual use of the forelimb held below the level of the shoulder and supporting the weight of a pendant limb. Humans also show smaller muscular masses compared with other apes (Ashton and Oxnard, 1963; Thorpe et al., 1999), which could explain the lesser forelimb joint excursion (especially in abduction and rotation) during prehension compared with non-human apes. The orientation of the trunk varied according to the species, yet all species flexed the trunk in reaching except humans in a quadrupedal position. The orientation of the trunk is related to both the length of the limbs and the behaviour of the species during prehension. Indeed, the intermembral index, based on the ratio of the length of the forelimb to that of the hind limb, indicates that the closer the index is to 1 or exceeds it, the more upright the posture (Jungers, 1985; Fleagle, 1999). Apes share an index close to or superior to 1 and have longer forelimbs that orient the trunk in a more upward posture. Capuchins and lemurs have longer hind limbs than forelimbs, orienting the head more downwards. Moreover, lemurs move their head close to the food, increasing the downward posture of the trunk and much affecting the forelimb joints. Indeed, lemurs smell the food during prehension as has been previously observed for the grey mouse lemur (Reghem et al., 2011), for carnivores such as raccoons (Iwaniuk and Whishaw, 1999), and rodents such as gerbils and beavers (Whishaw et al., 1998). The ranges of motion of the trunk in lemurs contribute the most to prehension, differing from what was observed in the other species. To conclude, a variety of morphological differences related to locomotor style and environmental adaptations undoubtedly contribute to differences in kinematics. In order to better understand the evolution of manual abilities, we suggest an investigation of the influence of the mode of locomotion, diet, social context (e.g., self- and social grooming, learning), manipulation activities, and technical practice (e.g., fine-scale object manipulations, nest building, tool use, precision grip) in a diversity of primates. Only through such integrative studies will we gain true insights into the evolution of grasping in primates. Acknowledgements This research was supported by a grant from the foundation ‘ Marcel Bleustein Blanchet pour la vocation’ and by the Action Transversale du Muséum National d’Histoire Naturelle (Paris, France) ‘Formes possibles, Formes réalisées’. We are especially grateful to the director of the zoo Jardin Zoologique Tropical, Mr. Dupuyo, and to the director of the zoo La Palmyre, Mr. Caillé, and the veterinary, Mr. Petit, who permitted us to conduct this study. We also wish to thank the staff of the zoo, particularly Ronald Bosse, Redouane Lajali and Michel Carette, for their assistance during this study. We also thank A. Herrel, A. Borel, P.A. Libourel, N. Louis and J. Jacquier-Bret for their help and relevant remarks. We finally thank the reviewers who helped in the improvement of the manuscript.
311
References Alstermark, B., Lundberg, A., Pettersson, L.G., Tantisira, B., Walkowska, M., 1993. Characteristics of target-reaching in cats. I. Individual differences and intraindividual constancy. Exp. Brain Res. 94, 279e286. Ashton, E.H., Oxnard, C.E., 1963. The musculature of the primate shoulder. Trans. Zool. Soc. Lond. 29, 554e650. Bishop, A., 1964. Use of the hand in lower primates. In: Buettner-Janush, J. (Ed.), Evolutionary and Genetic Biology of Primates. Academic Press, New York, pp. 135e225. Bootsma, R.J., Marteniuk, R.G., MacKenzie, C.L., Zaal, F.T.J.M., 1994. The speedaccuracy trade-off in manual prehension: effects of movement amplitude, object size and object width on kinematic characteristics. Exp. Brain Res. 98, 535e541. Castiello, U., Bennett, K.M.B., Paulignan, Y., 1992. Does the type of prehension influence the kinematics of reaching. Behav. Brain Res. 50, 7e15. Chan, L.K., 2007a. Scapular position in primates. Folia Primatol. 78, 19e35. Chan, L.K., 2007b. Glenohumeral mobility in primates. Folia Primatol. 78, 1e18. Christel, M.I., 1993. Grasping techniques and hand preferences in apes and humans. In: Preuschoft, H., Chivers, D.J. (Eds.), Hands of Primates. Springer-Verlag, NewYork, pp. 91e108. Christel, M.I., Billard, A., 2002. Comparison between macaques’ and humans’ kinematics of prehension: the role of morphological differences and control mechanisms. Behav. Brain Res. 131, 169e184. Corruccini, R.S., 1975. Morphometric affinities in the forelimb of anthropoid primates. Z. Morph. Anthropol. 67, 19e61. Costello, M.B., Fragaszy, D.M., 1988. Prehension in Cebus and Saimiri: grip type and hand preference. Am. J. Primatol. 15, 235e245. Fitts, P.A., 1954. The information capacity of the human motor system in controlling the amplitude of movement. J. Exp. Psychol. 47, 381e391. Fleagle, J.G., 1999. Primate Adaptation and Evolution. Academic Press, San Diego. Gentilucci, M., Castiello, U., Corradini, M.L., Scarpa, M., Umilta, C., Rizzolatti, G., 1991. Influence of different types of grasping on the transport component of prehension movements. Neuropsychologia 29, 361e378. Graham, K.M., Moore, K.D., Cabel, D.W., Gribble, P.L., Cisek, P., Scott, S.H., 2003. Kinematics and kinetics of multijoint reaching in non human primates. J. Neurophysiol. 89, 2667e2677. Grosskpof, A., Kuhtz-Buschbeck, J.P., 2006. Grasping with the left and right hand: a kinematic study. Exp. Brain Res. 168, 230e240. Hof, A.L., 1996. Scaling gait data to body size. Gait Posture 4, 222e223. Hopkins, W.D., 1999. On the other hand: statistical issues in the assessment and interpretation of hand preference data in nonhuman primates. Int. J. Primatol. 20, 851e866. Ivanco, T.L., Pellis, S.M., Whishaw, I.Q., 1996. Skilled forelimb movements in prey catching and in reaching by rats (Rattus norvegicus) and opossums (Monodelphis domestica): relations to anatomical differences in motor systems. Behav. Brain Res. 79, 163e181. Iwaniuk, A.N., Whishaw, I.Q., 1999. How skilled are the skilled limb movements of the raccoon (Procyon lotor)? Behav. Brain Res. 99, 35e44. Jeannerod, M., 1981. Intersegmental coordination during reaching at natural visual objects. In: Long, J., Baddeley, A. (Eds.), Attention and Performance IX. Erlbaum, Hillsdale, pp. 153e168. Jeannerod, M., 1984. The timing of natural prehension movements. J. Mot. Behav. 16, 235e254. Jenkins Jr., F.A., 1973. The functional anatomy and evolution of the mammalian humero-ulnar articulation. Am. J. Anat. 137, 281e298. Jindrich, D.L., Courtine, G., Liu, J.J., McKay, H.L., Moseanko, R., Bernot, T.J., Roy, R.R., Zhong, H., Tuszynski, M.H., Edgerton, V.R., 2011. Unconstrained threedimensional reaching in rhesus monkeys. Exp. Brain Res. 209, 35e50. Jungers, W.L., 1985. Body size and scaling of limb proportions in primates. In: Jungers, W.L. (Ed.), Size and Scaling in Primate Biology. Plenum Publishing, New York, pp. 345e381. Larson, S.G., 1993. Functional morphology of the shoulder in primates. In: Gebo, D.L. (Ed.), Postcranial Adaptation in Nonhuman Primates. Northnern Illinois University Press, DeKalb, pp. 45e89. Lewis, O.J., 1969. The hominoid wrist joint. Am. J. Phys. Anthropol. 30, 251e268. MacFarlane, N.B., Graziano, M.S., 2009. Diversity of grip in Macaca mulatta. Exp. Brain Res. 197, 255e268. Marteniuk, R.G., MacKenzie, C.L., Jeannerod, M., Athenes, S., Dugas, C., 1987. Constraints on human arm movement trajectories. Can. J. Psychol. 41, 365e378. Napier, J.R., 1960. Studies of the hand of living primates. Proc. Zool. Soc. Lond. 134, 647e657. O’Connor, B.L., Rarey, K.E., 1979. Normal amplitudes of radioulnar pronation and supination in several genera of anthropoid primates. Am. J. Phys. Anthropol. 51, 39e44. Paulignan, Y., Frak, V.G., Toni, I., Jeannerod, M., 1997. Influence of object position and size on human prehension movements. Exp. Brain Res. 114, 226e234. Pouydebat, E., Laurin, M., Gorce, P., Bels, V., 2008. Evolution of grasping among anthropoids. J. Evol. Biol. 21, 1732e1743. Pouydebat, E., Gorce, P., Coppens, Y., Bels, V., 2009. Biomechanical study of grasping according to the volume of the object: human versus non-human primates. J. Biomech. 42, 266e272. Pouydebat, E., Reghem, E., Borel, A., Gorce, P., 2011. Diversity of grip in adults and young humans and chimpanzees (Pan troglodytes). Behav. Brain Res. 218, 21e28.
312
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
R Development Core Team, 2009. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. http://www.Rproject.org. Reghem, E., Pouydebat, E., Tia, B., Bels, V., 2011. Food prehension and manipulation in Microcebus murinus (Prosimii, Cheirogaleidae). Folia Primatol. 82, 177e188. Rose, M.D., 1989. New postcranial specimens of catarrhines from the Middle Miocene Chinji Formation, Pakistan: descriptions and a discussion of proximal humeral functional morphology in anthropoids. J. Hum. Evol. 18, 131e162. Roy, A.C., Paulignan, Y., Farnè, A., Jouffrais, C., Boussaoud, D., 2000. Hand kinematics during reaching and grasping in the macaque monkey. Behav. Brain Res. 117, 75e82. Roy, A.C., Paulignan, Y., Meunier, M., Boussaoud, D., 2002. Prehension movements in the macaque monkey: effects of object size and location. J. Neurophysiol. 88, 1491e1499. Roy, A.C., Paulignan, Y., Meunier, M., Boussaoud, D., 2006. Prehension movements in the macaque monkey: effects of perturbation of object size and location. Exp. Brain Res. 169, 182e193. Santello, M., Flanders, M., Soechting, J.F., 2002. Patterns of hand motion during grasping and the influence of sensory guidance. J. Neurosci. 22, 1426e1435. Sartori, L., Camperio Ciani, A., Bulgheroni, M., Castiello, U., 2012. Reaching and grasping behavior in Macaca fascicularis: a kinematic study. Exp. Brain Res. 224, 119e124. Schmidt, M., Krause, C., 2011. Scapula movements and their contribution to threedimensional forelimb excursions in quadrupedal primates. In: D’Août, K., Vereecke, E.E. (Eds.), Primate Locomotion: Linking Field and Laboratory Research. Springer Publishing, New York, pp. 83e108. Schmidt, M., Voges, D., Fischer, M.S., 2002. Shoulder movements during quadrupedal locomotion in arboreal primates. Z. Morph. Anthropol. 83, 235e242.
Scott, S.H., Kalaska, J.F., 1997. Reaching movements with similar paths but different arm orientations. I. Activity of individual cells in motor cortex. Am. Physiol. Soc. 77, 826e852. Spinozzi, G., Truppa, V., Lagana, T., 2004. Grasping behavior in tufted capuchin monkeys (Cebus apella): grip types and manual laterality for picking up a small food item. Am. J. Phys. Anthropol. 125, 30e41. Swartz, S.M., 1989. The functional morphology of weight bearing: limb joint surface area allometry in anthropoid primates. J. Zool. 218, 441e460. Thorpe, S.K.S., Crompton, R.H., Günther, M.M., Ker, R.F., McNeil, A.R., 1999. Dimensions and moment arms of the hind-and forelimb muscles of common chimpanzees (Pan troglodytes). Am. J. Phys. Anthropol. 110, 179e199. Tuttle, R.H., 1969. Quantitative and functional studies on the hands of Anthropoidea. I. The Hominoidea. J. Morphol. 128, 309e364. Vereecke, E., D’Août, K., Aerts, P., 2006. Speed modulation in hylobatid bipedalism: a kinematic analysis. J. Hum. Evol. 51, 513e526. Whishaw, I.Q., Sarna, J.R., Pellis, S.M., 1998. Evidence for rodent-common and species-typical limb and digit use in eating, derived from a comparative analysis of ten rodent species. Behav. Brain Res. 96, 79e91. Whitehead, P.F., Larson, S.G., 1994. Shoulder motion during quadrupedal walking in Cercopithecus aethiops: integration of cineradiographic and electromyographic data. J. Hum. Evol. 26, 525e544. Wu, G., van der Helm, F.C., Veeger, H.E., Makhsous, M., Van Roy, P., Anglin, C., Nagels, J., Karduna, A.R., McQuade, K., Wang, X., Werner, F.W., Buchholz, B., 2005. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion-Part II: shoulder, elbow, wrist and hand. J. Biomech. 38, 981e992. Zilhman, A.L., McFarland, R.K., Underwood, C.E., 2011. Functional anatomy and adaptation of gorillas (Gorilla gorilla gorilla) with comparisons to male orangutans (Pongo pygmaeus). Anat. Rec. 294, 1842e1855.
Contents lists available at ScienceDirect
Journal of Human Evolution journal homepage: www.elsevier.com/locate/jhevol
Unconstrained 3D-kinematics of prehension in five primates: Lemur, capuchin, gorilla, chimpanzee, human Elodie Reghem a, b, Laurence Chèze c, Yves Coppens d, Emmanuelle Pouydebat a, * a
UMR 7179, Département Ecologie et Gestion de la Biodiversité, MNHN, 55 rue Buffon, case postale 55, FR-75231 Paris, France EA 4322 Handibio, Université du Sud Toulon Var, La Garde, France c IFSTTAR, LBMC, UMR_T9406, Université Lyon 1, 43 Boulevard du 11 novembre, 69622 Villeurbanne, France d Chaire de Paléoanthropologie et Préhistoire, Collège de France, 11 Place Marcelin Berthelot, 75005 Paris, France b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 February 2013 Accepted 26 June 2013 Available online 29 July 2013
Primates are known for their use of the hand in many activities including food grasping. Yet, most studies concentrate on the type of grip used. Moreover, kinematic studies remain limited to a few investigations of the distal elements in constrained conditions in humans and macaques. In order to improve our understanding of the prehension movement in primates, we analyse here the behavioural strategies (e.g., types of grip, body postures) as well as the 3D kinematics of the whole forelimb and the trunk during the prehension of small static food items in five primate species in unconstrained conditions. All species preferred the quadrupedal posture except lemurs, which used a typical crouched posture. Grasp type differed among species, with smaller animals (capuchins and lemurs) using a whole-hand grip and larger animals (humans, gorillas, chimpanzees) using predominantly a precision grip. Larger animals had lower relative wrist velocities and spent a larger proportion of the movement decelerating. Humans grasped food items with planar motions involving small joint rotations, more similar to the smaller animals than to gorillas and chimpanzees, which used greater rotations of both the shoulder and forearm. In conclusion, the features characterising human food prehension are present in other primates, yet differences exist in joint motions. These results provide a good basis to suggest hypotheses concerning the factors involved in driving the evolution of grasping abilities in primates. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Grasping evolution Kinematics Forelimb
Introduction All of the members of the Order Primates share the characteristic ability to grasp food or objects with one or both hands (Napier, 1960; Bishop, 1964; Costello and Fragaszy, 1988; Christel, 1993; MacFarlane and Graziano, 2009; Pouydebat et al., 2009; Reghem et al., 2011). This ability is a key behaviour implied in daily activities in both humans and non-human primates and its evolution remains unresolved (Pouydebat et al., 2008). Although all primates are able to grasp, different food prehension strategies have been identified with respect to both manual grip postures (Bishop, 1964; Christel, 1993; Spinozzi et al., 2004; Pouydebat et al., 2011) and the kinematics of the forelimb (Scott and Kalaska, 1997; Christel and Billard, 2002). In kinematic studies, prehension is often described as consisting of two phases. The first phase is the transport phase
* Corresponding author. E-mail address: [email protected] (E. Pouydebat). 0047-2484/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jhevol.2013.06.011
where the hand is transported to the object to be grasped. The second one is the grasping phase where the grip aperture between thumb and index opens and ultimately closes around the object (Jeannerod, 1981, 1984). The kinematics of prehension in primates have been investigated only in humans and macaques. In addition, most studies focus on the kinematical similarities; that is to say the kinematic parameters common between humans and macaques (Scott and Kalaska, 1997; Roy et al., 2000, 2002, 2006; Sartori et al., 2012). Both species show a similar bell-shaped pattern of the wrist velocity with a longer deceleration than acceleration phase, and a grip aperture varying according to the object size. However, most of these results were obtained for macaques in constrained conditions (e.g., intensive learning of the movement, grasping through a slot, parts of the body constrained, the head immobilised) that induce stereotyped movements, a bias in elbow motion, and which restrict shoulder motion despite the importance of the proximal joints in forelimb movement. In part because of these constrained conditions, many studies have focused only upon the distal components, i.e., the wrist kinematics and grip aperture. One study (Christel and
304
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Billard, 2002) investigated untrained macaques in unconstrained condition (i.e., in their group within their habitual enclosure) and quantified the ranges of motion of the trunk and the forelimb joints in comparison with humans. Despite the fact that similar results were obtained in comparison with prior studies on wrist velocity and grip aperture, Christel and Billard (2002) emphasised differences in kinematic strategies of both species. Indeed, macaques were faster than humans and showed more variability. Moreover, humans and macaques performed opposite movements. During the reaching phase, macaques adopted a larger elbow, wrist and trunk motion and a smaller shoulder motion, contrary to what is observed in humans. Therefore, the trunk of macaques contributes in a greater proportion to the prehension movement, compensating for the lower excursion at the shoulder. The authors suggested that variation in motor control, posture, and morphology could explain the differences. Indeed, macaques use their forelimb in locomotion, which may imply a different shoulder joint morphology, muscular strength, body posture, and a change in the motor control driving the observed differences (Christel and Billard, 2002). Whether these kinematic parameters are specific to particular primates or general features of non-human primates remains, however, unclear. Only very few kinematic studies have been conducted under unconstrained conditions involving spontaneous and free gestures. These conditions appear, however, fundamental to understanding both the source of variation in movement and the prehension mechanism. Thus, to address questions pertaining to the evolution of prehension and the human specificities such studies are needed. In this context, the following research questions are addressed in the present study: How do several primate species of different phylogenetically distinct taxa perform their prehension movement? Are the similarities in motor control observed in humans and macaques shared by all non-human primates? Here, we hypothesise that the velocity profiles and the range of motion will differ across the species, depending of their size and morphology. To address these questions, we examine here unconstrained prehension in five species representative of different groups of primates: the ring-tailed lemur (Lemur catta, prosimian), the yellow-breasted capuchin (Sapajus xanthosternos, New World monkey), gorillas (Gorilla gorilla), chimpanzees (Pan troglodytes) and humans (Homo sapiens). We quantify the 3D-kinematics of the reaching and the type of grip used during a prehension task in unconstrained conditions. Reaching is described by the ranges of motion of the forelimb joints and the trunk, and by the wrist velocity profile. Grasping is here described by the grip postures and not by the movement of the fingers. The behavioural strategies during prehension are quantified and include body posture, the position of the subjects relative to the food, and their distance to the food. Material and methods Species Seven adults humans (two women and five men, mean age: 33.0 7.3 years) were investigated. The study was carried out in accordance with ‘The ethical codes of the World Medical Association’ (Declaration of Helsinki). For the non-human primates, the experimental protocol used adhered to the legal requirements of the European Union and the American Association of Physical Anthropologists Code of Ethics. Three untrained adults were recorded for each species. Chimpanzees (P. troglodytes, three females, mean age: 19.7 3.8 years), gorillas (Gorilla gorilla, two females and one male, mean age: 17.6 4.1 years), and capuchins (Sapajus xanthosternos, two females and one male, mean age: 7.6 5.6 years)
were filmed in the zoological park of La Palmyre, France. Lemurs (Lemur catta, three males, mean age: 3.0 years) were filmed in the zoological park of Jardin Zoologique Tropical, France. Food selection and food size We selected raisins as our target food because all subjects consume fruit as part of their diet. The size of the food was standardised according to the length of the hand calculated based on the landmarks of the wrist and those on the metacarpal heads. The size of the raisin was one-quarter of this length for each species (Table 1). The lemurs were different when grasping raisins compared with the other species as they systematically grasped the raisins with the mouth. The size of food for which the lemurs systematically used their hand corresponded to a small apple morsel. We chose to keep the small food item as a basis of our study because it was comparable with the only study on macaques recorded in unconstrained conditions, thus allowing our data to be compared with that available for macaques (Christel and Billard, 2002). Furthermore, the small food size was chosen to test the type of grip (power or precision) used in different species. Indeed, not all of the non-human primates are able to use precision grips to grasp small food items (e.g., lemurs) and small food sizes can reveal more behavioural versatility (many types of precision grips) than large food sizes (mainly involving a power grip). Experimental set up for humans Reaching and grasping movements were recorded at 100 frames/second using a Motion AnalysisÒ system (LBMC, Lyon; Motion Analysis Corporation, Santa Rosa, USA) with eight EagleÒ cameras surrounding the subject. Subjects received no particular instruction except to reach, grasp, and move the raisin to the mouth. When in a sitting posture, participants sat on a stool with the hands rested on the table. In the quadrupedal posture, body weight was supported by the knees and the palms of the hands. Raisins were scattered on a surface of 50 cm2 in front of the subject. Twelve infrared reflective markers were positioned on anatomical landmarks of the forelimb and the trunk (Fig. 1A) as recommended by the International Society of Biomechanics (ISB) (Wu et al., 2005). Experimental set up for non-human primates Data were acquired using five video cameras (SanyoÒ Full HD) at 60 frames/second. The food was scattered on the ground on a surface of 50 50 cm. All of the non-human primates were in unconstrained conditions, meaning that they were not trained prior the recordings. They were in their habitual enclosure, in a group, and totally free to choose their body posture and their distance relative to the food. As no markers could be placed on the animals in zoos for reasons of security (and the animals would have taken off any external markers immediately), ten landmarks were manually digitised at the same body location as those in humans (see kinematic analysis for more details).
Table 1 Food size according to the species. Species Humans Chimpanzees Gorillas Capuchins Lemurs
Small food size (cm)
Large food size (cm3)
1.5 1.7 1.7 0.5 0.5
_ _ _ _ 2
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
305
Figure 1. Position of the landmarks (A) and the segment coordinate systems (B) on the forelimb. Two technical markers placed on the arm and one on the forearm helped to reconstruct the motion of forelimb in humans. For the five species (humans, chimpanzees, gorillas, capuchins, lemurs), only the trunk landmarks [1] and [2] were used for reasons of visibility. A segment coordinate system (SCS) was computed for each segment (trunk, arm, forearm) using a minimum of three landmarks per segment. Then, the orthogonal rotation matrix from proximal SCS to distal SCS was computed for each joint and interpreted so that the first rotation (flexion/extension) was around the z axis embedded on the proximal SCS, the last rotation (internal/external rotation) was around the y axis embedded on the distal SCS and the abduction/adduction took place around the x floating axis. Trunk angles were obtained using ground and trunk SCS, shoulder angles using trunk and arm SCS, and elbow angles using arm and forearm SCS. Landmark legend (A): [1] processus spinosus of the seventh cervical vertebra, [2] eighth thoracic vertebra, [3] suprasternal notch, [4] xiphoid process, [5, 6] left and right dorsal point on the acromioclavicular joint, [7, 8] medial and lateral epicondyles of humerus, [9, 10] radial and ulnar styloid processes, [11, 12] head of second and fifth metacarpus. Segment coordinate system legend (B): G: Ground, T: Trunk, A: Arm, F: Forearm.
Behavioural analysis of non-human primates Four behavioural variables were collected for 100 grasps per subject (total ¼ 1200 grasps) from video recordings (sessions of 20 minutes/day): body posture (sitting, squatting, quadrupedal), the position of the subject relative to the food (in front, at left or at right), hand preference, and preferred grip type. Precision grips are defined here as a grasp between the thumb and index tips on the lateral, pulp or medial side of the distal part of the fingers (Christel, 1993; Pouydebat et al., 2009). The power grips describe a grasp with the palm and all of the fingers, and the scissor grip describes a grasp between the index and the third finger. As the existence of kinematic differences between the use of the dominant and nondominant hand remains debated in the literature (Grosskpof and Kuhtz-Buschbeck, 2006), we selected the dominant hand of each subject in our kinematic analysis. Hand preferences were considered for subjects who had a significant z-score (Table 2). As only one chimpanzee subject did not show a dominant hand, we arbitrarily analysed its right hand. These behavioural variables helped us to select comparable trials among all of the species for the movement analysis. The preferred grip type and dominant hand used by each subject were retained for subsequent kinematic analysis. Kinematic analysis Kinematic data was analysed for five trials for each subject (15 grasp sequences per species for the non-humans and 35 grasp sequences for humans). First, as the non-humans preferentially grasped food in front of them (see results), we retained prehension movements in front of the subjects for all of the species to standardise the comparison. Second, we selected comparable food distances by calculating the ratio of each species’ forelimb length (arm þ forearm) to the food distance. The most used food distances were retained in which the food was no further than an arm’s
length away. Third, for all of the sequences selected, the hand was on the ground in a stationary position before being moved forward to reach. For human subjects, the software EvaRTÒ 5.0 reconstructed the marker trajectories and extracted their coordinates in 3D. For nonhumans, video calibration, manual digitisation (landmarks were manually digitised frame by frame) of the anatomical landmarks on each 2D video frame and the reconstruction of the 3D coordinates were performed using a custom-written MatlabÒ routine (Loco 3.3). For all of the species in our sample, kinematic data were processed using MatlabÒ (The MathWorks, Inc., Natick, Massachusetts). Data were low-pass filtered at a frequency of 6 Hz with a second order dual-pass Butterworth filter. Movement onset of the grasp (reaching phase) was identified based on the wrist velocity profile and was defined as the time when wrist velocity reached 5% of its peak (Alstermark et al., 1993; Santello et al., 2002; Graham et al., 2003). All variables were obtained in a trunk-centered frame of reference. The X axis was oriented toward the food, the Y axis was vertical and the Z axis was directed laterally (Fig. 1B). We quantified several variables for the wrist kinematics analysis. The wrist velocity was calculated from the filtered three dimensional Cartesian coordinates (x, y, z) of the mid-point of the two wrist landmarks. The wrist velocities were converted to dimensionless units to reduce the effect of the size of the species and allow inter-specific comparisons (Hof, 1996; Vereecke et al., 2006). Dimensionless velocities were obtained from velocities divided by the mean of the arm plus forearm length of all of the subjects for each species. The mean velocity, the maximal amplitude, and the time to the velocity peak were measured. The mean velocity and the maximal amplitude are reported both in absolute values and in dimensionless values. The dimensionless wrist velocity curves were then interpolated for each species allowing resampling on 100 points in order to superimpose the curves, to compute the mean patterns and represent them relative to the
306
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 2 List of the subjects of all of the species and scores of their manual preference (right-handers ¼ R, left-handers ¼ L or ambidextrous ¼ A). Species
Subjects
Sex
Age in years
Forelimb length (cm)
Mass (kg)
HI
Z-score
Manual preference
Humans
1 2 3 4 5 6 7 1 2 3 1 2 3 1 2 3 1 2 3
Male Female Female Male Male Female Male Female Female Female Male Female Female Male Female Female Male Male Male
28 29 38 22 33 44 37 18 17 24 21 13 19 14 6 3 3 3 3
57.5 58.0 59.0 57.0 57.0 57.0 60.0 65.4 48.3 67.6 87.5 72.1 81.5 24.8 19.1 18.3 22.9 20.7 20.4
68.0 57.0 73.0 62.0 62.0 53.0 72.0 44.0 35.0 55.0 150.0 145.5 65.5 3.3 3.1 2.0 2.3 2.5 3.0
1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.07 0.20 0.06 0.46 0.20 0.60 0.24 0.24 0.34 0.93 1.00 1.00
10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 10.1 1.96*** 0.7 NS 2.7 1.96** 0.6 NS 4.6 1.96** 2.0 1.96* 6.0 1.96** 2.4 1.96* 2.4 1.96* 3.1 1.96** 8.9 1.96** 8.5 1.96** 9.4 1.96**
R R R R R R R A R A R L L L L L L R L
Chimpanzees
Gorillas
Capuchins
Lemurs
All of the non-human species were in quadrupedal postures and humans in sitting and quadrupedal postures. The score of humans was the same in both postures. NS: not significant; *: significant at 0.05; **: significant at 0.01; ***: significant at 0.001.
overall movement duration (100%). The movement duration was the time measured from wrist movement onset to the actual grasp. Next, the angles of the trunk (relative to the ground) and the forelimb joints (shoulder, elbow) were computed using the ISB recommendations (Wu et al., 2005) (Fig. 1B). Eight degrees of freedom were quantified: (a) trunk flexion-extension (forward/ backward motion in sitting; downward/upward motion in quadrupedal), (b) trunk inclination (lateral motion of the trunk on the left or right side), (c) trunk rotation, (d) shoulder flexionextension, (e) shoulder abductioneadduction, (f) shoulder rotation, (g) elbow flexion-extension, (h) elbow rotation (i.e., pronation and supination). The ranges of motion were calculated from the absolute differences between minimal and maximal angles for the trunk, the shoulder, and the elbow joints. The curves of the forelimb joints and trunk motion were interpolated over 100 points and averaged for the subjects of each species to represent the patterns relative to the overall movement duration. Statistical analyses All analyses were performed using the R graphical and statistical package v.2.9.0 (R Development Core Team, 2009). The degree of manual asymmetry was calculated for each subject via the handedness index (HI) using the formula (R-L)/(R þ L), where R and L represent the total number of right and left responses (Hopkins, 1999). The HI values ranging from þ1.0 to 1.0 indicate a right hand preference for positive values and a left-hand preference for negative values (Hopkins, 1999). Then, the binomial zscore determined if the manual preference is significant. The subjects were classified with z score, z 1.96 as right-handed, z 1.96 as left-handed, and 1.96 < z < 1.96 as ambipreferent (Hopkins, 1999). Analyses of variance (ANOVA) with one factor (species) and Tukey’s HSD post-hoc tests were used to test for differences in kinematics among species. The absolute values of angular displacements, the time to wrist velocity peak (in % and ms), the duration of the deceleration phase (in % and ms), the movement duration, and the dimensionless values of wrist velocity were tested. A Shapiroe Wilks’ test was used to test the normality of the data and the Bartlett-test was used to check for homogeneity of variance.
Results Behaviour All species showed a hand preference except for two chimpanzees (Table 2). Moreover, all of the species, except lemurs, used precision grips and humans used only this type of grip. All other species used two or three different grip types (Table 3). Gorillas used few scissor grips compared with chimpanzees, which used both in the same proportion. Capuchins and lemurs were the only ones to use power grips. Capuchins mainly used the power grip compared with precision and mouth grips. Lemurs mainly used the power grip and showed the greatest use of the mouth. Considering the body posture adopted during grasping, the quadrupedal posture was preferred in most of the non-human species with lemurs using a typical crouched posture (98% of the cases). Capuchins were always in a quadrupedal posture (99.7%), whereas gorillas and chimpanzees were more diverse in their positions, including sitting (respectively, 47.4% and 34.4% of the time) and quadrupedal postures (respectively, 52.6% and 65.6% of the time). Concerning the position of the subject relative to the food location, all of the subjects preferentially placed themselves in front of the food (lemurs: 90.7%, capuchins: 61.3%, gorillas: 70.3%, chimpanzees: 44.0%) rather than at the left (lemurs: 4.0%, capuchins: 19.0%, gorillas: 17.0%, chimpanzees: 24.3%) or at the right (lemurs: 5.3%, capuchins: 19.7%, gorillas: 12.7%, chimpanzees: 31.7%). Food prehension sequences with the preferred hand and grip type (i.e., precision grip for chimpanzees, gorillas and humans; power grip for lemurs and capuchins) in a quadrupedal posture with the food located in front were retained for the kinematic study. Wrist velocity kinematics The statistics and the means of the values of the wrist velocity variables for all of the species are reported in the Tables 4 and 5. Fig. 2 shows the dimensionless curves of the wrist velocity for each species. Two groups were identified and showed no significant differences with respect to the mean and peak amplitudes of the wrist velocity: the first being humans and apes (humans, gorillas
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312 Table 3 Types of grasp used by the five species expressed in percentages.
Precision grasps Scissor grasps Whole hand grasps Mouth grasps
Humans (N ¼ 7)
Chimpanzees (N ¼ 3)
Gorillas (N ¼ 3)
Capuchins (N ¼ 3)
Lemurs (N ¼ 3)
100 _ _ _
56.66 42.33 _ 1.00
95.66 4.33 _ _
30.00 _ 65.33 4.66
_ _ 87.00 13.00
Precision grasp ¼ between the thumb and index, scissor grasp ¼ between the index and the third finger, whole hand grasp ¼ between all the fingers and the palm, mouth grasp ¼ the mouth alone. All of the non-human species were in quadrupedal postures and humans in sitting and quadrupedal postures. The score for humans is reported for both postures combined.
and chimpanzees), the second being the smaller species (capuchins and lemurs) (Table 4). Humans in both body postures displayed a similar time to peak wrist velocity as chimpanzees and gorillas (approximately 39e42%) yet an earlier peak compared with capuchins and lemurs (approximately 48%; Fig. 2 and Table 5). Consequently, apes, including humans, exhibited a longer deceleration phase. Note, however, that all species display a longer deceleration than acceleration phase. Considering the dimensionless velocities (means and peak amplitudes), capuchins and lemurs presented the highest ones compared with the non-human apes (Fig. 2 and Table 5). Consequently, they exhibited a shorter movement duration compared with apes. Humans showed the lowest velocities and exhibited the longest movement duration (Table 5). Ranges of motion of trunk and forelimb joints Concerning the general quadrupedal posture, the position of the head of lemurs near the food on the ground influenced the general posture of their trunk and forelimbs compared with the other species. The trunk and forelimb segments of lemurs were in a compact Z-like position (Fig. 3). The other species showed a more distant position of the head relative to the food, involving a more upright trunk, flexed shoulder, and reduced elbow flexion. However, a major difference appeared for the hand posture at the onset of motion, which changed according to the species. Indeed, nonhuman apes showed a knuckle walking posture, meaning that
307
they supported body weight on the back of their phalanges whereas humans, capuchins and lemurs used palmigrade postures with the palm and the finger tips in contact with the substrate (Fig. 3). The statistics and absolute values of the range of motion of the trunk and forelimb joints are presented in Tables 6 and 7. Each value represents the mean across all subjects. When comparing humans in both postures with the non-human species in a quadrupedal posture, humans showed the most significant differences independent of body postures (Table 6). Their shoulder and elbow rotation were significantly lower compared with the non-human species. Although no significant differences were detected for the other ranges of motion, humans often presented low ranges of motion especially in a quadrupedal posture (Table 7). Moreover, in a sitting posture, the human trunk flexion was as pronounced as that of non-human species and showed no significant differences. Yet, the human shoulder abduction was significantly lower than that of non-human species. In contrast, the range of trunk flexion of humans in a quadrupedal posture was significantly lower than that of the non-human species, but the shoulder abduction was not significantly different from that in non-human taxa (Tables 6 and 7). When comparing the non-human species, no significant differences in the ranges of motion with the exception of shoulder flexion (lemurs versus gorillas and chimpanzees) and elbow rotation (lemurs versus gorillas) were observed (Table 6). Indeed, the range of shoulder flexion in lemurs (65.3 ) was significantly greater than in chimpanzees (41.8 ) and gorillas (36.1 ). Only lemurs and gorillas were significantly different in terms of elbow rotation with wider ranges observed in gorillas (74.1 ) compared with lemurs (41.6 ). More generally, gorillas and lemurs presented the more extreme ranges of motion for most of the forelimb joints (Table 7). Despite the fact that the non-human species exhibited no significant differences, they did exhibit two different strategies during prehension. The ranges of rotation at the shoulder and elbow of the non-human apes (gorillas and chimpanzees) were greater compared with the smallest species (capuchins and lemurs) (Table 7). At the shoulder, the smallest species compensated by greater ranges of flexion-extension compared with the non-human apes, but at the elbow, the ranges were similar for all of the nonhuman species. When we compared humans in both postures
Table 4 Results of ANOVA’s and Tukey’s HSD post-hoc tests of the variables of wrist velocity and movement duration during prehension comparing the five species. Mean of the wrist velocity (s1)
Amplitude of the wrist velocity peak (s1)
Time to the wrist velocity peak (ms)
Time to the wrist velocity peak (%)
Duration of the deceleration phase (ms)
Duration of the deceleration phase (%)
F5,474.4 ¼ 55.5 p < 0.001***
F5,1202861 ¼ 37.88 p < 0.001***
F5,1468 ¼ 6.80 p < 0.001***
F5,3989124 ¼ 36.25 p < 0.001***
F5,1468 ¼ 6.80 p < 0.001***
F5,9527451 ¼ 47.30 p < 0.001***
0.93 NS 0.02* 0.013* 0.001*** 0.001*** 0.13 NS 0.09 NS 0.001*** 0.001*** 0.99 NS 0.001*** 0.001*** 0.001*** 0.001*** 0.41 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.22 NS 0.78 NS 0.08 NS 0.49 NS 0.93 NS
0.94 NS 0.98 NS 0.99 NS 0.007** 0.002** 0.69 NS 0.94 NS 0.001*** 0.001*** 0.99 NS 0.17 NS 0.10 NS 0.057 NS 0.03* 0.99 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.07 NS 0.17 NS 0.03* 0.07 NS 0.99 NS
0.94 NS 0.98 NS 0.99 NS 0.001*** 0.001*** 0.69 NS 0.94 NS 0.001*** 0.001*** 0.99 NS 0.17 NS 0.1 NS 0.057 NS 0.03* 0.99 NS
0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.99 NS 0.04* 0.21 NS 0.011* 0.07 NS 0.98 NS
Movement duration (ms)
ANOVA results F5,21.51 ¼ 31.05 p < 0.001*** Results of Tukey’s HSD post-hoc tests Humans(S)ehumans(Q) 0.99 NS Humans(S)echimpanzees 0.051 NS Humans(S)egorillas 0.021* Humans(S)ecapuchins 0.001*** Humans(S)elemurs 0.001*** Humans(Q)echimpanzees 0.14 NS Humans(Q)egorillas 0.06 NS Humans(Q)ecapuchins 0.001*** Humans(Q)elemurs 0.001*** Chimpanzeesegorillas 0.99 NS Chimpanzeesecapuchins 0.001*** Chimpanzeeselemurs 0.001*** Gorillasesapajus 0.001*** Gorillaselemurs 0.001*** Capuchinselemurs 0.79 NS
All of the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures. NS: not significant; p < 0.1*; p < 0.01**; p < 0.001***.
308
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 5 Means and standard deviations (SD) of the velocity of the wrist and the movement duration in the five species during prehension. Humans (S) (N ¼ 7) Humans (Q) (N ¼ 7) Chimpanzees (N ¼ 3) Gorillas (N ¼ 3) Capuchins (N ¼ 3) Lemurs (N ¼ 3) Mean wrist velocity (mms1) Dimensionless values (s1) Amplitude of the wrist velocity peak (mms1) Dimensionless values (s1) Time to the wrist velocity peak (ms) Time to the wrist velocity peak (%) Duration of the deceleration phase (ms) Duration of the deceleration phase (%) Movement duration (ms)
486 0.83 957 1.65 439 40.7 659 59.3 1098
143 0.2 248 0.4 107 6 227 6 304
537 0.92 1125 1.94 343 39.2 537 60.8 880
112 0.2 214 0.4 76 5 133 5 182
866 1.43 1780 2.94 239 42.2 339 57.8 578
329 0.5 613 1 49 9 107 9 117
1202 1.49 2415 3.00 252 41.0 359 59.0 611
465 0.5 916 1.1 91 9 111 9 165
705 3.36 1459 6.95 173 47.8 189 52.2 362
234 1.1 506 2.4 37 2 42 2 78
652 3.05 1293 6.05 201 48.4 211 51.6 412
287 1.3 531 2.5 51 7 35 7 67
All the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures. Absolute values (mms1 and ms) dimensionless values (s1) and percentages (%) are reported.
with the non-human taxa, humans appeared more similar to lemurs and capuchins as they display lower rotation at the shoulder and elbow (Table 7). Concerning the abduction of the shoulder, humans presented the lowest values in both postures with all nonhuman species being similar and showing larger values. Concerning the contribution of the trunk in flexion relative to the ground, the lemurs exhibited the widest range of all of the nonhuman species (22.3 ) and capuchins the lowest (14.5 ). Apes, including humans, shared a similar trunk flexion (approximately 16 , Table 7), except humans in quadrupedal posture, which displayed very small ranges of flexion (3.7 ). Discussion The aim of this study was to explore how several phylogenetically distinct primate species performed prehension movements. We also wanted to test whether the similarities in motor control observed in humans and macaques in previous studies (Scott and Kalaska, 1997; Roy et al., 2000, 2002, 2006; Christel and Billard, 2002; Sartori et al., 2012) were shared by all primates. We hypothesised that the velocity profiles and the ranges of motion would differ among species due to differences in size and morphology. Our study reveals three major results: the behavioural data show similarities among the species, but also differences with the excursions of the forelimb joints and the trunk show different patterns according to the species. The velocity profiles and ranges of motion also differ across species and depend on size and/or morphology as predicted. Whereas most species preferred the quadrupedal posture, lemurs used a typical crouched posture. The grasp type also differed among species, with smaller animals (capuchins and lemurs) using whole-hand grips and larger animals (humans, gorillas, chimpanzees) using more precision grips. Larger animals had lower relative wrist velocities and spent a greater
Figure 2. Dimensionless wrist velocity curves during the entire movement for the five species. The movement is initiated at 0% and grasping occurs at 100%.
proportion of the movement in deceleration. Humans grasped food items with planar motions involving small joint rotations, more similar to the smaller animals in contrast to gorillas and chimpanzees, which used higher rotations of both the shoulder and forearm. Behaviour The behavioural data reveal that all species choose similar positions prior to reaching for and grasping food: animals position themselves in front of the food at a distance, which can be described as ‘comfortable’. Moreover, all of the subjects prefer to use a quadrupedal posture to grasp in contrast to the findings of Christel and Billard (2002), where macaques in unconstrained condition mainly used a sitting posture. This suggests that body stability may differ according to the species and influence their choice of posture. Concerning the grip types used, apes always used a grip between two fingers (precision and scissor grips) as was observed in previous studies (Christel, 1993; Pouydebat et al., 2009) even if the variability of their type of grip in this study is low compared with that quantified in previous studies. Wrist velocity kinematics All species showed a bell-shaped, single-peaked profile of wrist velocity as is observed for humans (Jeannerod, 1981, 1984; Marteniuk et al., 1987; Paulignan et al., 1997), macaques (Roy et al., 2000; Christel and Billard, 2002), cats (Alstermark et al., 1993), rats and opossums (Ivanco et al., 1996). Our results confirm that the bell-shaped profile of the wrist velocity is a basic component of the object prehension mechanism. Moreover, the wrist velocity profile in humans (Jeannerod, 1981), macaques (Roy et al., 2000) and the species in our study is asymmetric with a deceleration phase that is always longer than the acceleration phase. Differences appeared in both movement duration and in the asymmetry of the wrist velocity profile curve. Indeed, the smallest species (capuchins and lemurs) showed shorter movement durations, a later peak in the velocity profile, and consequently a shorter deceleration phase compared with larger species (chimpanzees, gorillas and humans). The later occurrence of peak velocity is likely due to the difference in preferred grip type. Indeed, capuchins and lemurs mainly grasped with their whole hand. In humans and macaques, the use of a power grip has been reported to affect the wrist kinematics and is known to extend the timing of velocity peak and to shorten the deceleration phase (Gentilucci et al., 1991; Castiello et al., 1992; Roy et al., 2002). As this grip needs less accuracy (e.g., more contact surface for the fingers) than the precision grip, it does not need to decelerate as much as the precision grip does in order to adjust the grasp (Fitts, 1954). Investigating the
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
309
Figure 3. Posture of the different species prior to prehension movements in unconstrained conditions. Note the palmigrade hand posture of humans (A, B), capuchins (E) and lemurs (F), and the knuckle walking posture of chimpanzees (C) and gorillas (D).
grasping kinematics of capuchins when using a precision grip is a question that requires further attention, yet, is essential to improve our understanding of the mechanism and evolution of prehension in primates. The latest occurrence of velocity peak and the shortest deceleration phase exhibited by lemurs is likely due to the difference in the size of the object grasped. Indeed, lemurs grasped larger food items than capuchins and the other species. In humans and macaques, reaching also depends on the intrinsic properties of the object such as its size (Bootsma et al., 1994; Roy et al., 2002). An increase of the size of the object decreases the need for an accurate grasp and induces a later occurrence of velocity peak, and a shorter deceleration phase (Fitts, 1954; Gentilucci et al., 1991; Castiello
et al., 1992). Gorillas and chimpanzees exhibited similar precision grip types and wrist velocity profiles when grasping small food items. Their peak wrist velocity occurred at the same time and they were faster than humans probably due to the pressure exerted on the subjects for access to food. These similarities are shared with macaques, suggesting that monkeys and apes respond in the same way to the same stimulus. Although we did not compare different conditions (e.g., different food sizes and manual postures) in each species, the kinematic characteristics of these conditions found within species (macaques or humans) are also present across species (chimpanzees, gorillas, capuchins and lemurs). Indeed, the kinematic
310
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
Table 6 Results of ANOVA’s and Tukey’s HSD post-hoc tests on the ranges of motion during prehension comparing the five species. Trunk flexion
Trunk inclination
Trunk rotation
Shoulder flexion
Shoulder abduction
Shoulder -rotation
Elbow flexion
Elbow rotation
Wrist extension
Wrist deviation
ANOVA results F5,5073 ¼ 13.9 F5,410 ¼ 1.2 F5,1568 ¼ 4.7 F5,8942 ¼ 5.4 F5,3211 ¼ 6.2 F5,33769 ¼ 13.8 F5,7197 ¼ 10.2 F5,60046 ¼ 26.8 F5,51944 ¼ 32.1 F5,23706 ¼ 33.3 p < 0.001*** p < 0.3 ns p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** p < 0.001*** Results of Tukey Humans(S)e humans(Q) Humans(S)e chimpanzees Humans(S)e gorillas Humans(S)e capuchins Humans(S)e lemurs Humans(Q)e chimpanzees Humans(Q)e gorillas Humans(Q)e capuchins Humans(Q)e lemurs Chimpanzeese gorillas Chimpanzeese capuchins Chimpanzeese lemurs Gorillase capuchins Gorillase lemurs Capuchinse lemurs
HSD post-hoc tests 0.001*** 0.80 NS
0.0016**
0.47 NS
0.02*
0.99 NS
0.001***
0.95 NS
0.01**
0.99 NS
0.99 NS
0.69 NS
0.08 NS
0.09 NS
0.02*
0.001***
0.001***
0.001***
0.001***
0.001***
0.99 NS
0.99 NS
0.99 NS
0.004**
0.0012**
0.001***
0.16 NS
0.001***
0.001***
0.001***
0.96 NS
0.35 NS
0.011*
0.97 NS
0.004**
0.03*
0.7 NS
0.001***
0.001***
0.001***
0.25 NS
0.96 NS
0.87 NS
0.63 NS
0.001***
0.02*
0.64 NS
0.003**
0.001***
0.001***
0.001***
0.99 NS
0.99 NS
0.81 NS
0.96 NS
0.001***
0.99 NS
0.001***
0.001***
0.001***
0.001***
0.87 NS
0.14 NS
0.21 NS
0.56 NS
0.001***
0.09 NS
0.001***
0.001***
0.001***
0.001***
0.91 NS
0.99 NS
0.99 NS
0.78 NS
0.056 NS
0.006**
0.001***
0.001***
0.001***
0.001***
0.99 NS
0.37 NS
0.044*
0.43 NS
0.04*
0.008**
0.03*
0.001***
0.001***
0.99 NS
0.76 NS
0.48 NS
0.95 NS
0.98 NS
0.95 NS
0.44 NS
0.71 NS
0.58 NS
0.84 NS
0.99 NS
0.99 NS
0.99 NS
0.62 NS
0.99 NS
0.43 NS
0.09 NS
0.99 NS
0.99 NS
0.99 NS
0.35 NS
0.99 NS
0.76 NS
0.007**
0.95 NS
0.5 NS
0.11 NS
0.06 NS
0.99 NS
0.98 NS
0.99 NS
0.47 NS
0.18 NS
0.15 NS
0.99 NS
0.11 NS
0.96 NS
0.37 NS
0.75 NS
0.97 NS
0.39 NS
0.97 NS
0.99 NS
0.001***
0.99 NS
0.15 NS
0.98 NS
0.001***
0.37 NS
0.47 NS
0.13 NS
0.91 NS
0.39 NS
0.36 NS
0.99 NS
0.99 NS
0.99 NS
0.22 NS
0.99 NS
0.91 NS
All of the non-human species were in quadrupedal posture and humans in sitting (S) and quadrupedal (Q) postures. NS: not significant; p < 0.1*; p < 0.01**; p < 0.001***.
features of lemurs and capuchins are consistent with those of the prehension of large objects with the whole hand in humans and macaques. The kinematics of apes is consistent with those of humans and macaques grasping small objects between the thumb and index finger. Finally, this study and the previous ones revealed similarities in kinematics and grip types for macaques, gorillas, chimpanzees and humans (kinematics: Roy et al., 2000, 2002; Christel and Billard, 2002; Sartori et al., 2012; grip types: Christel, 1993; MacFarlane and Graziano, 2009; Pouydebat et al., 2009). This suggests that prehension skills have evolved early on in Old World monkeys but further analyses are needed to shed light on the development of grasping abilities in New World monkeys.
capuchins and lemurs preferred movements in a parasagittal plane (flexion-extension) to reach the food whereas gorillas and chimpanzees favour shoulder rotation. At the elbow, the same motion strategies were observed for these species, except for capuchins, which preferred rotational movements similar to those of the nonhuman apes. The two strategies seem to be correlated with the structure of the forelimb joints, especially that of the shoulder. Indeed, the overall mobility of the forelimb in primates is affected by scapular and glenohumeral orientation (Jenkins, 1973; Larson, 1993; Whitehead and Larson, 1994; Schmidt et al., 2002; Chan, 2007a, b; Schmidt and Krause, 2011). Ranges of motion of the shoulder of capuchins and lemurs are not restricted to the parasagittal plane as are those of the macaques (Christel and Billard, 2002; Jindrich et al., 2011; Schmidt and Krause, 2011). Lemurs seem to be more constrained by their shoulder morphology than capuchins, who display a glenohumeral joint that is more independent from the scapula (Schmidt and Krause, 2011). During prehension, lemurs use wider ranges of shoulder flexion than
Kinematics of trunk and forelimb joints When we compared the quadrupedal species, including humans in this posture, the forelimb joints contributed to the prehension movement following two strategies. At the shoulder, humans,
Table 7 Means and standard deviations of angular values (in degrees) of the ranges of motion during prehension in the five species. Trunk flexion Humans (S) Humans (Q) Chimpanzees Gorillas Capuchins Lemurs
16.6 3.7 16.1 16.3 14.5 22.3
8 2.1 7.8 10.6 13.5 11.0
Trunk inclination 15.6 18.1 19.3 15.3 20.6 17.6
4.5 5.2 7.5 8.8 14.5 11.7
Trunk rotation 20.2 12.5 13.4 18.7 11.7 17.4
6.5 6.7 5.2 5.2 11.7 10.8
Shoulder flexion 56.6 48.8 41.8 36.1 52.2 65.3
10.7 12.8 23.8 17.4 25.3 26.9
Shoulder abduction 12.3 19.8 22.4 25.0 23.9 25.6
6.6 10.5 9.8 11.2 15.1 8.4
Shoulder rotation 23.2 22.1 58.2 64.7 42.6 43.7
11.7 11.8 35.3 37.0 19.6 25.6
Elbow flexion 51.0 32.6 34.3 42.2 45.8 45.4
All of the non-human species were in quadrupedal postures and humans in sitting (S) and quadrupedal (Q) postures.
10.9 9.6 16.8 12.0 14.8 9.2
Elbow rotation 17.3 21.7 63.1 74.1 58.9 41.6
6.5 10.0 29.5 34.2 35.2 18.2
Wrist extension 19.0 33.7 61.9 72.5 63.6 59.7
8.0 5.3 27.6 25.4 29.2 19.5
Wrist deviation 10.1 9.1 35.3 40.5 37.0 32.7
6.4 3.8 17.9 20.6 17.2 9.0
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
capuchins but their trunk is more inclined downward, which may force the execution of greater shoulder movements to advance the arm. However, in our data, both species presented similar ranges of motion overall with the exception of elbow rotation being higher in capuchins. In contrast, apes have the most mobile forelimbs in primates with a dorsal position of the scapula and high degree of motion of the glenohumeral joint in relation to climbing (Lewis, 1969; Tuttle, 1969; O’Connor and Rarey, 1979; Rose, 1989; Swartz, 1989; Thorpe et al., 1999; Schmidt and Krause, 2011; Zilhman et al., 2011). While humans display shoulder morphology similar to that of non-human apes (Corruccini, 1975; Chan, 2007a, b), they do not use the same joint motion strategy. Humans, which are strictly terrestrial bipeds, have a habitual use of the forelimb held below the level of the shoulder and supporting the weight of a pendant limb. Humans also show smaller muscular masses compared with other apes (Ashton and Oxnard, 1963; Thorpe et al., 1999), which could explain the lesser forelimb joint excursion (especially in abduction and rotation) during prehension compared with non-human apes. The orientation of the trunk varied according to the species, yet all species flexed the trunk in reaching except humans in a quadrupedal position. The orientation of the trunk is related to both the length of the limbs and the behaviour of the species during prehension. Indeed, the intermembral index, based on the ratio of the length of the forelimb to that of the hind limb, indicates that the closer the index is to 1 or exceeds it, the more upright the posture (Jungers, 1985; Fleagle, 1999). Apes share an index close to or superior to 1 and have longer forelimbs that orient the trunk in a more upward posture. Capuchins and lemurs have longer hind limbs than forelimbs, orienting the head more downwards. Moreover, lemurs move their head close to the food, increasing the downward posture of the trunk and much affecting the forelimb joints. Indeed, lemurs smell the food during prehension as has been previously observed for the grey mouse lemur (Reghem et al., 2011), for carnivores such as raccoons (Iwaniuk and Whishaw, 1999), and rodents such as gerbils and beavers (Whishaw et al., 1998). The ranges of motion of the trunk in lemurs contribute the most to prehension, differing from what was observed in the other species. To conclude, a variety of morphological differences related to locomotor style and environmental adaptations undoubtedly contribute to differences in kinematics. In order to better understand the evolution of manual abilities, we suggest an investigation of the influence of the mode of locomotion, diet, social context (e.g., self- and social grooming, learning), manipulation activities, and technical practice (e.g., fine-scale object manipulations, nest building, tool use, precision grip) in a diversity of primates. Only through such integrative studies will we gain true insights into the evolution of grasping in primates. Acknowledgements This research was supported by a grant from the foundation ‘ Marcel Bleustein Blanchet pour la vocation’ and by the Action Transversale du Muséum National d’Histoire Naturelle (Paris, France) ‘Formes possibles, Formes réalisées’. We are especially grateful to the director of the zoo Jardin Zoologique Tropical, Mr. Dupuyo, and to the director of the zoo La Palmyre, Mr. Caillé, and the veterinary, Mr. Petit, who permitted us to conduct this study. We also wish to thank the staff of the zoo, particularly Ronald Bosse, Redouane Lajali and Michel Carette, for their assistance during this study. We also thank A. Herrel, A. Borel, P.A. Libourel, N. Louis and J. Jacquier-Bret for their help and relevant remarks. We finally thank the reviewers who helped in the improvement of the manuscript.
311
References Alstermark, B., Lundberg, A., Pettersson, L.G., Tantisira, B., Walkowska, M., 1993. Characteristics of target-reaching in cats. I. Individual differences and intraindividual constancy. Exp. Brain Res. 94, 279e286. Ashton, E.H., Oxnard, C.E., 1963. The musculature of the primate shoulder. Trans. Zool. Soc. Lond. 29, 554e650. Bishop, A., 1964. Use of the hand in lower primates. In: Buettner-Janush, J. (Ed.), Evolutionary and Genetic Biology of Primates. Academic Press, New York, pp. 135e225. Bootsma, R.J., Marteniuk, R.G., MacKenzie, C.L., Zaal, F.T.J.M., 1994. The speedaccuracy trade-off in manual prehension: effects of movement amplitude, object size and object width on kinematic characteristics. Exp. Brain Res. 98, 535e541. Castiello, U., Bennett, K.M.B., Paulignan, Y., 1992. Does the type of prehension influence the kinematics of reaching. Behav. Brain Res. 50, 7e15. Chan, L.K., 2007a. Scapular position in primates. Folia Primatol. 78, 19e35. Chan, L.K., 2007b. Glenohumeral mobility in primates. Folia Primatol. 78, 1e18. Christel, M.I., 1993. Grasping techniques and hand preferences in apes and humans. In: Preuschoft, H., Chivers, D.J. (Eds.), Hands of Primates. Springer-Verlag, NewYork, pp. 91e108. Christel, M.I., Billard, A., 2002. Comparison between macaques’ and humans’ kinematics of prehension: the role of morphological differences and control mechanisms. Behav. Brain Res. 131, 169e184. Corruccini, R.S., 1975. Morphometric affinities in the forelimb of anthropoid primates. Z. Morph. Anthropol. 67, 19e61. Costello, M.B., Fragaszy, D.M., 1988. Prehension in Cebus and Saimiri: grip type and hand preference. Am. J. Primatol. 15, 235e245. Fitts, P.A., 1954. The information capacity of the human motor system in controlling the amplitude of movement. J. Exp. Psychol. 47, 381e391. Fleagle, J.G., 1999. Primate Adaptation and Evolution. Academic Press, San Diego. Gentilucci, M., Castiello, U., Corradini, M.L., Scarpa, M., Umilta, C., Rizzolatti, G., 1991. Influence of different types of grasping on the transport component of prehension movements. Neuropsychologia 29, 361e378. Graham, K.M., Moore, K.D., Cabel, D.W., Gribble, P.L., Cisek, P., Scott, S.H., 2003. Kinematics and kinetics of multijoint reaching in non human primates. J. Neurophysiol. 89, 2667e2677. Grosskpof, A., Kuhtz-Buschbeck, J.P., 2006. Grasping with the left and right hand: a kinematic study. Exp. Brain Res. 168, 230e240. Hof, A.L., 1996. Scaling gait data to body size. Gait Posture 4, 222e223. Hopkins, W.D., 1999. On the other hand: statistical issues in the assessment and interpretation of hand preference data in nonhuman primates. Int. J. Primatol. 20, 851e866. Ivanco, T.L., Pellis, S.M., Whishaw, I.Q., 1996. Skilled forelimb movements in prey catching and in reaching by rats (Rattus norvegicus) and opossums (Monodelphis domestica): relations to anatomical differences in motor systems. Behav. Brain Res. 79, 163e181. Iwaniuk, A.N., Whishaw, I.Q., 1999. How skilled are the skilled limb movements of the raccoon (Procyon lotor)? Behav. Brain Res. 99, 35e44. Jeannerod, M., 1981. Intersegmental coordination during reaching at natural visual objects. In: Long, J., Baddeley, A. (Eds.), Attention and Performance IX. Erlbaum, Hillsdale, pp. 153e168. Jeannerod, M., 1984. The timing of natural prehension movements. J. Mot. Behav. 16, 235e254. Jenkins Jr., F.A., 1973. The functional anatomy and evolution of the mammalian humero-ulnar articulation. Am. J. Anat. 137, 281e298. Jindrich, D.L., Courtine, G., Liu, J.J., McKay, H.L., Moseanko, R., Bernot, T.J., Roy, R.R., Zhong, H., Tuszynski, M.H., Edgerton, V.R., 2011. Unconstrained threedimensional reaching in rhesus monkeys. Exp. Brain Res. 209, 35e50. Jungers, W.L., 1985. Body size and scaling of limb proportions in primates. In: Jungers, W.L. (Ed.), Size and Scaling in Primate Biology. Plenum Publishing, New York, pp. 345e381. Larson, S.G., 1993. Functional morphology of the shoulder in primates. In: Gebo, D.L. (Ed.), Postcranial Adaptation in Nonhuman Primates. Northnern Illinois University Press, DeKalb, pp. 45e89. Lewis, O.J., 1969. The hominoid wrist joint. Am. J. Phys. Anthropol. 30, 251e268. MacFarlane, N.B., Graziano, M.S., 2009. Diversity of grip in Macaca mulatta. Exp. Brain Res. 197, 255e268. Marteniuk, R.G., MacKenzie, C.L., Jeannerod, M., Athenes, S., Dugas, C., 1987. Constraints on human arm movement trajectories. Can. J. Psychol. 41, 365e378. Napier, J.R., 1960. Studies of the hand of living primates. Proc. Zool. Soc. Lond. 134, 647e657. O’Connor, B.L., Rarey, K.E., 1979. Normal amplitudes of radioulnar pronation and supination in several genera of anthropoid primates. Am. J. Phys. Anthropol. 51, 39e44. Paulignan, Y., Frak, V.G., Toni, I., Jeannerod, M., 1997. Influence of object position and size on human prehension movements. Exp. Brain Res. 114, 226e234. Pouydebat, E., Laurin, M., Gorce, P., Bels, V., 2008. Evolution of grasping among anthropoids. J. Evol. Biol. 21, 1732e1743. Pouydebat, E., Gorce, P., Coppens, Y., Bels, V., 2009. Biomechanical study of grasping according to the volume of the object: human versus non-human primates. J. Biomech. 42, 266e272. Pouydebat, E., Reghem, E., Borel, A., Gorce, P., 2011. Diversity of grip in adults and young humans and chimpanzees (Pan troglodytes). Behav. Brain Res. 218, 21e28.
312
E. Reghem et al. / Journal of Human Evolution 65 (2013) 303e312
R Development Core Team, 2009. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. http://www.Rproject.org. Reghem, E., Pouydebat, E., Tia, B., Bels, V., 2011. Food prehension and manipulation in Microcebus murinus (Prosimii, Cheirogaleidae). Folia Primatol. 82, 177e188. Rose, M.D., 1989. New postcranial specimens of catarrhines from the Middle Miocene Chinji Formation, Pakistan: descriptions and a discussion of proximal humeral functional morphology in anthropoids. J. Hum. Evol. 18, 131e162. Roy, A.C., Paulignan, Y., Farnè, A., Jouffrais, C., Boussaoud, D., 2000. Hand kinematics during reaching and grasping in the macaque monkey. Behav. Brain Res. 117, 75e82. Roy, A.C., Paulignan, Y., Meunier, M., Boussaoud, D., 2002. Prehension movements in the macaque monkey: effects of object size and location. J. Neurophysiol. 88, 1491e1499. Roy, A.C., Paulignan, Y., Meunier, M., Boussaoud, D., 2006. Prehension movements in the macaque monkey: effects of perturbation of object size and location. Exp. Brain Res. 169, 182e193. Santello, M., Flanders, M., Soechting, J.F., 2002. Patterns of hand motion during grasping and the influence of sensory guidance. J. Neurosci. 22, 1426e1435. Sartori, L., Camperio Ciani, A., Bulgheroni, M., Castiello, U., 2012. Reaching and grasping behavior in Macaca fascicularis: a kinematic study. Exp. Brain Res. 224, 119e124. Schmidt, M., Krause, C., 2011. Scapula movements and their contribution to threedimensional forelimb excursions in quadrupedal primates. In: D’Août, K., Vereecke, E.E. (Eds.), Primate Locomotion: Linking Field and Laboratory Research. Springer Publishing, New York, pp. 83e108. Schmidt, M., Voges, D., Fischer, M.S., 2002. Shoulder movements during quadrupedal locomotion in arboreal primates. Z. Morph. Anthropol. 83, 235e242.
Scott, S.H., Kalaska, J.F., 1997. Reaching movements with similar paths but different arm orientations. I. Activity of individual cells in motor cortex. Am. Physiol. Soc. 77, 826e852. Spinozzi, G., Truppa, V., Lagana, T., 2004. Grasping behavior in tufted capuchin monkeys (Cebus apella): grip types and manual laterality for picking up a small food item. Am. J. Phys. Anthropol. 125, 30e41. Swartz, S.M., 1989. The functional morphology of weight bearing: limb joint surface area allometry in anthropoid primates. J. Zool. 218, 441e460. Thorpe, S.K.S., Crompton, R.H., Günther, M.M., Ker, R.F., McNeil, A.R., 1999. Dimensions and moment arms of the hind-and forelimb muscles of common chimpanzees (Pan troglodytes). Am. J. Phys. Anthropol. 110, 179e199. Tuttle, R.H., 1969. Quantitative and functional studies on the hands of Anthropoidea. I. The Hominoidea. J. Morphol. 128, 309e364. Vereecke, E., D’Août, K., Aerts, P., 2006. Speed modulation in hylobatid bipedalism: a kinematic analysis. J. Hum. Evol. 51, 513e526. Whishaw, I.Q., Sarna, J.R., Pellis, S.M., 1998. Evidence for rodent-common and species-typical limb and digit use in eating, derived from a comparative analysis of ten rodent species. Behav. Brain Res. 96, 79e91. Whitehead, P.F., Larson, S.G., 1994. Shoulder motion during quadrupedal walking in Cercopithecus aethiops: integration of cineradiographic and electromyographic data. J. Hum. Evol. 26, 525e544. Wu, G., van der Helm, F.C., Veeger, H.E., Makhsous, M., Van Roy, P., Anglin, C., Nagels, J., Karduna, A.R., McQuade, K., Wang, X., Werner, F.W., Buchholz, B., 2005. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion-Part II: shoulder, elbow, wrist and hand. J. Biomech. 38, 981e992. Zilhman, A.L., McFarland, R.K., Underwood, C.E., 2011. Functional anatomy and adaptation of gorillas (Gorilla gorilla gorilla) with comparisons to male orangutans (Pongo pygmaeus). Anat. Rec. 294, 1842e1855.