Bipedal tool use strengthens chimpanzee hand preferences

Abstract

The degree to which non-human primate behavior is lateralized, at either individual or population levels, remains controversial. We investigated the relationship between hand preference and posture during tool use in chimpanzees (Pan troglodytes) during bipedal tool use. We experimentally induced tool use in a supported bipedal posture, an unsupported bipedal posture, and a seated posture. Neither bipedal tool use nor these supported conditions have been previously evaluated in apes. The hypotheses tested were 1) bipedal posture will increase the strength of hand preference, and 2) a bipedal stance, without the use of one hand for support, will elicit a right hand preference. Results supported the first, but not the second hypothesis: bipedalism induced the subjects to become more lateralized, but not in any particular direction. Instead, it appears that subtle pre-existing lateral biases, to either the right or left, were emphasized with increasing postural demands. This result has interesting implications for theories of the evolution of tool use and bipedalism, as the combination of bipedalism and tool use may have helped drive extreme lateralization in modern humans, but cannot alone account for the preponderance of right-handedness.

Fig. 1

Peanut butter tube suspension apparatus viewed from the side. The tube about to be put into the corral is in front of the suspension system and the fishing line is run through the larger PVC and wrapped around the top extension to secure it.

Fig. 2

Sketch of a chimpanzee performing the task for each condition. A) Baseline condition: seated tool use. B) Supported bipedal condition: bipedal tool use while using one arm for support. C) Bipedal condition: completely bipedal tool use without any support. Measurements indicated are d1: 75 cm maximum, d2: 100 cm minimum, h: 2.8 m.

Fig. 3

Mean absolute value of HI with error bars indicating mean standard error. The changes in mean absolute value of HI between the baseline and the supported condition was significant (*p < 0.02), as was that between the supported condition and the unsupported condition (*p < 0.01), and between the baseline and the unsupported bipedal posture (**p < 0.001).

Fig. 4

Number of right-handed (R), left-handed (L), and ambidextrous (A) subjects by condition. Note the steady decrease in the number of ambidextrous subjects as the task demanded increasing bipedality, and roughly equivalent numbers of right- and left-handed animals with a slight bias towards right-handedness.

Fig. 5

Mean absolute value of HI when re-examined per “bout” with error bars indicating mean standard error. All changes were significant: between the baseline and the supported condition (*p < 0.02), between supported and unsupported bipedal (**p < 0.001), and between baseline and unsupported bipedal (**p < 0.001).