From hand to mouth: monkeys require greater effort in motor preparation for voluntary control of vocalization than for manual actions.

Hiroki Koda,Takeshi Nishimura,Takumi Kunieda

doi:10.1098/rsos.180879

Abstract

Voluntary control of vocal production is an essential component of the language faculty, which is thought to distinguish humans from other primates. Recent experiments have begun to reveal the capability of non-human primates to perform vocal control; however, the mechanisms underlying this ability remain unclear. Here, we revealed that Japanese macaque monkeys can learn to vocalize voluntarily through a different mechanism than that used for manual actions. The monkeys rapidly learned to touch a computer monitor when a visual stimulus was presented and showed a capacity for flexible adaptation, such that they reacted when the visual stimulus was shown at an unexpected time. By contrast, successful vocal training required additional time, and the monkeys exhibited difficulty with vocal execution when the visual stimulus appeared earlier than expected; this occurred regardless of extensive training. Thus, motor preparation before execution of an action may be a key factor in distinguishing vocalization from manual actions in monkeys; they do not exhibit a similar ability to perform motor preparation in the vocal domains. By performing direct comparisons, this study provides novel evidence regarding differences in motor control abilities between vocal and manual actions. Our findings support the suggestion that the functional expansion from hand to mouth might be a critical evolutionary event for the acquisition of voluntary control of vocalizations.

Highlights

The infant begins to produce acoustically resonant vocalizations, known as cooing or babbling; these are regarded as speech sounds that lead to the capability for language production with the maturation of voluntary vocal control through infant –mother interactions [26]; notably, these later vocalizations are decoupled from emotional states
In the ‘direct connections’ hypothesis, which suggests that direct connections from the motor cortex onto terminal motor neurons provide enhanced voluntary control over action, learning control of motor timing would depend on the degree of motor connectivity
Direct assessments of the learning processes were seemingly difficult due to the unbalanced histories of the training parameters used for each subject; we examined differences in reaction times between vocal and manual actions during the training session, by using general linear mixed models (GLMMs) with considerations for action type as a fixed-effect term and the subject and sessions as random-effect terms with the lmer method in the lme4 package of R. ver 3.4.2

Summary

Introduction

Monkey vocalizations have been regarded as homologous to uncontrollable emotional sounds, such as infant cries, which are independent of controlled speech This notion of system dichotomy is insufficient to explain the recent progress of vocal controllability in non-human primates. In the ‘direct connections’ hypothesis, which suggests that direct connections from the motor cortex onto terminal motor neurons provide enhanced voluntary control over action, learning control of motor timing would depend on the degree of motor connectivity These dependencies have not previously been sufficiently examined. A hand-and-arm action (i.e. a manual action), requires voluntary motor control, which involves both direct and indirect projections from the motor cortex to motoneurons in macaques [47] This is equivalent to human vocal/manual control and contrasts with vocal control in non-human primates. The current attempt to compare these two motor learning processes facilitates a clearer understanding of the natures of both manual and vocal motor actions; it enables further characterization of the processes involved in vocal control in macaques

Objectives

Methods

Conclusion