A neurocognitive reward-based model of language evolution

Abstract

The evolution of language is often framed in terms of social utility, emphasizing communication, cooperation, or courtship. Here, we propose a reward-centered framework that integrates both social and intrinsic reinforcement mechanisms as drivers of linguistic evolution. We first highlight intrinsic reward arising from representation matching i.e. the satisfaction generated when internal models align with sensory inputs, a process previously shown to be supported by reward prediction error circuits across species. By analogy, we then propose the existence at early evolutionary times of another category of intrinsic reward: the reinforcement triggered when self-generated, perceivable signals (vocalizations, gestures, or other expressions) correspond to external or internal referents. We suggest that such intrinsic reward generated by symbolic matching preceded and facilitated the emergence of symbolic competence and, through exaptation, the development of complex language. Supporting our reward-centered hypothesis, comparative neuroanatomical and genomic evidence indicate that humans possess a uniquely developed reward system. By situating intrinsic reward at the core of language evolution, this framework extends beyond social explanations, offering a unifying account of how symbolic cognition paved the way for symbolic communication, ultimately shaping the cognitive and social capacities of Homo sapiens.