Anthropology, University of California, Los Angeles
April 16, 2020 , Platinum Ballroom
A fundamental feature of human social cognition is the capacity to read others’ minds. Despite social cognition being mediated by language, no studies have examined whether language communities vary in their mental-state talk. We measured mental-state language in a small-scale society with implicit norms against attributing mental states to others. A simple response-elicitation task was administered in Spanish to bilingual Shuar / Spanish speakers in a small-scale, hunter-horticulturalist society in Amazonian Ecuador (N=40) as well as in English to a sample of American undergraduates (N=26). Participants were shown a set of nine short videos after which participants were asked to describe what had happened in the scene. Verbal responses were transcribed and coded according to a scheme adapted from earlier literature that categorized mental-state terms as affective, perception, desire, or epistemic. Word counts in each category were scaled to control for differences in description length. Hierarchical Poisson Regression models of word counts were run with culture and video as fixed factors and participant as a random factor. American participants used perception, affective, and epistemic words to describe characters in the videos significantly more often than Shuar participants did (p < 0.001). In contrast, Shuar participants used desire words more frequently (p < 0.001). Data on cross-cultural differences in mental state talk are important for a complete understanding of the evolution of language as they may illuminate the extent to which human language is constrained or free to vary in terms of how mental states are conceptualized and communicated.