Traits are the currency of personality and social psychologists; they help us understand andrepresent both real and perceived differences between people and groups. Here, we advance ourunderstanding of traits by revealing new insights into their content, structure, and evolution usinglarge language models and human raters. Across four studies, we: (a) introduce one of the mostcomprehensive trait lists to-date (2,847 trait words); (b) establish normative human ratings ofthese traits (from N = 3,070 English speakers) along 24 empirically-derived dimensions; (c)detect a four-factor structure of Fitness, Agency, Communion, and Traditionalism (FACT) fromthese dimensions; and (d) reveal systematic differences in the coherence, valence, and historicalevolution of these dimensions and factors. We show that the FACT model covers significantlymore traits than previous models like the Big Five (96% vs. 79%). Moreover, these four factorsdiffer in their coherence (e.g., Communion is the most semantic coherent while Traditionalism isthe least), valence (e.g., Communion and Agency show more alignment between their semanticdirection and valence than Traditionalism and Fitness; Communion has more negative wordswhile Agency and Traditionalism have more positive ones); and historical usage (e.g., Fitnessand Communion are older than Traditionalism; Communion is converging in semantic spacewhile Agency is diverging). Our paper showcases new methods for discovering and analyzingtrait words, and provides a new functional constructionist theoretical perspective on the structure,content, and evolution of trait language.
Hou et al. (Thu,) studied this question.