Open Access

The affective connotations of words are central to meaning and important predictors of many social processes. As such, understanding the degree to which commercially-available generative language models (LLMs) replicate human judgements of affective connotations may help better understand human-model interactions. LLMs may also serve as useful tools for researchers seeking affective meaning estimates. We test the ability of three LLMs – GPT-4o, Mistral Large, and Llama 3.1 – to estimate human affective connotation ratings of words representing social identities, behaviours, modifiers, and settings in three language cultures: English (US), French (France), and German (Germany). We find that LLM ratings of terms correlate strongly with human ratings. However, their ratings tend to be overly extreme and patterns of correlations between meaning dimensions only loosely approximate those of human ratings. Consistent with previous findings of English-language and American biases in LLMs, we find that LLMs tend to perform better on English terms, though this pattern varies somewhat by meaning dimension and the type of term in question. We explore how LLMs might contribute to scholarship on affective connotations – by acting as tools for measurement – and how scholarship on affective connotations might contribute to generative language models – by guiding exploration of model biases.

This study explores how social identities, behaviors, and settings interact to shape urban dynamics, with gentrification and neighborhood stigmatization as example phenomena of city life. Using affect control theory (ACT) simulations, we demonstrate how social identities and behaviors recursively influence the affective meanings attributed to urban settings, thus driving local cultural change. In the case of gentrification, the simulation illustrates how positive identities, such as those related to the creative class, transform stigmatized neighborhoods into attractive spaces. Analyses of segregated areas reveal the self-reinforcing cycle of stigma. The simulations are based on a comparative empirical analysis of impression formation in settings in French and German language cultures. Our findings highlight overall similarities in impression formation across these cultures, with some subtle differences: Context effects of behavior are somewhat stronger among German survey respondents, and French participants rely more heavily on fundamental sentiments. This study shows how ACT simulations bridge the gap between qualitative and quantitative analysis and between micro and macro level of analysis, offering a novel framework for explaining urban interactions and local cultural change.