Is GPT a Computational Model of Emotion? Detailed Analysis

TL;DR

This paper evaluates GPT models' ability to reason about emotions, revealing strengths in appraisals but weaknesses in predicting emotion intensity and coping, highlighting the importance of componential analysis.

Contribution

It provides a detailed component-wise analysis of GPT's emotional reasoning, emphasizing the need for targeted evaluation of emotional understanding in language models.

Findings

GPT predictions align with human emotional labels

GPT struggles with emotion intensity and coping prediction

GPT-4 performs best but still has limitations

Abstract

This paper investigates the emotional reasoning abilities of the GPT family of large language models via a component perspective. The paper first examines how the model reasons about autobiographical memories. Second, it systematically varies aspects of situations to impact emotion intensity and coping tendencies. Even without the use of prompt engineering, it is shown that GPT's predictions align significantly with human-provided appraisals and emotional labels. However, GPT faces difficulties predicting emotion intensity and coping responses. GPT-4 showed the highest performance in the initial study but fell short in the second, despite providing superior results after minor prompt engineering. This assessment brings up questions on how to effectively employ the strong points and address the weak areas of these models, particularly concerning response variability. These studies underscore the merits of evaluating models from a componential perspective.