Large Language Models are Capable of Offering Cognitive Reappraisal, if Guided
This work addresses the need for long-term mental well-being support through AI, offering a novel approach to emotional self-regulation, though it is a first step and incremental in applying known psychological principles to LLMs.
The paper tackled the problem of enabling large language models (LLMs) to provide cognitive reappraisal, a psychological strategy for emotional self-regulation, by developing RESORT, a set of psychologically grounded instructions. The result showed that even 7B-scale LLMs guided by RESORT could generate empathic responses that help users reappraise their situations, as evaluated by clinical psychologists.
Large language models (LLMs) have offered new opportunities for emotional support, and recent work has shown that they can produce empathic responses to people in distress. However, long-term mental well-being requires emotional self-regulation, where a one-time empathic response falls short. This work takes a first step by engaging with cognitive reappraisals, a strategy from psychology practitioners that uses language to targetedly change negative appraisals that an individual makes of the situation; such appraisals is known to sit at the root of human emotional experience. We hypothesize that psychologically grounded principles could enable such advanced psychology capabilities in LLMs, and design RESORT which consists of a series of reappraisal constitutions across multiple dimensions that can be used as LLM instructions. We conduct a first-of-its-kind expert evaluation (by clinical psychologists with M.S. or Ph.D. degrees) of an LLM's zero-shot ability to generate cognitive reappraisal responses to medium-length social media messages asking for support. This fine-grained evaluation showed that even LLMs at the 7B scale guided by RESORT are capable of generating empathic responses that can help users reappraise their situations.