Aware Yet Biased: Investigating Emotional Reasoning and Appraisal Bias in Large Language Models

认知心理学 心理学 计算机科学 认知科学
作者
Ala N. Tak,Jonathan Gratch,Klaus R. Scherer
出处
期刊:IEEE Transactions on Affective Computing [Institute of Electrical and Electronics Engineers]
卷期号:16 (4): 2871-2880 被引量:4
标识
DOI:10.1109/taffc.2025.3581461
摘要

This paper reports two studies investigating the emotional reasoning of Large Language Models (LLM). Previous research has suggested that LLMs are surprisingly accurate at predicting human emotions from text descriptions of situations and reason in a way that is consistent with appraisal theory—a leading theory of emotion. Study 1 tests this claim with a large multilingual corpus (English, French, and German) of autobiographical descriptions of emotionally charged events. We confirm that GPT-4, one of the most advanced and widely studied LLMs, shows a remarkable ability to predict emotion and appraisals. We further show this ability is language-independent, with accuracy being consistent across languages and unaffected by the language of the prompt. However, GPT-4 struggles to accurately predict certain emotions (shame, fear, and irritation) and fails to understand appraisal dimensions related to control and power. We repeat the experiments with Gemini-2.0-Flash and find a remarkably similar pattern of strengths and weaknesses, although it consistently outperforms GPT-4. Study 2 examines a possible mechanism for these failures based on the idea of cognitive appraisal bias. In psychological appraisal theory, appraisal bias is the idea that people evaluate situations in biased, often unrealistic ways. By testing both models on a set of situations designed to identify appraisal bias, we find they exhibit strong—but similar—appraisal bias; for example, evaluating situations as if they were a person high in agreeableness and low in power. We further offer evidence suggesting that LLMs could be debiased by incorporating a person's personality in the prompt. This research underscores LLMs' capabilities and limitations in emotional reasoning, though highlights one mechanism underlying this limitation and suggests an approach for addressing these limits.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
礼拜九完成签到,获得积分20
刚刚
刚刚
风驻云停完成签到,获得积分10
1秒前
科研通AI2S的应助被andjdd采纳,获得10
1秒前
彩色德天发布了新的文献求助10
1秒前
1秒前
1秒前
流水落花发布了新的文献求助10
1秒前
充电宝的应助被秾晓豆采纳,获得10
2秒前
竹墨笙箫完成签到,获得积分10
2秒前
于心发布了新的文献求助30
3秒前
3秒前
别来找喔完成签到 ,获得积分10
3秒前
开心鱼完成签到,获得积分20
3秒前
Hantheex完成签到,获得积分20
3秒前
3秒前
QIAO发布了新的文献求助10
3秒前
.....发布了新的文献求助10
3秒前
3秒前
科研通AI6.2的应助被草本采纳,获得10
3秒前
无极微光的应助被小鹿啊采纳,获得20
4秒前
QingY发布了新的文献求助10
4秒前
杨克学发布了新的文献求助10
4秒前
4秒前
一切顺利完成签到,获得积分10
5秒前
bkagyin的应助被合梦易采纳,获得10
5秒前
玛卡巴卡完成签到,获得积分20
5秒前
6秒前
共享精神的应助被zz采纳,获得10
6秒前
科研小能手的应助被lokiuiw采纳,获得80
6秒前
chrisio完成签到,获得积分10
6秒前
香蕉觅云的应助被lokiuiw采纳,获得10
6秒前
田様的应助被务实野狼采纳,获得20
6秒前
7秒前
7秒前
铁蛋儿完成签到,获得积分10
7秒前
温馨的雨发布了新的文献求助10
8秒前
1qq发布了新的文献求助10
8秒前
8秒前
科目三的应助被西红柿鸡蛋饼采纳,获得30
8秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1500
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8162621
求助须知:如何正确求助?哪些是违规求助? 9556889
关于积分的说明 21737583
捐赠科研通 7753002
什么是DOI,文献DOI怎么找? 3427836
关于科研通互助平台的介绍 2552719
邀请新用户注册赠送积分活动 2472507