研究并改进人类和机器的推理能力

Nicolas Yax, Hernán Anlló, Stefano Palminteri
{"title":"研究并改进人类和机器的推理能力","authors":"Nicolas Yax, Hernán Anlló, Stefano Palminteri","doi":"10.1038/s44271-024-00091-8","DOIUrl":null,"url":null,"abstract":"In the present study, we investigate and compare reasoning in large language models (LLMs) and humans, using a selection of cognitive psychology tools traditionally dedicated to the study of (bounded) rationality. We presented to human participants and an array of pretrained LLMs new variants of classical cognitive experiments, and cross-compared their performances. Our results showed that most of the included models presented reasoning errors akin to those frequently ascribed to error-prone, heuristic-based human reasoning. Notwithstanding this superficial similarity, an in-depth comparison between humans and LLMs indicated important differences with human-like reasoning, with models’ limitations disappearing almost entirely in more recent LLMs’ releases. Moreover, we show that while it is possible to devise strategies to induce better performance, humans and machines are not equally responsive to the same prompting schemes. We conclude by discussing the epistemological implications and challenges of comparing human and machine behavior for both artificial intelligence and cognitive psychology. Some large language models show reasoning errors akin to humans in cognitive bias tasks. However, humans and models respond differently to prompting strategies, highlighting differences in cognitive processing.","PeriodicalId":501698,"journal":{"name":"Communications Psychology","volume":" ","pages":"1-16"},"PeriodicalIF":0.0000,"publicationDate":"2024-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s44271-024-00091-8.pdf","citationCount":"0","resultStr":"{\"title\":\"Studying and improving reasoning in humans and machines\",\"authors\":\"Nicolas Yax, Hernán Anlló, Stefano Palminteri\",\"doi\":\"10.1038/s44271-024-00091-8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In the present study, we investigate and compare reasoning in large language models (LLMs) and humans, using a selection of cognitive psychology tools traditionally dedicated to the study of (bounded) rationality. We presented to human participants and an array of pretrained LLMs new variants of classical cognitive experiments, and cross-compared their performances. Our results showed that most of the included models presented reasoning errors akin to those frequently ascribed to error-prone, heuristic-based human reasoning. Notwithstanding this superficial similarity, an in-depth comparison between humans and LLMs indicated important differences with human-like reasoning, with models’ limitations disappearing almost entirely in more recent LLMs’ releases. Moreover, we show that while it is possible to devise strategies to induce better performance, humans and machines are not equally responsive to the same prompting schemes. We conclude by discussing the epistemological implications and challenges of comparing human and machine behavior for both artificial intelligence and cognitive psychology. Some large language models show reasoning errors akin to humans in cognitive bias tasks. However, humans and models respond differently to prompting strategies, highlighting differences in cognitive processing.\",\"PeriodicalId\":501698,\"journal\":{\"name\":\"Communications Psychology\",\"volume\":\" \",\"pages\":\"1-16\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-06-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.nature.com/articles/s44271-024-00091-8.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Communications Psychology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.nature.com/articles/s44271-024-00091-8\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Communications Psychology","FirstCategoryId":"1085","ListUrlMain":"https://www.nature.com/articles/s44271-024-00091-8","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在本研究中,我们利用一系列传统上专门用于研究(有界)理性的认知心理学工具,对大型语言模型(LLMs)和人类的推理能力进行了研究和比较。我们向人类参与者和一系列经过预训练的大型语言模型展示了经典认知实验的新变体,并对它们的表现进行了交叉比较。我们的结果表明,大多数模型都出现了推理错误,这些错误与那些经常被归咎于容易出错的、基于启发式的人类推理的错误类似。尽管表面上有相似之处,但对人类和 LLMs 的深入比较表明,两者的推理与人类的推理有很大不同,在最近发布的 LLMs 中,模型的局限性几乎完全消失了。此外,我们还表明,虽然有可能设计出诱导更好表现的策略,但人类和机器对相同提示方案的反应并不相同。最后,我们讨论了比较人类和机器行为对人工智能和认知心理学的认识论意义和挑战。一些大型语言模型在认知偏差任务中表现出与人类类似的推理错误。然而,人类和模型对提示策略的反应不同,凸显了认知处理过程的差异。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

摘要图片

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Studying and improving reasoning in humans and machines
In the present study, we investigate and compare reasoning in large language models (LLMs) and humans, using a selection of cognitive psychology tools traditionally dedicated to the study of (bounded) rationality. We presented to human participants and an array of pretrained LLMs new variants of classical cognitive experiments, and cross-compared their performances. Our results showed that most of the included models presented reasoning errors akin to those frequently ascribed to error-prone, heuristic-based human reasoning. Notwithstanding this superficial similarity, an in-depth comparison between humans and LLMs indicated important differences with human-like reasoning, with models’ limitations disappearing almost entirely in more recent LLMs’ releases. Moreover, we show that while it is possible to devise strategies to induce better performance, humans and machines are not equally responsive to the same prompting schemes. We conclude by discussing the epistemological implications and challenges of comparing human and machine behavior for both artificial intelligence and cognitive psychology. Some large language models show reasoning errors akin to humans in cognitive bias tasks. However, humans and models respond differently to prompting strategies, highlighting differences in cognitive processing.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Parent-child relationship quality predicts higher subjective well-being in adulthood across a diverse group of countries Sensory stimulation enhances visual working memory capacity Psychoacoustic and Archeoacoustic nature of ancient Aztec skull whistles Three diverse motives for information sharing Cultural contexts differentially shape parents’ loneliness and wellbeing during the empty nest period
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1