Addressing cognitive bias in medical language models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schmidgall, Samuel, Harris, Carl, Essien, Ime, Olshvang, Daniel, Rahman, Tawsifur, Kim, Ji Woong, Ziaei, Rojin, Eshraghian, Jason, Abadir, Peter, Chellappa, Rama
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909114431438848
author Schmidgall, Samuel
Harris, Carl
Essien, Ime
Olshvang, Daniel
Rahman, Tawsifur
Kim, Ji Woong
Ziaei, Rojin
Eshraghian, Jason
Abadir, Peter
Chellappa, Rama
author_facet Schmidgall, Samuel
Harris, Carl
Essien, Ime
Olshvang, Daniel
Rahman, Tawsifur
Kim, Ji Woong
Ziaei, Rojin
Eshraghian, Jason
Abadir, Peter
Chellappa, Rama
contents There is increasing interest in the application large language models (LLMs) to the medical field, in part because of their impressive performance on medical exam questions. While promising, exam questions do not reflect the complexity of real patient-doctor interactions. In reality, physicians' decisions are shaped by many complex factors, such as patient compliance, personal experience, ethical beliefs, and cognitive bias. Taking a step toward understanding this, our hypothesis posits that when LLMs are confronted with clinical questions containing cognitive biases, they will yield significantly less accurate responses compared to the same questions presented without such biases. In this study, we developed BiasMedQA, a benchmark for evaluating cognitive biases in LLMs applied to medical tasks. Using BiasMedQA we evaluated six LLMs, namely GPT-4, Mixtral-8x70B, GPT-3.5, PaLM-2, Llama 2 70B-chat, and the medically specialized PMC Llama 13B. We tested these models on 1,273 questions from the US Medical Licensing Exam (USMLE) Steps 1, 2, and 3, modified to replicate common clinically-relevant cognitive biases. Our analysis revealed varying effects for biases on these LLMs, with GPT-4 standing out for its resilience to bias, in contrast to Llama 2 70B-chat and PMC Llama 13B, which were disproportionately affected by cognitive bias. Our findings highlight the critical need for bias mitigation in the development of medical LLMs, pointing towards safer and more reliable applications in healthcare.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08113
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Addressing cognitive bias in medical language models
Schmidgall, Samuel
Harris, Carl
Essien, Ime
Olshvang, Daniel
Rahman, Tawsifur
Kim, Ji Woong
Ziaei, Rojin
Eshraghian, Jason
Abadir, Peter
Chellappa, Rama
Computation and Language
Human-Computer Interaction
There is increasing interest in the application large language models (LLMs) to the medical field, in part because of their impressive performance on medical exam questions. While promising, exam questions do not reflect the complexity of real patient-doctor interactions. In reality, physicians' decisions are shaped by many complex factors, such as patient compliance, personal experience, ethical beliefs, and cognitive bias. Taking a step toward understanding this, our hypothesis posits that when LLMs are confronted with clinical questions containing cognitive biases, they will yield significantly less accurate responses compared to the same questions presented without such biases. In this study, we developed BiasMedQA, a benchmark for evaluating cognitive biases in LLMs applied to medical tasks. Using BiasMedQA we evaluated six LLMs, namely GPT-4, Mixtral-8x70B, GPT-3.5, PaLM-2, Llama 2 70B-chat, and the medically specialized PMC Llama 13B. We tested these models on 1,273 questions from the US Medical Licensing Exam (USMLE) Steps 1, 2, and 3, modified to replicate common clinically-relevant cognitive biases. Our analysis revealed varying effects for biases on these LLMs, with GPT-4 standing out for its resilience to bias, in contrast to Llama 2 70B-chat and PMC Llama 13B, which were disproportionately affected by cognitive bias. Our findings highlight the critical need for bias mitigation in the development of medical LLMs, pointing towards safer and more reliable applications in healthcare.
title Addressing cognitive bias in medical language models
topic Computation and Language
Human-Computer Interaction
url https://arxiv.org/abs/2402.08113