# Cognitive bias in LLM reasoning compromises interpretation of clinical oncology notes

arXiv: https://arxiv.org/abs/2511.20680 · HTML: https://mlsanity.com/paper/2511.20680/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 19 of its 62 references that are in the map.
- First posted on arXiv 16 Nov 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)

## Builds on (most cited first, 19 of 19)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Capabilities of GPT-4 on Medical Challenge Problems](https://mlsanity.com/paper/2303.13375/) (2023)
- [Adapted Large Language Models Can Outperform Medical Experts in Clinical Text Summarization](https://mlsanity.com/paper/2309.07430/) (2023)
- [Can large language models reason about medical questions?](https://mlsanity.com/paper/2207.08143/) (2022)
- [Towards Accurate Differential Diagnosis with Large Language Models](https://mlsanity.com/paper/2312.00164/) (2023)
- [Diagnostic Reasoning Prompts Reveal the Potential for Large Language Model Interpretability in Medicine](https://mlsanity.com/paper/2308.06834/) (2023)
- [Unleashing the potential of prompt engineering for large language models](https://mlsanity.com/paper/2310.14735/) (2023)
- [Multi-LogiEval: Towards Evaluating Multi-Step Logical Reasoning Ability of Large Language Models](https://mlsanity.com/paper/2406.17169/) (2024)
- [Code Comprehension then Auditing for Unsupervised LLM Evaluation](https://mlsanity.com/paper/2410.03131/) (2024)
- [Few shot chain-of-thought driven reasoning to prompt LLMs for open ended medical question answering](https://mlsanity.com/paper/2403.04890/) (2024)
- [Limitations of Large Language Models in Clinical Problem-Solving Arising from Inflexible Reasoning](https://mlsanity.com/paper/2502.04381/) (2025)
- [Step-by-Step Reasoning to Solve Grid Puzzles: Where do LLMs Falter?](https://mlsanity.com/paper/2407.14790/) (2024)
- [Benchmark datasets driving artificial intelligence development fail to capture the needs of medical professionals](https://mlsanity.com/paper/2201.07040/) (2022)
- [Uncertainty Estimation of Large Language Models in Medical Question Answering](https://mlsanity.com/paper/2407.08662/) (2024)
- [CORAL: Expert-Curated medical Oncology Reports to Advance Language Model Inference](https://mlsanity.com/paper/2308.03853/) (2023)
- [Quality of Answers of Generative Large Language Models vs Peer Patients for Interpreting Lab Test Results for Lay Patients: Evaluation Study](https://mlsanity.com/paper/2402.01693/) (2024)
- [Large Language Models in Healthcare](https://mlsanity.com/paper/2503.04748/) (2025)
- [Medical Reasoning in LLMs: An In-Depth Analysis of DeepSeek R1](https://mlsanity.com/paper/2504.00016/) (2025)

## Abstract (arXiv)

Despite high performance on clinical benchmarks, large language models may reach correct conclusions through faulty reasoning, a failure mode with safety implications for oncology decision support that is not captured by accuracy-based evaluation. In this two-cohort retrospective study, we developed a hierarchical taxonomy of reasoning errors from GPT-4 chain-of-thought responses to real oncology notes and tested its clinical relevance. Using breast and pancreatic cancer notes from the CORAL dataset, we annotated 600 reasoning traces to define a three-tier taxonomy mapping computational failures to cognitive bias frameworks. We validated the taxonomy on 822 responses from prostate cancer consult notes spanning localized through metastatic disease, simulating extraction, analysis, and clinical recommendation tasks. Reasoning errors occurred in 23 percent of interpretations and dominated overall errors, with confirmation bias and anchoring bias most common. Reasoning failures were associated with guideline-discordant and potentially harmful recommendations, particularly in advanced disease management. Automated evaluators using state-of-the-art language models detected error presence but could not reliably classify subtypes. These findings show that large language models may provide fluent but clinically unsafe recommendations when reasoning is flawed. The taxonomy provides a generalizable framework for evaluating and improving reasoning fidelity before clinical deployment.