Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident, Especially When They are Wrong

TL;DR

This paper investigates how reasoning before answering affects LLM confidence in MCQ tests, revealing that reasoning inflates confidence, especially when wrong, and degrades calibration.

Contribution

It demonstrates that reasoning increases LLM confidence levels and worsens calibration, highlighting the need for cautious interpretation of probabilities in MCQ evaluations.

Findings

Models are more confident when reasoning before answering.

Confidence increase is larger when the answer is incorrect.

Chain-of-Thought prompting degrades calibration accuracy.

Abstract

Multiple Choice Question (MCQ) tests are among the most used methods for evaluating large language models (LLMs). Besides checking the correctness of the selected answer, evaluations often consider the model's confidence through the probability assigned to its response. In this work, we investigate how LLM confidence is influenced by the answering approach when the model answers directly or reasons before responding. Experiments on a general knowledge benchmark, covering 57 subjects and seven LLMs, show that models are systematically more confident when providing reasoning before answering, and that this confidence increase is larger when the selected answer is incorrect than when it is correct. We hypothesize that the reasoning process alters token probabilities, as the final answer prediction depends jointly on the question and the model's self-generated reasoning, leading to inflated confidence estimates. Using standard calibration metrics such as Expected Calibration Error and Brier score, we further show that Chain-of-Thought (CoT) prompting degrades calibration by increasing the proportion of high-confidence wrong answers. These findings indicate that, in MCQ evaluation settings with CoT prompting, LLM-estimated probabilities should be used with caution as a basis for evaluation and metacognitive mechanisms.