Explaining Too Much? Understanding How Large Language Model Reasoning Traces Influence Performance and Metacognition

2026-05-25

AI Summary

This study investigates how reasoning traces generated by large language models influence user performance, trust, enjoyment, and self-assessment accuracy in LSAT-style logical reasoning tasks. In a preregistered between-subjects experiment (N = 559), participants were randomly assigned to one of three conditions: answer-only, full reasoning trace, or summarized reasoning trace. Departing from prior work that treats reasoning traces as windows into model cognition, this research conceptualizes them as interface design elements. Results show that summarized traces enhance trust and enjoyment without compromising task performance, whereas full traces actually impair performance. Across all conditions, participants significantly overestimated their own accuracy, and no trace format improved calibration. Mediation analysis further reveals that enjoyment—not trust—drives this self-assessment bias.

Abstract

Large Language Model interfaces are increasingly verbose, exposing intermediate reasoning traces alongside final answers. Traces are framed as transparency mechanisms, yet it is unclear how people use them to solve problems. We report a preregistered between-subjects study (N = 559) in which participants solved ten LSAT-style reasoning problems under one of three conditions: an Answer-only baseline, a Full-trace revealed before the answer, and a Summary-trace presented alongside the answer. Summaries preserved task performance at the no-trace baseline while significantly elevating trust and hedonic appeal, establishing that trace exposure shifts subjective appraisal of the interaction without bringing performance benefits. Under an open-weight reasoning model exposing verbose intermediate output, full traces additionally impaired performance relative to the answer-only baseline. Across all conditions, participants substantially overestimated their performance, and no trace format supported calibrated self-evaluation. Further analysis indicates that hedonic appeal, not trust, carries the indirect path to overestimation, consistent with a processing-fluency account. Reasoning traces are best understood as user-facing interface artifacts rather than transparent windows into model cognition, and calibration is unlikely to emerge from the traces themselves and may best be scaffolded by interactions that elicit users' own reasoning first.

Problem

• reasoning traces
• large language models
• metacognition
• user trust
• performance calibration

Innovation

• reasoning traces
• metacognition
• large language models
• human-AI interaction
• calibration