# Learning to Trust: How Humans Mentally Recalibrate AI Confidence Signals

arXiv: https://arxiv.org/abs/2603.22634 · HTML: https://mlsanity.com/paper/2603.22634/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 35 references that are in the map.
- First posted on arXiv 23 Mar 2026.

## Builds on (most cited first, 9 of 9)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Inherent Trade-Offs in the Fair Determination of Risk Scores](https://mlsanity.com/paper/1609.05807/) (2016)
- [How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering](https://mlsanity.com/paper/2012.00955/) (2020)
- [Revisiting the Calibration of Modern Neural Networks](https://mlsanity.com/paper/2106.07998/) (2021)
- [Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making](https://mlsanity.com/paper/2001.02114/) (2020)
- [Calibration in Deep Learning: A Survey of the State-of-the-Art](https://mlsanity.com/paper/2308.01222/) (2023)
- ["Are You Really Sure?" Understanding the Effects of Human Self-Confidence Calibration in AI-Assisted Decision Making](https://mlsanity.com/paper/2403.09552/) (2024)
- [Capturing Humans' Mental Models of AI: An Item Response Theory Approach](https://mlsanity.com/paper/2305.09064/) (2023)
- [Understanding the Effects of Miscalibrated AI Confidence on User Trust, Reliance, and Decision Efficacy](https://mlsanity.com/paper/2402.07632/) (2024)

## Built on by (most cited first, 2 of 2)

- [Align AI to Dynamic Human-AI Workflows](https://mlsanity.com/paper/2607.14240/) (2026)
- [Referential Uncertainty in Human--AI Collaboration](https://mlsanity.com/paper/2609.39518/) (2026)

## Abstract (arXiv)

Productive human-AI collaboration requires appropriate reliance, yet contemporary AI systems are often miscalibrated, exhibiting systematic overconfidence or underconfidence. We investigate whether humans can learn to mentally recalibrate AI confidence signals through repeated experience. In a behavioral experiment (N = 200), participants predicted the AI's correctness across four AI calibration conditions: standard, overconfidence, underconfidence, and a counterintuitive "reverse confidence" mapping. Results demonstrate robust learning across all conditions, with participants significantly improving their accuracy, discrimination, and calibration alignment over 50 trials. We present a computational model utilizing a linear-in-log-odds (LLO) transformation and a Rescorla-Wagner learning rule to explain these dynamics. The model reveals that humans adapt by updating their baseline trust and confidence sensitivity, using asymmetric learning rates to prioritize the most informative errors. While humans can compensate for monotonic miscalibration, we identify a significant boundary in the reverse confidence scenario, where a substantial proportion of participants struggled to override initial inductive biases. These findings provide a mechanistic account of how humans adapt their trust in AI confidence signals through experience.