# When to Call an Apple Red: Humans Follow Introspective Rules, VLMs Don't

arXiv: https://arxiv.org/abs/2604.06422 · HTML: https://mlsanity.com/paper/2604.06422/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 21 of its 30 references that are in the map.
- First posted on arXiv 7 Apr 2026.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 21 of 21)

- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Models](https://mlsanity.com/paper/2402.04614/) (2024)
- [Are self-explanations from Large Language Models faithful?](https://mlsanity.com/paper/2401.07927/) (2024)
- [More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models](https://mlsanity.com/paper/2505.21523/) (2025)
- [Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs](https://mlsanity.com/paper/2510.17771/) (2025)
- [Measuring and Improving Chain-of-Thought Reasoning in Vision-Language Models](https://mlsanity.com/paper/2309.04461/) (2023)
- [Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts](https://mlsanity.com/paper/2505.17127/) (2025)
- [ROME: Evaluating Pre-trained Vision-Language Models on Reasoning beyond Visual Common Sense](https://mlsanity.com/paper/2310.19301/) (2023)
- [FaithCoT-Bench: Benchmarking Instance-Level Faithfulness of Chain-of-Thought Reasoning](https://mlsanity.com/paper/2510.04040/) (2025)
- [Do Vision & Language Decoders use Images and Text equally? How Self-consistent are their Explanations?](https://mlsanity.com/paper/2404.18624/) (2024)
- [Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction](https://mlsanity.com/paper/2502.06882/) (2025)
- [Privileged Self-Access Matters for Introspection in AI](https://mlsanity.com/paper/2508.14802/) (2025)
- [On the Faithfulness of Visual Thinking: Measurement and Enhancement](https://mlsanity.com/paper/2510.23482/) (2025)
- [Counterfactual Simulation Training for Chain-of-Thought Faithfulness](https://mlsanity.com/paper/2602.20710/) (2026)
- [A Closer Look at Bias and Chain-of-Thought Faithfulness of Large (Vision) Language Models](https://mlsanity.com/paper/2505.23945/) (2025)
- [Explanation-Driven Counterfactual Testing for Faithfulness in Vision-Language Model Explanations](https://mlsanity.com/paper/2510.00047/) (2025)
- [On Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMs](https://mlsanity.com/paper/2602.12506/) (2026)
- [SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models](https://mlsanity.com/paper/2602.07833/) (2026)
- [Balancing Faithfulness and Performance in Reasoning via Multi-Listener Soft Execution](https://mlsanity.com/paper/2602.16154/) (2026)

## Built on by (most cited first, 1 of 1)

- [I Don't Miss You, but I Do: Self-Explanation Faithfulness of Modality Missingness in Vision-Language Models](https://mlsanity.com/paper/2609.07596/) (2026)

## Abstract (arXiv)

Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment. To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules. GCA consists of line drawings that vary pixel-level color coverage across three conditions: world-knowledge recolorings, counterfactual recolorings, and shapes with no color priors. Using GCA, we ask both VLMs and human participants to state a threshold rule: the share of an object's pixels that must be a given color for the object to receive that color label. We then compare these rules with their subsequent color attribution decisions. Our findings reveal that models systematically violate their own introspective rules. For example, GPT-5-mini violates its stated introspection rules in nearly 60% of cases on objects with strong color priors. Human participants remain faithful to their stated rules, with any apparent violations being explained by a well-documented tendency to overestimate color coverage. In contrast, we find that VLMs can accurately estimate color coverage, yet directly contradict their own reasoning in their final responses. Across all models and strategies for eliciting introspective rules, world-knowledge priors systematically degrade faithfulness in ways that do not mirror human cognition. Our findings challenge the view that VLM reasoning failures are difficulty-driven and suggest that VLM introspective self-knowledge is miscalibrated, with direct implications for high-stakes deployment.