# From Activation to Specificity: Automating Counterfactual Testing of Visual Representations in the Human Brain

arXiv: https://arxiv.org/abs/2605.23895 · HTML: https://mlsanity.com/paper/2605.23895/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 42 references that are in the map.
- First posted on arXiv 22 May 2026.

## Builds on (most cited first, 17 of 17)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023)
- [MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data](https://mlsanity.com/paper/2403.11207/) (2024)
- [Hypothesis Generation with Large Language Models](https://mlsanity.com/paper/2404.04326/) (2024)
- [Paint by Inpaint: Learning to Add Image Objects by Removing Them First](https://mlsanity.com/paper/2404.18212/) (2024)
- [Brain Diffusion for Visual Exploration: Cortical Discovery using Large Scale Generative Models](https://mlsanity.com/paper/2306.03089/) (2023)
- [NeuroGen: activation optimized image synthesis for discovery neuroscience](https://mlsanity.com/paper/2105.07140/) (2021)
- [A Survey on Hypothesis Generation for Scientific Discovery in the Era of Large Language Models](https://mlsanity.com/paper/2504.05496/) (2025)
- [The Wisdom of a Crowd of Brains: A Universal Brain Encoder](https://mlsanity.com/paper/2406.12179/) (2024)
- [MindSimulator: Exploring Brain Concept Localization via Synthetic FMRI](https://mlsanity.com/paper/2503.02351/) (2025)
- [Brain Mapping with Dense Features: Grounding Cortical Semantic Selectivity in Natural Images With Vision Transformers](https://mlsanity.com/paper/2410.05266/) (2024)
- [Transformer brain encoders explain human high-level visual responses](https://mlsanity.com/paper/2505.17329/) (2025)
- [In Silico Mapping of Visual Categorical Selectivity Across the Whole Brain](https://mlsanity.com/paper/2510.21142/) (2025)
- [Brain-IT: Image Reconstruction from fMRI via Brain-Interaction Transformer](https://mlsanity.com/paper/2510.25976/) (2025)
- [BrainExplore: Large-Scale Discovery of Interpretable Visual Representations in the Human Brain](https://mlsanity.com/paper/2512.08560/) (2025)

## Abstract (arXiv)

Identifying which brain regions represent a visual concept in the human brain is a central challenge in neuroscience. Existing approaches have localized coarse functional regions (e.g., faces, places) through activation maximization, identifying regions that activate strongly for a target concept relative to other concepts. Yet strong activation alone does not establish that a region represents the concept itself, as responses may instead be driven by correlated visual or semantic cues. We introduce BrainTRACE (Testing Representations through Counterfactual Evidence), an automated framework that combines generative and brain models to synthesize controlled stimuli and validate neural representations through targeted counterfactual-specificity testing. Given a query specifying a concept of interest, our framework constructs targeted stimulus sets comprising concept images, counterfactual edits that remove the target concept while preserving other image content, and images with candidate correlated distractors. It then uses an image-to-fMRI encoding model to predict brain responses and searches for representations that respond specifically to the target concept over correlated alternatives. BrainTRACE returns validated candidate representations and proposes follow-up fMRI experiments to further test or extend its discoveries. Our approach successfully recovers known functional localizations and identifies new candidate representations across dozens of concepts, validated on both predicted and measured fMRI data. Critically, we show that without counterfactual evaluation, a large fraction of localizations would be false positives, confirming that activation alone is insufficient evidence of representation.