Brain alignment of reasoning and action representations from vision-language and action models during naturalistic gameplay

• URL: http://arxiv.org/abs/2605.19352v1
• Date: Tue, 19 May 2026 04:40:14 GMT
• Title: Brain alignment of reasoning and action representations from vision-language and action models during naturalistic gameplay
• Abstract summary: We study brain alignment of representative models from two foundation-model families using fMRI recordings.<n>We find that both vision-language models (VLMs) and large-action models (LAMs) exhibit significantly exhibit voxel-wise encoding performance.
• Score: 40.44107959566121
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Understanding how humans and artificial intelligence systems predict and plan by interacting with their environment is a fundamental challenge at the intersection of neuroscience and machine learning. Most brain-encoding studies focus on aligning artificial models with brain activity during language comprehension or passive visual processing, while interactive brain-alignment studies have to date been largely limited to reinforcement-learning (RL) agents and theory-based models. To address this gap, we study brain alignment of representative models from two foundation-model families, namely vision-language models (VLMs) and large-action models (LAMs), using fMRI recordings from participants playing naturalistic Atari-style video games. Specifically, we examine how action-focused and reasoning-focused prompts shape model's internal representations and align with fMRI brain activity. First, we find that both VLMs and LAMs exhibit significantly exhibit voxel-wise encoding performance than RL baselines, with the advantage holding even under matched feature dimensionality. Second, prompt-driven gains scale with the cortical processing hierarchy: the largest improvements appear in frontal-parietal and motor-planning regions, while early visual cortex gains roughly half as much. Third, variance partitioning reveals a qualitatively different representational organization: VLM is prompt-symmetric (12.5% unique action vs. 13.6% unique reasoning), whereas LAM is prompt-asymmetric (27% unique action vs. -5% unique reasoning), with the asymmetry strongest in frontal-motor cortex. Together, these results demonstrate that action-specialized fine-tuning reorganizes multimodal representations toward action-relevant neural computations even when whole-brain prediction accuracy is statistically equivalent between VLM and LAM.

Related papers

• Toward Cognitive Supersensing in Multimodal Large Language Model [67.15559571626747]
  We introduce Cognitive Supersensing, a training paradigm that endows MLLMs with human-like visual imagery capabilities.<n>In experiments, MLLMs trained with Cognitive Supersensing significantly outperform state-of-the-art baselines on CogSense-Bench.<n>We will open-source the CogSense-Bench and our model weights.
  arXiv  Detail & Related papers  (2026-02-02T02:19:50Z)
• Probing the Representational Geometry of Color Qualia: Dissociating Pure Perception from Task Demands in Brains and AI Models [6.165387850279033]
  We perform a rigorous comparison of the representational geometry of color qualia between state-of-the-art AI models and the human brain.<n>Our work contributes a new benchmark task for color qualia to the field, packaged in a Brain-Score compatible format.
  arXiv  Detail & Related papers  (2025-10-26T19:13:16Z)
• Uncovering Brain-Like Hierarchical Patterns in Vision-Language Models through fMRI-Based Neural Encoding [34.313883741642066]
  Current understanding of the parallels between artificial neural networks (ANNs) and human brain processing remains limited.<n>We propose a novel neuron-level analysis framework that investigates the multimodal information processing mechanisms in vision-language models (VLMs) through the lens of human brain activity.
  arXiv  Detail & Related papers  (2025-10-19T15:11:03Z)
• Disentangling the Factors of Convergence between Brains and Computer Vision Models [11.560007214914465]
  We show that the largest DINOv3 models trained with the most human-centric images reach the highest brain-similarity.<n>These findings disentangle the interplay between architecture and experience in shaping how artificial neural networks come to see the world as humans do.
  arXiv  Detail & Related papers  (2025-08-25T17:23:27Z)
• Do Large Language Models Think Like the Brain? Sentence-Level Evidence from fMRI and Hierarchical Embeddings [28.210559128941593]
  This study investigates how hierarchical representations in large language models align with the dynamic neural responses during human sentence comprehension.<n>Results show that improvements in model performance drive the evolution of representational architectures toward brain-like hierarchies.
  arXiv  Detail & Related papers  (2025-05-28T16:40:06Z)
• Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning [58.86928947970342]
  Embodied-R is a framework combining large-scale Vision-Language Models for perception and small-scale Language Models for reasoning.<n>After training on only 5k embodied video samples, Embodied-R with a 3B LM matches state-of-the-art multimodal reasoning models.<n>Embodied-R also exhibits emergent thinking patterns such as systematic analysis and contextual integration.
  arXiv  Detail & Related papers  (2025-04-17T06:16:11Z)
• Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs [65.93003087656754]
  VisFactor is a benchmark that digitizes 20 vision-centric subtests from a well-established cognitive psychology assessment.<n>We evaluate 20 frontier Multimodal Large Language Models (MLLMs) from GPT, Gemini, Claude, LLaMA, Qwen, and SEED families.<n>The best-performing model achieves a score of only 25.19 out of 100, with consistent failures on tasks such as mental rotation, spatial relation inference, and figure-ground discrimination.
  arXiv  Detail & Related papers  (2025-02-23T04:21:32Z)
• Brain-like Functional Organization within Large Language Models [58.93629121400745]
  The human brain has long inspired the pursuit of artificial intelligence (AI) Recent neuroimaging studies provide compelling evidence of alignment between the computational representation of artificial neural networks (ANNs) and the neural responses of the human brain to stimuli. In this study, we bridge this gap by directly coupling sub-groups of artificial neurons with functional brain networks (FBNs) This framework links the AN sub-groups to FBNs, enabling the delineation of brain-like functional organization within large language models (LLMs)
  arXiv  Detail & Related papers  (2024-10-25T13:15:17Z)
• Modelling Multimodal Integration in Human Concept Processing with Vision-Language Models [7.511284868070148]
  We investigate whether integration of visuo-linguistic information leads to representations that are more aligned with human brain activity.<n>Our findings indicate an advantage of multimodal models in predicting human brain activations.
  arXiv  Detail & Related papers  (2024-07-25T10:08:37Z)
• Multimodal foundation models are better simulators of the human brain [65.10501322822881]
  We present a newly-designed multimodal foundation model pre-trained on 15 million image-text pairs. We find that both visual and lingual encoders trained multimodally are more brain-like compared with unimodal ones.
  arXiv  Detail & Related papers  (2022-08-17T12:36:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.