Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The ISLab Solution to the Algonauts Challenge 2025: A Multimodal Deep Learning Approach to Brain Response Prediction

Andrea Corsico, Giorgia Rigamonti, Simone Zini, Luigi Celona, Paolo Napoletano

TL;DR

This work presents a network-specific approach for predicting brain responses to complex multimodal movies, leveraging the Yeo 7-network parcellation of the Schaefer atlas, and demonstrates that this clustered strategy significantly enhances prediction accuracy across the 1,000 cortical regions of the Schaefer atlas.

Abstract

In this work, we present a network-specific approach for predicting brain responses to complex multimodal movies, leveraging the Yeo 7-network parcellation of the Schaefer atlas. Rather than treating the brain as a homogeneous system, we grouped the seven functional networks into four clusters and trained separate multi-subject, multi-layer perceptron (MLP) models for each. This architecture supports cluster-specific optimization and adaptive memory modeling, allowing each model to adjust temporal dynamics and modality weighting based on the functional role of its target network. Our results demonstrate that this clustered strategy significantly enhances prediction accuracy across the 1,000 cortical regions of the Schaefer atlas. The final model achieved an eighth-place ranking in the Algonauts Project 2025 Challenge, with out-of-distribution (OOD) correlation scores nearly double those of the baseline model used in the selection phase. Code is available at https://github.com/Corsi01/algo2025.

The ISLab Solution to the Algonauts Challenge 2025: A Multimodal Deep Learning Approach to Brain Response Prediction

TL;DR

This work presents a network-specific approach for predicting brain responses to complex multimodal movies, leveraging the Yeo 7-network parcellation of the Schaefer atlas, and demonstrates that this clustered strategy significantly enhances prediction accuracy across the 1,000 cortical regions of the Schaefer atlas.

Abstract

In this work, we present a network-specific approach for predicting brain responses to complex multimodal movies, leveraging the Yeo 7-network parcellation of the Schaefer atlas. Rather than treating the brain as a homogeneous system, we grouped the seven functional networks into four clusters and trained separate multi-subject, multi-layer perceptron (MLP) models for each. This architecture supports cluster-specific optimization and adaptive memory modeling, allowing each model to adjust temporal dynamics and modality weighting based on the functional role of its target network. Our results demonstrate that this clustered strategy significantly enhances prediction accuracy across the 1,000 cortical regions of the Schaefer atlas. The final model achieved an eighth-place ranking in the Algonauts Project 2025 Challenge, with out-of-distribution (OOD) correlation scores nearly double those of the baseline model used in the selection phase. Code is available at https://github.com/Corsi01/algo2025.

Paper Structure

This paper contains 16 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. The Algonauts Challenge
  4. Problem Definition
  5. Data
  6. Feature extraction
  7. Visual Features
  8. Audio features
  9. Language Features
  10. Multi subjects model
  11. Network memory modeling
  12. Experiments
  13. Brain responses to in-distribution movies
  14. Brain responses to out-of-distribution movies
  15. Results
  16. ...and 1 more sections

Figures (3)

  • Figure 1: Yeo 7-network parcellation of Schaefer atlas.
  • Figure 2: Temporal response patterns across Yeo networks and modalities. Correlation performance as a function of HRF delay (0-50 time points) for individual modalities and combined features across the seven functional networks.
  • Figure 3: Brain encoding performance comparison between in-distribution and out-of-distribution evaluation. Correlation maps showing prediction accuracy across cortical regions, averaged across subjects (ID) and across cortical regions, movies and subjects (OOD).