Meta-Learning an In-Context Transformer Model of Human Higher Visual Cortex

TL;DR

This paper introduces BraInCoRL, a transformer-based in-context learning model that predicts human visual cortex responses from limited data, generalizes across subjects and stimuli, and enhances interpretability of neural signals.

Contribution

It presents a novel in-context learning approach for neural response prediction that requires no fine-tuning and generalizes well across datasets and subjects.

Findings

Outperforms existing models in low-data regimes

Generalizes to new datasets with different subjects and parameters

Enables interpretable mappings from language to neural responses

Abstract

Understanding functional representations within higher visual cortex is a fundamental question in computational neuroscience. While artificial neural networks pretrained on large-scale datasets exhibit striking representational alignment with human neural responses, learning image-computable models of visual cortex relies on individual-level, large-scale fMRI datasets. The necessity for expensive, time-intensive, and often impractical data acquisition limits the generalizability of encoders to new subjects and stimuli. BraInCoRL uses in-context learning to predict voxelwise neural responses from few-shot examples without any additional finetuning for novel subjects and stimuli. We leverage a transformer architecture that can flexibly condition on a variable number of in-context image stimuli, learning an inductive bias over multiple subjects. During training, we explicitly optimize the model for in-context learning. By jointly conditioning on image features and voxel activations, our model learns to directly generate better performing voxelwise models of higher visual cortex. We demonstrate that BraInCoRL consistently outperforms existing voxelwise encoder designs in a low-data regime when evaluated on entirely novel images, while also exhibiting strong test-time scaling behavior. The model also generalizes to an entirely new visual fMRI dataset, which uses different subjects and fMRI data acquisition parameters. Further, BraInCoRL facilitates better interpretability of neural signals in higher visual cortex by attending to semantically relevant stimuli. Finally, we show that our framework enables interpretable mappings from natural language queries to voxel selectivity.