The Compositional Nature of Event Representations in the Human Brain

TL;DR

This study investigates how the human brain encodes complex event representations by decoding fMRI data to identify constituents and their compositions, demonstrating that independent decoding of components can reconstruct sentential descriptions across subjects.

Contribution

It shows that the brain's representation of event constituents is compositional and that independent decoding of components can effectively reconstruct complex event descriptions.

Findings

Independent constituent decoding is comparable to joint classification accuracy.

Brain regions for constituent decoding are largely disjoint from those for joint concepts.

Cross-subject classifiers can recognize event descriptions across different individuals.

Abstract

How does the human brain represent simple compositions of constituents: actors, verbs, objects, directions, and locations? Subjects viewed videos during neuroimaging (fMRI) sessions from which sentential descriptions of those videos were identified by decoding the brain representations based only on their fMRI activation patterns. Constituents (e.g., "fold" and "shirt") were independently decoded from a single presentation. Independent constituent classification was then compared to joint classification of aggregate concepts (e.g., "fold-shirt"); results were similar as measured by accuracy and correlation. The brain regions used for independent constituent classification are largely disjoint and largely cover those used for joint classification. This allows recovery of sentential descriptions of stimulus videos by composing the results of the independent constituent classifiers. Furthermore, classifiers trained on the words one set of subjects think of when watching a video can recognise sentences a different subject thinks of when watching a different video.