Situation Models in the Brain are Used to Resolve Word References
Aliko, S.; Wittenberg, E.; Skipper, J. I.; Small, S.
Show abstract
Resolving the meaning of a pronoun requires retrieving information about a unique individual (a referent) from many remembered experiences and inferences throughout sensory and cognitive modalities, from vision and audition to episodic memory and social cognition. Here, we hypothesize that the functional neuroanatomy of pronoun interpretation involves interaction across distributed neural networks, during which each referent activates its own unique sensorimotor neural fingerprint associated with these experiences. To test this hypothesis, we collected fMRI data from 20 people watching a full-length movie, and developed a 3D branched convolutional neural network to distinguish movie characters from the fMRI signal across distributed sensorimotor regions. The same model distinguished the characters referenced by pronouns, using these same sensorimotor regions, supplemented by the hippocampus, precuneus, and medial prefrontal cortices. This work has far-reaching implications for understanding the relations among the many domains and modalities of neural representation required for ecological language comprehension. In particular, the demonstration that situation models, implemented in distributed sensory-motor and association cortices, are involved in resolving reference, suggests a whole brain distribution for language processing.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Encoding neural representations of time-continuous stimulus-response transformations in the human brain with advanced deep neural networks 97%
- Identifying and characterizing scene representations relevant for categorization behavior 96%
- Broad Brain Networks Support Curiosity-Motivated Incidental Learning Of Naturalistic Dynamic Stimuli With And Without Monetary Incentives 96%
Similar papers in this journal
Similar papers in this journal
- A Distributed Network for Multimodal Experiential Representation of Concepts 97%
- Unconscious manipulation of conceptual representations with decoded neurofeedback impacts search behaviour 97%
- Relating scene memory and perception activity to functional properties, networks, and landmarks of posterior cerebral cortex - A probabilistic atlas 96%
Similar papers in this journal
- Functional selectivity for naturalistic social interaction perception in the human superior temporal sulcus 97%
- Stimulus-independent neural coding of event semantics: Evidence from cross-sentence fMRI decoding 97%
- Language beyond the language system: dorsal visuospatial pathways support processing of demonstratives and spatial language during naturalistic fast fMRI 96%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.