Constructing and Forgetting Temporal Context in the Human Cerebral Cortex

• Published in: Neuron (Cambridge, Mass.) Vol. 106; no. 4; pp. 675 - 686.e11
• Main Authors: Chien, Hsiang-Yun Sherry, Honey, Christopher J.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 20.05.2020; Elsevier Limited
• Subjects: Adolescent; Adult; Brain; Cerebral cortex; Cerebral Cortex - physiology; Cognition - physiology; computational modeling; Cortex (somatosensory); Cortex (temporal); Echo-Planar Imaging; event boundary; Female; fMRI; hierarchy; Humans; Information processing; Integration; inter-subject correlation; Male; Memory - physiology; Models, Neurological; prediction error; Quantitative psychology; sequence processing; temporal context; temporal integration; Time; timescales; Young Adult; event boundary; fMRI; temporal integration; hierarchy; inter-subject correlation; temporal context; timescales; prediction error; computational modeling; sequence processing
• ISSN: 0896-6273; 1097-4199; 1097-4199
• DOI: 10.1016/j.neuron.2020.02.013

How does information from seconds earlier affect neocortical responses to new input? We found that when two groups of participants heard the same sentence in a narrative, preceded by different contexts, the neural responses of each group were initially different but gradually fell into alignment. We observed a hierarchical gradient: sensory cortices aligned most quickly, followed by mid-level regions, while some higher-order cortical regions took more than 10 seconds to align. What computations explain this hierarchical temporal organization? Linear integration models predict that regions that are slower to integrate new information should also be slower to forget old information. However, we found that higher-order regions could rapidly forget prior context. The data from the cortical hierarchy were instead captured by a model in which each region maintains a temporal context representation that is nonlinearly integrated with input at each moment, and this integration is gated by local prediction error. •Distinct cortical responses when the same stimulus is preceded by different contexts•Responses align as common input continues: sensory cortex, then higher-order cortex•Cortical regions maintain a distributed and hierarchical representation of context•Distributed cortical memory is gated and prior context can be flexibly forgotten Chien and Honey measured how sequential information in a spoken narrative is integrated and separated in the human cerebral cortex. They observed a hierarchical representation of temporal context, distributed across the cortex. Computational modeling suggests the distributed context representation is flexibly updated or reset based on surprise.