Information flow across the cortical timescale hierarchy during narrative construction

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• The brain integrates information across multiple timescales for narrative comprehension.
• A hierarchical cortical processing chain with increasing temporal receptive windows is suggested for naturalistic narratives.

Purpose of the Study:

• To investigate the hypothesized sequence of response lags between functional brain networks during narrative comprehension.
• To determine if narrative structure elicits a specific temporal lag gradient across cortical areas.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was employed to measure brain activity.
• Intersubject cross-correlation analysis was used to assess network connectivity driven by shared narrative stimuli.
• Response lags between functional networks were estimated during narrative processing.

Main Results:

• A consistent temporal sequence of response lags, spanning several seconds, was identified.
• This gradient initiated in early auditory areas, progressed through language and attention networks, and concluded in the default mode network.
• The observed lag gradient was specific to coherent narrative comprehension, absent during rest or scrambled stimuli.

Conclusions:

• Narrative construction generates a distinct internetwork lag gradient, reflecting hierarchical processing.
• The findings support a model where gradual information accumulation in nested linguistic events drives this temporal sequence.
• This hierarchical lag structure is fundamental to understanding how the brain processes complex narratives.