Large-scale dynamics of perceptual decision information across human cortex

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Perceptual decisions involve accumulating sensory evidence.
• A common model posits sensory cortex encodes evidence, while motor regions accumulate it.
• The brain-wide distribution of this process is not well understood.

Purpose of the Study:

• To map the dynamics of stimulus- and choice-specific signals across the human cortex during visual decision-making.
• To differentiate between stimulus-dependent and endogenous components of choice-predictive activity.
• To investigate the role of early visual cortex in decision formation.

Main Methods:

• Developed a regionally-specific magnetoencephalography (MEG) decoding approach.
• Mapped neural signals across the human cortical surface during a visual decision task.
• Compared neural data with behaviorally inferred evidence accumulation dynamics.

Main Results:

• Identified an endogenous component of choice-predictive activity in early visual cortex (including V1).
• This activity was expressed in a low (<20 Hz) frequency band.
• The early visual cortex signal lagged behind the build-up of activity in (pre-)motor regions.

Conclusions:

• Early visual cortex plays a role in decision formation beyond simply encoding sensory evidence.
• Results support a model involving frequency-specific cortical feedback signaling during decision-making.
• Suggests a more distributed neural architecture for perceptual decisions than previously thought.