Simultaneous EEG-fMRI Study Reveals Separate Bias Mechanisms Modulated by Stimulus Evidence During Rapid Face...

Area of Science:

• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Perceptual choice is influenced by stimulus evidence and internal biases.
• The drift-diffusion model (DDM) offers frameworks for bias in decision-making, either via drift rate or starting point.
• Understanding the neural basis of these distinct bias mechanisms is crucial.

Purpose of the Study:

• To spatially and temporally dissociate the neural correlates of decision bias in a visual categorization task.
• To investigate how stimulus evidence level modulates the neural basis of bias.
• To differentiate the neural underpinnings of drift rate vs. starting point bias within the DDM framework.

Main Methods:

• Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) were acquired during a three-choice visual task (Face vs. Car vs. House).
• The drift-diffusion model (DDM) was fitted to behavioral data to quantify decision bias parameters.
• EEG single-trial variability was used to temporally link fMRI signals to bias parameters.

Main Results:

• DDM fitting indicated that drift rate changes best explained face bias with abundant sensory evidence.
• Starting point shifts best explained face bias with inadequate sensory evidence.
• fMRI results revealed a double dissociation: drift rate bias correlated with early sensory networks, while starting point bias engaged late decision-related networks.

Conclusions:

• Decision bias in perceptual tasks has distinct neural correlates depending on the level of sensory evidence.
• Bias in drift rate is associated with early sensory processing, whereas bias in starting point is linked to later decision-making networks.
• This study provides spatio-temporal evidence for separate neural mechanisms underlying different bias parameters in the DDM.