Predicting Perceptual Decisions Using Visual Cortical Population Responses and Choice History

Area of Science:

• Neuroscience
• Cognitive Science
• Visual Perception

Background:

• Understanding the neural basis of perceptual decision-making relies on linking neural activity fluctuations to behavioral choices.
• Previous research often compares single neuron responses across different experimental conditions, limiting direct comparisons.

Purpose of the Study:

• To directly compare choice signals in early (V1) and midlevel (V4) visual cortex during a perceptual task.
• To assess the benefit of analyzing neuronal populations versus individual neurons for predicting decisions.
• To determine if choice and reward history improve decision prediction models.

Main Methods:

• Simultaneous recording of neuronal activity in V1 and V4 of macaque monkeys during an orientation discrimination task.
• Analysis of individual neuron and small neuronal population responses.
• Incorporation of choice and reward history into predictive models.

Main Results:

• Individual V4 neurons showed weak predictive power for decisions, primarily between stimulus offset and choice.
• V1 neurons exhibited minimal decision-related activity.
• Neuronal population analysis revealed strong choice-related signals in V4 and weaker signals in V1.
• Including choice and reward history significantly improved decision prediction, especially with limited neuronal information.

Conclusions:

• Neuronal population activity and decision history are crucial for accurately relating neural responses to perceptual decisions.
• V4 plays a more significant role than V1 in representing choice information late in the decision process.
• Integrating multi-neuronal signals and behavioral history enhances our understanding of decision-making mechanisms.