Notes on Perceptual Decision Making and Evidence Accumulation

Perceptual Decision Making Overview

• Focus on decision making mechanisms in the human brain.
• Discussed stages: evidence representation, decision making, and outcome evaluation.

Evidence Representation

• Importance of having separate representations for different decision alternatives (e.g., faces, houses).
• Utilization of specific cortical regions to analyze different categories.

Electrophysiology vs. Neuroimaging

• Issues with using MT for micro-level observations.
• Suggested alternative: separate cortical regions that specialize in processing specific stimuli.

Visual Discrimination Tasks

• Emphasis on the parametrically modulated responses based on external noise affecting stimulus evidence.
• Correlation between firing patterns of neurons and behavioral outcomes, using monkeys as a study model.

Drift Diffusion Model (DDM)

• DDM outlines how different neural representations are compared and accumulated over time to make decisions.
• Decision boundaries represent potential outcomes (right vs. left choices).
• Rate of evidence accumulation relates to stimulus difficulty; easier tasks yield faster accumulations.
• Need to identify brain regions where this integration of evidence occurs.

fMRI and Accumulation Modelo

• fMRI studies predict behavioral outcomes by analyzing fMRI signal at decision-making pressures.
• Distinct trial types (reaction time tasks) provide comparative insight into neural activity patterns related to decision making.

Left Dorsolateral Prefrontal Cortex (DLPFC)

• Found to be crucial in decision making and evidence accumulation processes.
• Facilitation of actions based on determined choices, suggesting an interconnectedness of decision-making and motor response systems.
• Experiments indicated that certain brain regions might govern decisions independent of the required motor responses.

Experiment Design in Decision Making

• Tasks designed to ensure decisions are made independently of motor plans, enhancing clarity in decision-making processes.
• Findings show that decision-making processes are discernibly distinguishable from motor responses.

Confidence in Decision Making

• Confidence measured pre-decision as a possible indicator of future learning and performance.
• Evidence suggests that confidence may also develop from the ongoing accumulation of evidence during the decision-making process.
• Proposed that pre-decisional confidence assessments can be implicitly motivational, signaling the need for improvement in decision-making accuracy.

Post-Decisional Confidence

• Studies reveal neural correlates of confidence ratings occur after the decision-making phase.
• Right rostrolateral prefrontal cortex implicated in post-decision confidence assessments.
• The left DLPFC, which signifies early accumulation, appears to communicate with the right rostrolateral area for full confidence evaluations.

Speed-Accuracy Tradeoff (SAT)

• Importance highlighted in ensuring decisions are made appropriately under speed vs. accuracy conditions.
• Findings indicate that longer integration times yield higher accuracy at the cost of response speed.
• Brain regions responsible for managing the decision boundaries during SAT conditions, feeding information to the areas responsible for initiating responses.

Learning and Decision-Making

• Studies indicate synaptic improvements in decision-related areas over time, emphasizing a functional shift towards decision-making efficiency rather than raw perceptual acuity.

Conclusion

• Collective insights from behavior, neuroimaging, and modeling perspectives underline the complexity of decision-making processes in the brain.
• Encouraged to refer to summarized conclusions for exam preparation, connecting adaptive responses within decision-making interpretations.