Searching for the majority: algorithms of voluntary control

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Human Information Processing

Background:

• Voluntary control over information processing is essential for resource allocation and prioritizing tasks.
• Understanding the underlying algorithms of this control is crucial but often unclear.
• The majority function task serves as a model for investigating these mental operations.

Purpose of the Study:

• To investigate and compare different algorithms of voluntary information processing.
• To determine the computational mechanisms underlying performance in the majority function task.
• To test competing hypotheses about mental operations based on input manipulation.

Main Methods:

• Participants performed a majority function task, identifying the dominant category (left/right arrows).
• Input amount (set size: 1, 3, 5) and content (arrow ratios) were systematically manipulated.
• Reaction time was measured and analyzed using a novel computational load metric.

Main Results:

• Reaction time was best predicted by a grouping search algorithm.
• This algorithm involves iterative sampling and resampling of input data before decision-making.
• Alternative algorithms, such as exhaustive or self-terminating search, were less predictive.

Conclusions:

• The grouping search algorithm provides a viable model for voluntary information processing in this task.
• Findings underscore the significance of algorithmic approaches to understanding voluntary cognitive control.
• Further research should explore the implications of these mental operation algorithms.