A phase transition model for the speed-accuracy trade-off in response time experiments

Area of Science:

• Cognitive Psychology
• Psychophysics
• Computational Neuroscience

Background:

• Traditional models assume a continuous speed-accuracy trade-off in cognitive tasks.
• This implies gradual accuracy sacrifice for increased response speed under stress.
• This trade-off is thought to span from accurate-slow to fast-guessing behavior.

Purpose of the Study:

• To challenge the assumption of continuity in speed-accuracy trade-off models.
• To propose and validate a phase transition model for response times (RTs) and accuracy.
• To differentiate the proposed model from existing fast guess and sequential sampling models.

Main Methods:

• Utilized catastrophe theory to derive predictions for the phase transition model.
• Designed experiments to test for hysteresis in processing mode transitions.
• Designed experiments to detect bimodal RT distributions under intermediate response conditions.

Main Results:

• Confirmed hysteresis in transitions between guessing and stimulus-controlled behavior when reward structures were manipulated.
• Confirmed bimodal RT distributions when participants were instructed to adopt an intermediate response strategy.
• Findings support a phase transition model over continuous trade-off assumptions.

Conclusions:

• The speed-accuracy trade-off in elementary cognitive tasks may not be continuous but exhibit phase transitions.
• A model with distinct guessing and stimulus-controlled modes better explains observed RT and accuracy patterns.
• Future research should explore the implications of this phase transition model for understanding cognitive control.