Response-time dynamics: evidence for linear and low-dimensional nonlinear structure in human choice sequences.

Area of Science:

• Cognitive Psychology
• Dynamical Systems Theory
• Neuroscience

Background:

• Response time (RT) is a standard metric for cognitive performance, typically viewed as random.
• Hidden deterministic patterns may exist within the fluctuations of RT.
• Dynamical systems analysis offers methods to explore these complex, potentially deterministic RT patterns.

Purpose of the Study:

• To investigate if response time fluctuations exhibit low-dimensional chaotic characteristics.
• To determine the influence of task demands, specifically inter-stimulus interval (ISI) duration, on RT dynamics.
• To explore the hypothesis that demanding tasks may induce a response strategy reflected in RT dynamics.

Main Methods:

• Applied dynamical systems analysis techniques to RT data from a forced-pace serial RT task.
• Utilized careful task construction, noise-reduction, and surrogate series tests for rigorous analysis.
• Manipulated the inter-stimulus interval (ISI) duration across two experiments.

Main Results:

• In Experiment 1, 80% of subjects' filtered RT series displayed low-dimensional chaotic characteristics, including sensitive dependence and stable attractor geometry.
• Experiment 2 revealed that a small ISI led to 100% of subjects exhibiting chaotic RT dynamics.
• Conversely, a large ISI resulted in only 25% of subjects showing low-dimensional chaotic RT components.

Conclusions:

• RT fluctuations can possess low-dimensional chaotic properties, challenging the purely stochastic view.
• Task demands, particularly short ISIs, significantly increase the prevalence of chaotic RT dynamics.
• These findings suggest that RT variability may reflect a deterministic response strategy employed under demanding cognitive conditions.