Speed accuracy trade-off under response deadlines

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Decision Science

Background:

• Perceptual decision making is often modeled as evidence accumulation to a threshold.
• Maximizing rewards under response deadlines necessitates dynamic threshold adjustments to ensure timely responses.

Purpose of the Study:

• To test predictions that participants dynamically collapse decision thresholds under time pressure.
• To investigate the influence of deadline duration and timing uncertainty on threshold collapsing strategies.

Main Methods:

• Human participants performed a random dot motion discrimination task.
• Experimental conditions included free-response, short (800 ms), and long (1000 ms) deadlines.
• Response time and accuracy were analyzed in relation to deadline conditions.

Main Results:

• Accuracy did not decline to chance levels near the deadline as predicted by optimal performance.
• Observed accuracy declines with response time were explainable by a diffusion model with constant parameters.
• No significant relationship was found between accuracy-RT function declines and endogenous timing uncertainty.

Conclusions:

• Participants did not appear to dynamically adjust decision thresholds within trials under deadline conditions.
• Findings suggest that typical decision durations may be insufficient for within-trial adaptation of decision parameters.
• The standard diffusion model with constant parameters adequately describes the observed behavior.