An evidence accumulation model for conflict detection performance in a simulated air traffic control task

Area of Science:

• Cognitive psychology
• Human factors engineering
• Decision science

Background:

• Human performance in dynamic decision-making tasks is often modeled using evidence accumulation.
• This approach assumes iterative sampling and accumulation of evidence over time.
• Decision thresholds dynamically adjust based on task urgency.

Purpose of the Study:

• To develop a formal computational model of conflict detection performance.
• To investigate the influence of conflict geometry and timing on human responses.

Main Methods:

• A formal evidence accumulation model was developed.
• Two experiments were conducted to collect data on response proportions and reaction times.
• Model predictions were compared against empirical data.

Main Results:

• The model accurately predicts observed response times, including complex relationships with approach parameters.
• Nonmonotonic relationships between distance at closest approach and response time were captured.
• Effects of angle of approach and relative velocity on performance were successfully modeled.

Conclusions:

• Evidence accumulation models provide a robust framework for understanding conflict detection.
• These models, a type of dynamic signal detection theory, can simulate human performance in dynamic decision tasks.
• The findings support the utility of evidence accumulation models for analyzing both response accuracy and timing.