Behavioral training to improve collision detection.

Area of Science:

• Cognitive Psychology
• Human Factors Engineering
• Transportation Safety

Background:

• Young drivers represent a high-risk demographic for vehicle crashes.
• Inexperience in detecting impending collisions contributes to crash risk.
• Perceptual learning (PL) offers a potential training avenue to mitigate this risk.

Purpose of the Study:

• To evaluate the efficacy of perceptual learning (PL) in enhancing collision detection performance.
• To determine if PL training can improve young drivers' ability to identify collision and non-collision events.
• To assess the transferability of PL-induced improvements to different visual conditions.

Main Methods:

• A two-alternative forced choice procedure was used to measure collision detection thresholds.
• Participants underwent seven 1-hour training sessions focused on near-threshold collision detection.
• A second experiment contrasted near-threshold training with practice-based training above threshold.

Main Results:

• Post-training, participants demonstrated a significant reduction in the time required for collision detection at the trained speed.
• Collision detection improvements transferred to a higher observer speed condition.
• Training with stimuli well above threshold did not yield significant performance improvements.

Conclusions:

• Perceptual learning effectively enhances collision detection capabilities in young adults.
• PL training leads to transferable improvements, suggesting robust skill acquisition.
• The benefits of PL are specific to training near perceptual thresholds, not merely task practice.