Response preparation in a lane change task

Abstract

Anticipation of future events is crucial for driving performance and safety. The aim of this study is to assess the relevance of theoretical frameworks of response preparation (response priming, movement integration theory) for driving. In a customised lane change task, valid and invalid primes were used to indicate the direction of the forthcoming lane changes. Reaction time (RT) and phase durations from steering movements served as dependent measures. In agreement with the theoretical considerations, we found a clear effect of validity on RT and steering kinematics. RTs were faster and the duration of the initial steering phase was shorter with valid than with invalid advance information. The experimental outcomes suggest that the theoretical considerations about benefits and costs of response preparation can be generalised to driving manoeuvres. Therefore, response priming paradigms might be well suited to investigate preparatory effects of advance information, e.g. in the context of advanced driving assistance systems.

Practitioner summary: Benefits and costs of response preparation were assessed in the context of driving. The findings suggest that the understanding of preparatory processes is of relevance to enhance driving performance and safety. It is possible to derive some implications that may be useful for the design of assistance and information systems.

Keywords:

• response preparation
• validity effect
• steering dynamics
• lane change manoeuvres
• driving simulator

Acknowledgements

This research was supported by grants from the German Research Council (DFG: GU 331/6-1, 6-2). We thank Marc Grosjean and Andreas Bremer for helpful discussions, and two anonymous reviewers for valuable comments. We are also grateful to Daria Yakovleva for assistance with data collection. A summary of these data was presented as a poster at the 56th congress of the Gesellschaft für Arbeitswissenschaften, 24–26 March 2010, in Darmstadt, Germany.

Notes

^1. Participants on average responded 8 ms faster to the right (RT = 332 ms) than to the left side (340 ms), F(1,11) = 6.9, p = 0.02, η_p ² = 0.39. There was no significant interaction with the factor validity (p>0.58) or the factor proportion of validity (p>0.32) or both factors (p>0.13). The reason for the right-side bias may be due to the faster proceeding of arrows pointing to the right or due to faster preparation and/or execution of steering movements to the right side.