Individual glance strategies and their effect on the NHTSA visual manual distraction test Robert Broström a,b,⇑ , Peter Bengtsson b,1 , Mikael Ljung Aust a,2 a Volvo Car Corporation, SE-405 31 Göteborg, Sweden b Luleå University of Technology, SE-971 87 Luleå, Sweden article info Article history: Received 8 April 2015 Received in revised form 20 August 2015 Accepted 11 October 2015 Available online 21 December 2015 Keywords: Individual differences Driving simulator testing Distraction guidelines Glance behaviour Design of in-vehicle technology abstract The purpose of this paper was to investigate how individual differences in glance strategy could impact the glance performance test defined in the NHTSA visual manual distraction guidelines. Better understanding of the test procedure could help development of new tech- nology for safe driving. A custom in-vehicle information system was developed and assessed in a driving simulator by eighteen participants. The interfaces were designed according to recommendations in the NHTSA guidelines and contained manual radio-tuning tasks, sound settings tasks and six letter spelling tasks. Two of the six tested interfaces fully complied with the test. In addition, clear individual differences in glance strategy were found among the participants. Four individual glance strategies were identified. Two of these, long glan- cers and frequent glancers, highly affected the outcome of the compliance test. Participants belonging to the long glancers and the frequent glancers categories were iden- tified as statistical outliers in many test cases. For example, if the individual values of these participants were replaced with sample mean, the number of complying interfaces would increase to five out of six, which is more in line with expectations for these interfaces. The results of this study show that individual variations in glance strategy exist. Also, these indi- vidual variations seem to have a non- negligible influence on the result when performance- testing of in-vehicle interfaces is done according to the NTHSA guidelines. Ó 2015 Elsevier Ltd. All rights reserved. 1. Introduction It has long been common sense to assume that it is risky to look away from the forward roadway while driving. More recently, this assumption has been confirmed by several naturalistic driving studies, which empirically show that looking away from the forward roadway for more than a certain period of time is a key crash contributing factor (Dingus et al., 2006; Klauer, Dingus, Neale, Sudweeks, & Ramsey, 2006; Liang, Lee, & Yekhshatyan, 2012; Victor & Dozza, 2011). Since the assumption that not looking at the road while driving has been around for a long time, there are many initiatives to reduce glance times for in-vehicle interfaces. One way in which these efforts have taken shape is through the formulation http://dx.doi.org/10.1016/j.trf.2015.10.017 1369-8478/Ó 2015 Elsevier Ltd. All rights reserved. ⇑ Corresponding author at: Driver Interaction & Infotainment Department 94700, Volvo Car Corporation PV32, SE-405 31 Göteborg, Sweden. Tel.: +46 31 32 50 146. E-mail addresses: robert.brostrom@volvocars.com (R. Broström), pebn@ltu.se (P. Bengtsson), mikael.ljung.aust@volvocars.com (M.L. Aust). 1 Tel.: +46 92 049 24 74. 2 Tel.: +46 31 59 63 52. Transportation Research Part F 36 (2016) 83–91 Contents lists available at ScienceDirect Transportation Research Part F journal homepage: www.elsevier.com/locate/trf