The Smartphone and the Driver’s Cognitive Workload: A Comparison of Apple, Google, and Microsoft’s Intelligent Personal Assistants

The goal of this research was to examine the impact of voice-based interactions using 3 different intelligent personal assistants (Apple’s Siri, Google’s Google Now for Android phones, and Microsoft’s Cortana) on the cognitive workload of the driver. In 2 experiments using an instrumented vehicle on suburban roadways, we measured the cognitive workload of drivers when they used the voice-based features of each smartphone to place a call, select music, or send text messages. Cognitive workload was derived from primary task performance through video analysis, secondary-task performance using the Detection Response Task (DRT), and subjective mental workload. We found that workload was significantly higher than that measured in the single-task drive. There were also systematic differences between the smartphones: The Google system placed lower cognitive demands on the driver than the Apple and Microsoft systems, which did not differ. Video analysis revealed that the difference in mental workload between the smartphones was associated with the number of system errors, the time to complete an action, and the complexity and intuitiveness of the devices. Finally, surprisingly high levels of cognitive workload were observed when drivers were interacting with the devices: “on-task” workload measures did not systematically differ from that associated with a mentally demanding Operation Span (OSPAN) task. The analysis also found residual costs associated using each of the smartphones that took a significant time to dissipate. The data suggest that caution is warranted in the use of smartphone voice-based technology in the vehicle because of the high levels of cognitive workload associated with these interactions. Le but de cette recherche consistait à examiner, au moyen de trois différents assistants personnels (Siri de Apple, Google Now de Google pour téléphone Androïde et Cortana de Microsoft), l’impact d’interactions vocales sur la charge de travail cognitive du conducteur. À l’aide de deux expériences employant un véhicule instrumenté sur des routes de banlieue, nous avons mesuré la charge de travail cognitive de conducteurs alors qu’ils utilisaient les fonctionnalités vocales de chacun des téléphones intelligents pour effectuer un appel, sélectionner de la musique ou envoyer un message texte. La charge de travail cognitive a pu être déterminée après évaluation de la performance de la tâche principale par analyse-vidéo, de la performance de la tâche secondaire par tâche de détection-réponse (DRT) puis, de la charge de travail mentale subjective. Nous avons constaté que la charge de travail y était nettement plus élevée que celle associée à la tâche simple de conduire. Il y avait aussi des différences systématiques entre les téléphones intelligents. Le système Google était moins demandant cognitivement sur le conducteur que les systèmes Apple et Microsoft, lesquels avaient le même effet. L’analyse vidéo a montré que la différence au niveau de la charge de travail mentale entre téléphones intelligents était associée au nombre d’erreurs de système, à la durée de temps requise pour mener à bien une action et à la complexité et à l’intuitivité des appareils. Finalement, des niveaux étonnamment élevés de charge de travail cognitive ont été observés lorsque les conducteurs étaient en interaction avec leurs appareils : Les mesures de la charge de travail associée à la concentration sur une tâche ne différaient pas systématiquement de celles associées à une tâche (OSPAN) exigeante sur le plan mental. L’analyse a aussi révélé la présence de coûts résiduels associée à l’utilisation de chacun des téléphones intelligents, lesquels ont pris un temps considérable pour se dissiper. Les données suggèrent que la prudence est de mise en ce qui a trait à l’utilisation de technologie vocale sur téléphone intelligent dans un véhicule étant donné les niveaux élevés de la charge de travail cognitive associée à ces interactions.

[1]  D. Strayer,et al.  The crosstalk hypothesis: why language interferes with driving. , 2013, Journal of experimental psychology. General.

[2]  Andrew Heathcote,et al.  Working memory’s workload capacity , 2015, Memory & cognition.

[3]  Shayne Loft,et al.  Slow down and remember to remember! A delay theory of prospective memory costs. , 2015, Psychological review.

[4]  Donald L Fisher,et al.  Modeling situation awareness and crash risk. , 2014, Annals of advances in automotive medicine. Association for the Advancement of Automotive Medicine. Annual Scientific Conference.

[5]  D. Kahneman,et al.  Attention and Effort , 1973 .

[6]  Mica R. Endsley,et al.  Toward a Theory of Situation Awareness in Dynamic Systems , 1995, Hum. Factors.

[7]  A. Hartley,et al.  Age-related differences and similarities in dual-task interference. , 1999, Journal of experimental psychology. General.

[8]  J. Cerella,et al.  Aging and dual-task performance: a meta-analysis. , 2003, Psychology and aging.

[9]  S. Monsell,et al.  Costs of a predictible switch between simple cognitive tasks. , 1995 .

[10]  S. Hart,et al.  Development of NASA-TLX (Task Load Index): Results of Empirical and Theoretical Research , 1988 .

[11]  Raymond J. Shaw,et al.  Attention and Aging: A Functional Perspective , 2000 .

[12]  David L Strayer,et al.  Towards an understanding of driver inattention: taxonomy and theory. , 2014, Annals of advances in automotive medicine. Association for the Advancement of Automotive Medicine. Annual Scientific Conference.

[13]  David L. Strayer,et al.  Is the Technology in Your Car Driving You to Distraction? , 2015 .

[14]  David L. Strayer,et al.  Assessing Cognitive Distraction in the Automobile , 2015, Hum. Factors.

[15]  M. Endsley Situation Awareness Misconceptions and Misunderstandings , 2015 .

[16]  Donald L. Fisher,et al.  SPIDER: A Framework for Understanding Driver Distraction , 2016, Hum. Factors.

[17]  Dario D. Salvucci,et al.  Threaded cognition: an integrated theory of concurrent multitasking. , 2008, Psychological review.

[18]  David L. Strayer,et al.  Measuring Cognitive Distraction in the Automobile II: Assessing In-Vehicle Voice-Based InteractiveTechnologies , 2014 .

[19]  Daniel V. McGehee,et al.  Using Naturalistic Driving Data toAssess the Prevalence of EnvironmentalFactors and Driver Behaviors in TeenDriver Crashes , 2015 .

[20]  Charlene Hallett,et al.  Driver distraction and driver inattention: definition, relationship and taxonomy. , 2011, Accident; analysis and prevention.

[21]  Christopher D. Wickens,et al.  The Structure of Attentional Resources , 1980 .

[22]  David L. Strayer,et al.  Measuring Cognitive Distraction in the Automobile , 2013 .

[23]  D. Strayer,et al.  Passenger and Cell-Phone Conversations in Simulated Driving , 2004, Journal of experimental psychology. Applied.

[24]  David L. Strayer Attention and driving. , 2015 .

[25]  Johan Engström,et al.  Effects of visual and cognitive load in real and simulated motorway driving , 2005 .

[26]  Dario D. Salvucci Modeling Driver Behavior in a Cognitive Architecture , 2006, Hum. Factors.

[27]  D. Strayer,et al.  Supertaskers: Profiles in extraordinary multitasking ability , 2010, Psychonomic bulletin & review.

[28]  Raja Parasuraman,et al.  Varieties of attention , 1984 .

[29]  J. Gregory Trafton,et al.  Situation Awareness Recovery , 2014, Hum. Factors.

[30]  David L. Strayer,et al.  Driven to Distraction: Dual-Task Studies of Simulated Driving and Conversing on a Cellular Telephone , 2001, Psychological science.