The smartphone and the driver's cognitive workload: A comparison of Apple, Google, and Microsoft's intelligent personal assistants
DOI: 10.1037/cep0000104
archive: archived pipeline: cataloged verified
Summary
Strayer, Cooper, Turrill, Coleman, and Hopman compared the driver workload imposed by Apple's Siri, Google Now, and Microsoft's Cortana intelligent personal assistants. In two on-road instrumented-vehicle experiments on suburban Salt Lake City roadways, participants used each smartphone's voice features to place calls, select music, and send text messages while measures of primary-task performance (video coding), DRT response time, and NASA-TLX were collected; OSPAN served as a high-cognitive-demand benchmark. All three smartphones produced significantly higher workload than single-task driving and approached the OSPAN benchmark; Google was less demanding than Apple and Microsoft (which did not differ), with workload differences tracking system errors, completion time, and interface complexity, and residual costs persisting after each interaction.
Key finding
Voice-based interactions with Siri, Google Now, and Cortana while driving produce workload comparable to the high-demand OSPAN task, with Google less demanding than Apple/Microsoft and residual costs persisting after the interaction ends.
Methodology
Two on-road instrumented-vehicle experiments with within-subjects voice-IVIS interactions across three smartphones; primary-task video coding, secondary-task DRT response time, and NASA-TLX, anchored to single-task baseline and OSPAN cognitive referent.
Sample size: Exp 1: N=10; Exp 2: N=20
Quality score: 5 / 5