Effects of Conversation on Situation Awareness and Working Memory in Simulated Driving

Abstract

OBJECTIVE: In the present research, we investigated the hypothesis that working memory mediates conversation-induced impairment of situation awareness (SA) while driving. BACKGROUND: Although there is empirical evidence that conversation impairs driving performance, the cognitive mechanisms that mediate this relationship remain underspecified. Researchers have reported that a phonological working memory task decreased drivers' SA for vehicles located behind them whereas a visuospatial working memory task impaired SA for vehicles ahead. Conversation, therefore, might impair SA for vehicles behind the driver by preferentially taxing the phonological loop. METHOD: A 20-questions task was used as a proxy for natural conversation. In Experiment I, driving performance was measured across three within-subjects conversation conditions (i.e., no conversation, driver asks questions, driver answers questions) with the use of a driving simulator. In Experiment 2, participants drove in the same simulator while either conversing (20-questions task) or not Participants estimated the positions of other vehicles after the screens were blanked at the end of each trial. RESULTS: Speed monitoring and responses to visual probes were impaired by the 20-questions conversation task (Experiment 1). As predicted, conversation impaired SA for the location of other vehicles more for vehicles located behind the driver than for those in front (Experiment 2). CONCLUSION: Conversation impairs drivers' SA of vehicles behind them by taxing working memory's phonological loop and impairs SA generally by taxing working memory's central executive. APPLICATION: Provides a theoretical framework that links driver SA to working memory and a mechanism for understanding why conversation impairs driving performance.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 60 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.