The impact of hands-free message reception/response on driving task performance

https://doi.org/10.1016/S0001-4575(01)00083-5Get rights and content

Abstract

A series of closed-course driving experiments were conducted in which 41 drivers ranging in age from 19 to 70 were put through a series of increasingly challenging driving performance tasks both in the presence and absence of audible messages. The messages required specific responses and these, along with driving performance measures based on driver/vehicle response characteristics, were recorded. The results clearly showed a negative impact of the message task on driver decision-making performance when this involved the more complex tasks of weaving and especially left-turning. Such decision-making decrements in the presence of the messages were exacerbated by adverse pavement surface conditions.

Introduction

Various models of driving behavior have been proposed which attempt to explain crash risk and causation (Ranney, 1994). Many of these have focused on individual abilities having to do with visual searching and cognitive processing. Complex task performance, such as driving, requires rapid switching of attention and measures of individual attention-switching ability have been found to be effective predictors of accident involvement (Ranney, 1994). Both visual attention and selective attention have been found to be more associated with crash risk in older than in younger drivers (Ranney et al., 2000).

Much of driving is characterized by basic vehicle control activities (such as maintaining appropriate lane position and headway within surrounding traffic) which, with experience, take up very little mental processing and allow some attention to be diverted to other tasks (such as reading signs or talking on a telephone). Such “automatic” processing is learned “skill-based” behavior, but in unfamiliar situations drivers may resort to rule-based behavior or, where no applicable rules are apparent, less predictable knowledge-based processing will have to occur. Unanticipated events juxtaposed with driver inattention, can cause distraction-related crashes (Ranney et al., 2000).

The concept of a control-based hierarchy for decision-making coupled with recognition of the need for attention-switching has led to what has been called the Multiple Resource Theory (Lee et al., 2000). Under this theory each driver has access to supplies of cognitive resources in several areas including modality (e.g. auditory, visual, etc.) and processing code (e.g. spacial, verbal, etc.). In a dual task situation requiring selective attention interference will be more probable if the individual tasks access the same pool of processing resources (Ranney, 1994, Lee et al., 2000). Such “time-sharing” of cognitive resources must occur whenever a secondary task is performed concurrently with a primary task (Davies et al., 1984) and studies have shown that time-sharing involving similar sources of processing capacity in working memory may impact older persons more than the young (Davies et al., 1984).

Distraction of drivers by secondary tasks or events can result in general withdrawal of attention (e.g. eye glances away from road) or in selective withdrawal of attention in which “automatic” behaviors (such as those involved in vehicle speed control and lane position), are maintained but event detection is degraded (Tijerina, 2000). Selective withdrawal of attention to driving (as the primary task) would result from attention to thoughts or verbal communication. Under such conditions drivers may selectively filter information based on expectations rather than actual situations (Tijerina, 2000). But, overall, the degree to which the primary driving task may be expected to degrade due to performance of a secondary task is a function of the attentional demands required in each task (Ranney et al., 2000, Lee et al., 2000). So one could expect that a situation requiring controlled communication by subjects while driving would interfere with driving less in situations involving “automatic” behaviors and more when higher levels of choice and decision-making are required.

Also, in situations of consistent driving task characteristics, the impact of such things as phone use should be related to the type of communication/thought process involved. In particular, interference caused by a secondary task seems to be greater if both share the same type of content (i.e. if both tasks involve imagery-based information versus one involving imagery-based information and the other abstract-based information). This notion was confirmed by an on-road study that found that eye movements executed while driving were more restricted in range horizontally and vertically if drivers responded to a concurrent auditory task involving spatial/imagery content than one involving verbal/abstract content (Recarte and Nunes, 2000). Thus, spatial/imagery phone messages should have a greater impact on driving performance than verbal/abstract phone messages.

The Insurance Corporation set out in 1999 to test these expectations. The impetus for such research came from recent reports (Redelmeier and Tibshirani, 1997, Cain and Burris, 1999, Laberge-Nadeau et al., 1998) which have linked driver distraction resulting from in-vehicle phone use to increased crash risk. Further, it has been suggested that the cognitive distraction of conversations may be at least as important in crash causation as the biomechanical distraction of operating the phone unit (Ranney et al., 2000, Redelmeier and Tibshirani, 1997).

Numerous studies have been conducted in order to investigate the effects of telephone and speech-based information interfaces on driving. Most studies report significant effects of the conversation task on some aspects of driving performance under certain conditions but the results are somewhat variable depending on the specific tasks and driving situations enacted. For example, Vollrath and Totzke (2000) concluded that acoustic information processing does not interfere with driving and Noy et al. (1999) concluded that their results could not support the hypothesis that cell phone use impedes safe driving. Other studies (such as Lee et al., 2000) have concluded that speech-based interfaces are distracting to drivers in a manner which could compromise safety. In between these extremes lie the majority of studies which demonstrate some impact of phone use on such things as reaction to speed variations in a preceding vehicle (headway maintenance) but not much effect on measures such as lateral positioning and not much differential effect in relation to driving task complexity (for example, Parkes and Hooijmeijer, 2000, Alm and Nilsson, 1994, Brookhuis et al., 1991). Alm and Nilsson (1994) speculated that in their “hard” driving task, subjects may have assigned a low status to the telephone task and “not allowed” it to influence their behavior to any great extent. Another possibility though, is that the higher-difficulty driving tasks administered in these simulator or closed-course driving experiments do not usually come close enough to the unexpected situations, control decisions or critical choices which substantially tax the brain’s time-sharing capacity and disrupt automatic processing (Ranney, 1994).

One such situation is the “unprotected” intersection cross-traffic turn task (the left-turn in North America) which has been described as one of the most hazardous maneuvers in driving (Hancock et al., 1991) and especially for older drivers, with whom we associate attention and information processing decrements (Preusser et al., 1998, Matthias et al., 1996). Drivers must perceive and integrate multiple information sources before making a decision (Hancock et al., 1991). This information relates to oncoming vehicle gaps and speeds, but may be modified by weather conditions, size/type of oncoming vehicle and turning driver characteristics such as age and decision time (Andrea et al., 1999, Staplin, 1995, Hancock et al., 1991). Nevertheless, it may also be true that older drivers often compensate to a certain extent (although evidently not fully) for decreased information processing ability by making more conservative gap acceptance decisions than younger drivers (Hancock and Caird, 1993) The consistent degradation of left-turning safety and efficiency with age has reinforced the notion that complex driving tasks such as this put strain on individuals’ information processing skills and cognitive abilities which are components of “working memory” discussed earlier (Guerrier et al., 1999).

In summary, we may expect that cell phone use should affect drivers more in situations requiring a high level of information processing and decision-making (such as left-turns) and less in situations where choices are simpler (such as stopping at a signal light). We may also anticipate some impact of message type and some differential effect on driving tasks performance by subject age.

Section snippets

Overview

Forty-one subjects (volunteers from a larger group of respondents to a telephone survey of cell phone users) participated in a series of closed-course driving tasks. All subjects were required to be both experienced drivers (more than 3 years) and experienced cell phone users. Of the 41, 30 were male and 11 female, 7 were aged 19–24 (x̄=21.6), 25 were aged 25–44 (x̄=35.3), and 9 were aged 45–70 (x̄=60.0). Previous research (Cooper, 1990, for example) has shown that relative crash risk increases

Traffic signal task

In this there were two distinct driver actions possible: stop or go through. Each ended up being represented by a substantial number of occurrences and thus it was considered more appropriate to analyze the results separately. So the analysis was broken down by long- versus short-trigger, vehicle stopped versus vehicle ran light, and message versus no-message. Only 28% of subjects ran the long-trigger light while 45% ran the short-trigger light.

First, the data were examined as a function of

Discussion

The results from all three driving tasks gave no substantive evidence of subject “learning” adjustments to the tasks over repeated exposure. There was also no indication that listening/responding to the messages had any meaningful differential impact in this regard.

The overall effect of the messages in the case of the traffic signal task was to produce a more conservative response on the part of subject drivers. With the message/response task activated, the drivers were more likely to stop (as

Conclusions

Listening and responding to relatively complex messages, as might occur when using a hands-free cellular telephone to conduct business or deal with important domestic issues, was found to significantly degrade driving performance in a series of driving tasks. The extent to which this degradation occurred seemed connected to the complexity of the driving maneuver. Commonly encountered traffic signal-related choices tended to elicit conservative decision-making in the presence of the messages but

Acknowledgements

The research study reported in this paper was funded by the Insurance Corporation of British Columbia. Test set-up and data collection were performed by MacInnis Engineering Associates Ltd., with subject recruitment and on-site coordination by McIntyre and Mustel Research Ltd. The test facility was made available by the Pacific Traffic Education Centre. The authors would especially like to thank Linda Andersen of ICBC for her invaluable assistance in dealing with the many administrative aspects

References (25)

  • Davies, D.R., Jones, D.M., Taylor, A., 1984. Selective and Sustained Attention Tasks: Individual and Group Differences....
  • Hancock, P.A., Caird, J.K., 1993. The Left-Turn. Human Factors Research Laboratory, University of...
  • Cited by (60)

    • Ecological interface design effectively reduces cognitive workload – The example of HMIs for speed control

      2020, Transportation Research Part F: Traffic Psychology and Behaviour
      Citation Excerpt :

      For example, cognitively loaded drivers rely on this fallback level and would most likely brake sufficiently before an obstacle because looming effects automatically trigger a braking reaction (Franconeri & Simons, 2003). However, cognitively loaded drivers might not be able to properly adapt to predictive cues, such as road signs and roadside instructions (Baumann, Petzoldt, Hogema, & Krems, 2008; Engström & Markkula, 2007) or dangerous road conditions (Cooper et al., 2003). Moreover, capture errors, as described by Norman (1981), might happen more often to cognitively loaded drivers, when a partially familiar situation unconsciously activates an inappropriate automatized behavior.

    • Mobile phone use while driving-literary review

      2017, Transportation Research Part F: Traffic Psychology and Behaviour
      Citation Excerpt :

      Strayer and Johnston (2001), found out that mobile phone use while driving has an influence on driver's reaction time, i.e. reaction time increases with a degree of complexity of a traffic situation. This is also supported by the results obtained by Cooper et al. (2003), who showed that reading and sending text messages (sms), while driving may lead to a significant reduction in driving performance. Moreover, this reduction degree is linked with the complexity of activity that is being performed (turning left/stopping at the traffic lights).

    View all citing articles on Scopus
    View full text