Content area
Issue Title: Special Issue on Cognitive ergonomics for work, education and everyday life (pp 233-318)
This study assessed whether the increased demand of listening in hearing impaired individuals exacerbates the detrimental impact of auditory distraction on a visual task (useful field of view test), relative to normally hearing listeners. Auditory distraction negatively affects this visual task, which is linked with various driving performance outcomes. Hearing impaired and normally hearing participants performed useful field of view testing with and without a simultaneous listening task. They also undertook a cognitive test battery. For all participants, performing the visual and auditory tasks together reduced performance on each respective test. For a number of subtests, hearing impaired participants showed poorer visual task performance, though not to a statistically significant extent. Hearing impaired participants were significantly poorer at a reading span task than normally hearing participants and tended to score lower on the most visually complex subtest of the visual task in the absence of auditory task engagement. Useful field of view performance is negatively affected by auditory distraction, and hearing loss may present further problems, given the reductions in visual and cognitive task performance suggested in this study. Suggestions are made for future work to extend this study, given the practical importance of the findings.
http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = Cogn Tech Work (2016) 18:393402
DOI 10.1007/s10111-016-0366-2
http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = ORIGINAL ARTICLE
http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10111-016-0366-2&domain=pdf
Web End = The effect of auditory distraction on the useful eld of viewin hearing impaired individuals and its implications for driving
Nicholas C. Herbert1 Nicholas J. Thyer2 Sarah J. Isherwood2
Natasha Merat1
Received: 11 August 2015 / Accepted: 5 February 2016 / Published online: 18 February 2016 Springer-Verlag London 2016
Abstract This study assessed whether the increased demand of listening in hearing impaired individuals exacerbates the detrimental impact of auditory distraction on a visual task (useful eld of view test), relative to normally hearing listeners. Auditory distraction negatively affects this visual task, which is linked with various driving performance outcomes. Hearing impaired and normally hearing participants performed useful eld of view testing with and without a simultaneous listening task. They also undertook a cognitive test battery. For all participants, performing the visual and auditory tasks together reduced performance on each respective test. For a number of subtests, hearing impaired participants showed poorer visual task performance, though not to a statistically signicant extent. Hearing impaired participants were significantly poorer at a reading span task than normally hearing participants and tended to score lower on the most visually complex subtest of the visual task in the absence of auditory task engagement. Useful eld of view performance is negatively affected by auditory distraction, and hearing loss may present further problems, given the reductions in visual and cognitive task performance suggested in this study. Suggestions are made for future work to extend this study, given the practical importance of the ndings.
Keywords Useful eld of view (UFOV) Distraction
Hearing loss Listening effort Attention Driving
1 Introduction
The effect of hearing impairment on driving is an area of research which has received little attention to date, although a small number of studies have begun to suggest that there may be a relationship between the two. A large proportion of this work has not measured driving performance directly, instead focussing on road trafc accident or driving cessation rates for hearing impaired individuals compared to those with normal hearing. The outcomes of these studies are heterogeneous (see Table 1), with some exhibiting an increased risk of negative driving outcomes as a result of hearing loss (Barreto et al. 1997; Ivers et al. 1999; Gilhotra et al. 2001; Picard et al. 2008), but others showing no such association (McCloskey et al. 1994; Sims et al. 2000; Unsworth et al. 2007; Green et al. 2013). Accordingly, there has been no consensus reached over whether hearing loss has an impact on driving safety.
Despite the possibility that hearing loss might affect driving safety, there is little suggestion as to why, exactly, hearing loss affects driving performance. Some authors have suggested that hearing impaired individuals are simply unable to hear salient auditory information in the driving environment (Picard et al. 2008). However, another recent explanation is that hearing impaired individuals may be more susceptible to the effects of auditory distraction whilst driving in comparison with normally hearing individuals (Hickson et al. 2010). It has been shown that the extra cognitive effort required in understanding a distorted auditory signal impacts on operations at later stages of information processing (Rabbitt 1968), and when the source of this auditory distortion is hearing impairment, a negative impact on memory span tasks has been found (Rabbitt 1991; McCoy et al. 2005). This suggests that hearing impairment may affect the performance of tasks
& Nicholas C. Herbert [email protected]
1 Institute for Transport Studies, University of Leeds, Leeds,UK
2 School of Medicine, University of Leeds, Leeds, UK
123
394 Cogn Tech Work (2016) 18:393402
Table 1 An overview of casecontrol studies investigating the relationship between hearing loss and driving outcomes
Study and total sample size
Measure of hearing loss used Driving data collected Outcome
McCloskey et al. (1994)
N = 683
Pure tone audiometry; speech reception thresholds; hearing aid ownership and use
Crash resulting in a medical claim within 7 days of the road trafc accident
Hearing aid owners and users whilst driving more likely to be involved in a crash
Deaths as a result of motor vehicle injuries
Those with hearing loss more at risk of death as a result of road trafc injury
Barreto et al. (1997)
N = 145
Hearing loss (yes or no) obtained from health records, but no denition given
Ivers et al. (1999)
N = 2326
Self-reported degree of hearing difculty
Self-reported road trafc accidents within previous year
Those self-reporting a severe hearing loss more likely to have had a road trafc accident in the previous year
Sims et al. (2000)
N = 174
Self-reported hearing difculty (classied as yes or no) and hearing aid use
Road trafc accidents within the previous 5 years
No difference in risk of road trafc accident as a result of hearing loss or hearing aid use
Self-reported driving cessation Those with a severe self-reported hearing difculty more likely to have ceased driving
Gilhotra et al. (2001)
N = 2831
Self-reported degree of hearing difculty
Unsworth et al. (2007)
N = 538
Self-reported degree of hearing difculty
Self-reported driving cessation or modication to driving behaviour
No change in driving behaviour or extra risk of cessation as a result of self-reported hearing difculty
Picard et al.
(2008)
N = 46,030
Pure tone audiometry. Only cases of normal hearing or noise-induced hearing loss included
Motor vehicle accidents, speeding violations, and all other violations
Risk of accident increased by hearing loss. Those with a hearing loss have a reduced risk of speeding violations, but increased risk of all other violations
Green et al.
(2013)
N = 2000
Self-reported hearing loss (yes or no) History of a motor vehicle collision in the previous 5 years
No increased risk of motor vehicle crashes in the previous 5 years as a result of hearing loss
relying on audition or those performed concurrently with such tasks (e.g. driving whilst conversing with a passenger).
To explore this theory, Hickson et al. (2010) asked normally hearing and hearing impaired individuals to drive a closed-road circuit, undertaking various tasks set up along the course (such as reporting the presence and content of road signs). Whilst driving, participants were asked to concurrently perform a listening task (adding together two aurally presented numbers). The authors hypothesised that hearing impaired individuals would be more affected by this, because of an increase in mental effort associated with the auditory task (Wingeld et al. 2005). They noted that road sign recognition whilst driving was more affected by auditory distraction in their hearing impaired subjects than their normally hearing group. Although it cannot be directly inferred from the results of Hickson et al. (2010), this nding corresponds with research showing an effect of auditory distraction on the useful eld of view: the visual eld area over which information can be acquired in a brief glance without eye or head movements (Edwards et al.
2006, p. 275). This research suggests that the useful eld of view is reduced when a cognitively engaging auditory task is being simultaneously undertaken (Wood et al. 2006) and that this is likely to be more marked the more challenging the auditory task becomes (Pomplun et al. 2001). Because hearing impairment is thought to increase the cognitive demands of listening (Shinn-Cunningham and Best 2008), the current study hypothesised that hearing impaired individuals should experience an even greater reduction in their useful eld of view than normally hearing individuals whilst engaged in an auditory task. Conrmation of this hypothesis would suggest an explanation for the nding of Hickson et al. (2010).
The useful eld of view can be assessed using a computer-based test (UFOV), which measures skills thought to be used during driving (Ball and Owsley 1993). Thus, the assessment has been employed extensively in studies investigating the driving ability of older adults. Evidence has shown that UFOV performance predicts driving competence (Owsley et al. 1998), vehicle crashes (Ball et al. 1993; Owsley et al. 1998), and driver safety (Clay et al.
123
Cogn Tech Work (2016) 18:393402 395
2005). This evidence, and the fact that UFOV can be administered without specialist training in a short amount of time (Classen et al. 2009), has led to suggestions that UFOV may be suitable as a tool to quickly and reliably identify at-risk drivers (Bdard et al. 2008). This suggested predictive nature of UFOV is likely due to its hypothesised reliance on both visual sensory abilities and higher-order attentional skills (Owsley 1994). Indeed, previous work suggests that the functional visual eld is reduced by increases in cognitive load (Rantanen and Goldberg 1999; Williams 1982; Williams and Lefton 1981).
Data presented by Wood et al. (2006) agrees, showing that the concurrent performance of an auditory task resulted in more perceptual errors being made in the visual eld. They argued that their nding had implications for safe driving. Other studies investigating the effect of auditory task engagement on the functional visual eld have produced similar results, which have then been extrapolated to the driving domain (Atchley and Dressel 2004). These two studies suggest that auditory distraction will impact on scores obtained on UFOV, and, by association, lower driving safety and competence under these conditions. The effect of auditory distraction on UFOV is likely to be exacerbated in hearing impaired individuals, as hearing loss places an extra cognitive demand on listening (Shinn-Cunningham and Best 2008), essentially increasing the difculty of the auditory task. However, this is yet to be empirically investigated; Atchley and Dressel (2004) and Wood et al. (2006) only studied individuals with normal hearing.
The proposition that hearing impairment may exacerbate the effects of in-vehicle auditory distraction is pertinent given the increasing complexity of the in-car environment (Hickson et al. 2010). Furthermore, hearing impairment is a prevalent condition, estimated to affect approximately 10 % of the population in Western countries (Arlinger 2003). Therefore, it is important to assess the effect of distraction on tasks relevant to driving in hearing impaired individuals, so that any detrimental effect can be identied and interventions can be devised, should there be a need for them. The aim of this study was to extend the ndings of Wood et al. (2006) by establishing whether hearing loss exacerbates the effect that auditory distraction has on the performance of UFOV, as they found for normally hearing subjects.
2 Materials and methods
2.1 Participants
Sixteen hearing impaired participants were recruited to this study from the Audiology department at Linkping University Hospital. Sixteen individuals reporting normal hearing (matched in terms of age and gender to the hearing
impaired cohort) were recruited from the local community. All participants were in good general health, and free from eye and ear disease. They held a current, valid drivers licence and wore any optical correction normally worn for driving. Pure tone audiometry was conducted on each participant in accordance with the British Society of Audiology guidelines (2011) in order to conrm the presence of hearing loss. Participants were split into two groups: a normal hearing group who all had hearing thresholds better than or equal to 20 dB HL across octave frequencies between 250 and 4 kHz, and a hearing impaired group who did not full this criterion. Hearing thresholds for both groups are given in Fig. 1 and demographic information is given in Table 2. One of the hearing impaired participants had an average hearing threshold of 74 dB HL, and so was classied as having a severe level of hearing loss (British Society of Audiology 2011). Since presentation levels were specied in dB sensation level, this participant was able to hear all auditory stimuli during the experiment. Three of the participants in the hearing loss group had a congenital hearing loss, the other fourteen had an acquired hearing impairment. Fourteen of the group owned bilateral hearing aids, two owned unilateral hearing aids, and only one participant with a hearing loss did not own a hearing aid. Of the sixteen participants who owned hearing aids, eleven wore them all of the time, three wore them occasionally, and two never wore them. Nobody reported differing behaviour with regard to hearing aid use whilst driving.
2.2 Stimuli
2.2.1 Cognitive testing
Given the reliance of UFOV on higher-order processing abilities (Edwards et al. 2006), a cognitive test battery was
Fig. 1 Mean pure tone audiometry thresholds (standard error) for both experimental groups
123
396 Cogn Tech Work (2016) 18:393402
Table 2 Demographic information of the two groups included in this study
Number of participants (males/females)
Mean age (SD) Mean number of years drivers licence owned (SD)
Mean annual mileage (SD)
Hearing impaired group 17 (8 #; 9 $) 57.88 (12.7) 37.94 (13.3) 1505 (640) Normally hearing group 15 (5 #; 10 $) 51.20 (9.3) 34.33 (9.9) 1760 (1110)
employed. KIPS, a developed, abbreviated version of the cognitive test battery TIPS (Lyxell et al. 1998), was used in order to assess working memory capacity, lexical access speed, and phonological skills (Borg et al. 2008). The test was administered on a personal computer, requiring responses using the mouse and/or keyboard. The test battery consisted of four sections and lasted approximately 17 min in total:
1. Physical matching: participants had to decide whether two letters appearing on the monitor looked the same or different.
2. Lexical text: participants had to decide whether words that appeared on the screen one at a time were real or invented.
3. Rhyme: participants had to decide whether two words displayed simultaneously on the monitor rhymed with each other or not.
4. Reading span: sets of two 35 word sentences were displayed on the monitor one word at a time. Participants had to decide whether or not each sentence made sense or was nonsense. Once this choice had been made, the participant was asked to recall either the rst or the last word in both preceding sentences.
2.2.2 Visual task
UFOV test software (v6.1.1, Visual Awareness Research Group Inc.) was used to assess the useful eld of view of participants. This software consists of three screening subtests, each of which was used in this study. Each subtest consisted of a number of trials beginning with an empty screen with a central outline of a white square subtending approximately 3.5 at the eye. Visual targets were then presented followed by a noise masking screen, and then nally the response screen(s) (see Fig. 2). The central task was to specify which of two pictures (always a car or a truck) ashed up in the centre square, and the peripheral task was to indicate at which of eight possible locations a picture (always of a car), at a visual eccentricity of approximately 29, was presented. Each subtest varied slightly in terms of the targets presented:
1. Subtest 1, central task: perform the central task alone (no peripheral target is presented).
2. Subtest 2, central and peripheral task: perform the central and the peripheral task simultaneously.
3. Subtest 3, central and peripheral task with visual distracters: the same stimuli as subtest 2; however, a distracter array of 47 triangles was presented simultaneously with the stimuli (see Fig. 2).
Each subtest consisted of a variable number of trials (range 1351) as the number of presentations was controlled by participant consistency. The epoch of stimulus presentation also varied between trials (range 17500 ms), again depending on the accuracy of responses given by respective participants (see Sect. 2.4.2 for details). Participants were automatically presented with a response screen following each stimulus (see Fig. 2). Responses were made with a computer mouse by navigating to their chosen answer shown on the screen and left-clicking. This method of response has been shown in past research to have a high testretest repeatability of .884 (Edwards et al. 2005).
2.2.3 Auditory task
The auditory distraction task used in this study was a dichotic listening test developed by Hallgren et al. (1998), which consisted of two ve-word, low-redundancy, sentences being played to opposing ears simultaneously. This task was chosen because such a dichotic listening task that requires a response has been shown to affect the span of visual search (Wood et al. 2006). Furthermore, the test provides a level of face validity for driving under certain circumstances (e.g. conversing with a passenger whilst listening to a radio programme), and its temporal properties make it ideal for coinciding auditory stimuli with UFOV stimuli. The auditory stimuli were presented using Telephonics TDH-39P headphones. Subjects were required to listen to the stimuli in full before repeating back as much of both sentences as they had heard. As these stimuli were being presented through headphones, participants were not permitted to wear hearing aids during the experiment, even if they did so under normal driving conditions. For this reason, auditory stimuli were presented at a level of 50 dB HL sensation level, so that sounds were played at an audible level for all participants, regardless of hearing loss. To a certain extent, this approach emulates the primary goal of hearing aids (Hogan and Turner 1998). In cases
123
Cogn Tech Work (2016) 18:393402 397
Fig. 2 Presentation sequence of one trial in the UFOV software, in this case from subtest 3 (incorporating a central and peripheral target with a distracter array)
where the extent of hearing loss made this sensation level uncomfortably loud, stimuli were adjusted to an intensity deemed comfortable by participants.
2.3 Procedure
Audiological testing and the cognitive test battery were both undertaken prior to starting UFOV testing. Participants were then seated 60 cm away from the 17 in. UFOV computer monitor. Participants were instructed how to perform UFOV with the aid of sample stimuli contained within the software and were then given a practice as per the test instructions. Practice continued until 75 % of trials were correctly performed, or until 16 trials had been presented. Participants were then given the opportunity to practise UFOV simultaneously with the auditory task, using the same stimuli as in the experimental session. Again practice was stopped once 75 % of trials had been successfully completed or 16 trials had been presented. Following the training and practice session, participants went on to complete the three UFOV subtests described above both with and without the auditory task presented simultaneously. This resulted in six experimental conditions, which were partially counterbalanced using the balanced Latin square method. A baseline measure of auditory task performance on its own
was also taken whereby participants responded to ten auditory stimuli. Half of each experimental group undertook this baseline measure before performing the six experimental conditions, the other half performed it at the end of the experimental session.
2.4 Measures
2.4.1 Cognitive testing
The KIPS software measured participant performance (percentage correct) on each of the cognitive battery sub-tests (as described in Sect. 2.2.1). Each individual section of the test battery could be analysed independently.
2.4.2 Visual task
Visual task scores ranged between 17 and 500 ms. Scores were given as the stimulus epoch required to achieve 75 % successful performance of UFOV trials. Therefore, a lower score meant better subtest performance. The UFOV software derived visual task scores by varying the stimulus presentation duration depending upon correct/incorrect responses, presenting stimuli using a double-staircase method. Subtests ended automatically once the software had a stable estimate of the required stimulus epoch.
123
398 Cogn Tech Work (2016) 18:393402
2.4.3 Auditory task
Auditory task responses were also recorded when present in the experimental condition, and during the baseline measure of auditory task performance. A percentage correct score was calculated for each participant during each subtest by counting the number of correct words repeated following each stimulus presentation. As there were ve words in each sentence, the maximum score for each stimulus was ten. A similar approach to marking this auditory task has been taken in past research, which asked participants to report the sentence from one ear only (Hallgren et al. 2001). However, the current study asked participants to recall as much of both sentences as possible. This approach was taken in order to avoid the possibility of cueing participants towards a certain side of their visual eld as a result of directed auditory stimuli (Ho et al. 2006). Accordingly, sentences were analysed such that if a participant only gave one sentence as a response, marks were not awarded for words from both stimuli sentences. Instead, marks were only given from the sentence that scored highest. For example:
Stimuli: Elsa borrowed three dark gloves and Bosse owned six beautiful rings.
Response 1: Bosse owned six beautiful rings and *no response*. Score given = 5/10.
Response 2: Bosse borrowed three beautiful rings and *no response*. Score given = 3/10.
Response 1 scores 5/10, as the participant has repeated only one sentence in its entirety, but the sentence given is correct. Response 2 only scores 3/10, as the participant has only repeated one sentence, and the answer given is a mixture of the two stimuli sentences. The 3 marks given are, therefore, for the participant saying Bosse, beautiful and rings, all three of which are present in the second stimuli sentence. Marks are awarded from this particular sentence as the responses recorded to the other stimulus sentence would have resulted in a lower score of 2/10. This approach was taken in order to reect the difculty of the listening task. It was not considered feasible to give the same mark to somebody repeating a mixture of the two stimuli sentences, and another person successfully ignoring an interfering stimulus, listening to one of the sentences, and repeating it in its entirety.
3 Results
The mean scores obtained by both experimental groups on the KIPS cognitive test battery are shown in Fig. 3. Performance on the majority of these tests was accurate, with mean scores of around 90 % or above. However, reading span scores were generally lower than the other three
sections for participants of both groups. In fact, there was a signicant difference in scores on the reading span section of the KIPS test battery between the hearing impaired group (M = 50.94, SEM = 2.57) and the normally hearing group (M = 61.07, SEM = 2.53); t(30) = -2.793, p = .009, r = .45. The hearing impaired group also scored lower on average for the rhyme section of the KIPS test than did the normally hearing group, though this difference was not signicant. Performance on the other two test sections was very similar between the groups.
The mean UFOV test scores for both groups in each individual experimental condition are shown in Fig. 4. A two-way repeated-measures ANOVA with two within-subjects factors (UFOV subtest and auditory task presence) and one between-subject factor (hearing loss presence) indicated a main effect of auditory task presence [F(1,30) = 24.733, p \ .001, g2 = .452] and UFOV sub-test [F(2,60) = 75.265, p \ .001, g2 = .715]. UFOV test scores became poorer for both groups when the auditory task was performed simultaneously, and when more visual stimuli were added to the UFOV subtests. Although the interaction between UFOV subtest and auditory task presence tended towards signicance, no effect was found [F(2,60) = 2.946, p = .060, g2 = .089]. The data show a trend for UFOV test scores to be poorer for the hearing impaired group than they were for their normally hearing counterparts when the test was performed simultaneously with an auditory task (see Fig. 4). Despite this trend, no statistically signicant interaction between hearing loss presence and performance was found. Measures of baseline performance on UFOV without the presence of an auditory distracter were not signicantly different, although for subtest 3 (incorporating the distracter array of 47 triangles) a lower average score was noted for the hearing impaired participants (181.3 ms) than for the normally hearing group (133.4 ms), though this was not signicant; t(30) = 1.37,
Fig. 3 Mean scores (standard error) on each individual section of the KIPS cognitive test battery
123
Cogn Tech Work (2016) 18:393402 399
Fig. 4 Mean scores for each UFOV subtest (standard error). Scores are shown for the test being performed in isolation (black symbols) and with the auditory task present (white symbols)
Fig. 5 Mean percentage of auditory responses (standard error) that were correct during each UFOV subtest and with the auditory task performed in isolation (baseline condition)
p = .183. When controlling for the effects of cognitive differences in the participants of each group (assessed by KIPS test battery scores), there remained no signicant interaction between UFOV performance and hearing loss presence.
A comparison of the change in UFOV scores as a result of auditory task engagement between normally hearing and hearing impaired individuals showed a marginal, but not signicant, difference between the two groups. There was a worse performance decrement for hearing impaired participants when they were required to complete subtest 2 of UFOV in the presence of an auditory task.
The mean percentage scores for the auditory task performed as a baseline measure and during each UFOV subtest are shown in Fig. 5. A main effect of the UFOV subtest on the accuracy of auditory responses was noted [F(1.709,49.574) = 7.378, p = .03, g2 = .203]. Mauchlys test indicated that the assumption of sphericity had been violated, v2 (5) = 36.873, p [ .001; hence, the
Greenhouse-Geisser correction was used. Contrast analysis indicated an increasing linear effect of visual task complexity [F(1,29) = 14.57, p = .001, g2 = .045], such that the accuracy of auditory responses was progressively reduced with increasing visual task demand. Although under each visual condition the mean score of the hearing impaired group was lower than that of the normally hearing group, no main effect of hearing loss presence was observed.
4 Discussion
The aim of this study was to assess the effect of auditory distraction on the performance of a complex visual task, and to test whether this effect was more pronounced in hearing impaired compared to normally hearing participants. The results show that the performance of a simultaneous auditory task degrades performance on UFOV and that the more complex the visual task becomes, the greater the effect of the auditory distracter. These ndings concur
with those of Wood et al. (2006) who found that the simultaneous performance of an auditory task signicantly reduced performance on a visual task analogous to UFOV, particularly when the visual task incorporated an array of visual distracters. Wood et al. (2006) argue that results such as these are of great practical importance for driving as they may be suggestive of poorer hazard and sign detection and loss of vehicle control during periods of auditory engagement. This studys results advocate that auditory task engagement whilst driving may well decrease road safety. It has been shown here that auditory distraction reduces scores for a test on which poor performance has been linked with various driving performance measures. The concurrence of these results with past research suggests that caution should be exercised with regard to complex auditory task engagement whilst driving. Indeed, given the increasing availability and use of in-car systems which function using the auditory modality, these ndings are of clear practical importance. This data seems to suggest that these types of device may have an adverse effect on driving ability, if they are actively engaged with whilst on the road.
In terms of the exacerbating effect of hearing loss, the effect of the concurrent auditory task was the same for both groups. This was not likely to be due to a withdrawal from the auditory task in favour of the visual task; although visual task complexity affected auditory task scores, this effect was the same for both experimental groups. Concurrent engagement in the auditory task did not result in any signicant differences in UFOV performance between the two groups in this study. These results are in contrast to past research, which has found a disproportionate effect of auditory task engagement in those with hearing loss (Hick and Tharpe 2002; Hickson et al. 2010). However, since there was a trend for worse UFOV performance by the hearing impaired participants in this study (even without an auditory task), the results warrant highlighting, because it
123
400 Cogn Tech Work (2016) 18:393402
is possible that this group nd visually complex environments challenging in both dual- and single-task conditions.
The implications of complex in-car systems operating in the auditory modality have already been discussed above. In concurrence with Hickson et al. (2010), we suggest that it could be particularly problematic for hearing impaired drivers to engage with multiple cognitively demanding tasks in the presence of such in-vehicle systems. Further research to identify and test alternative solutions for this population is therefore warranted.
Hickson et al. (2010) noted that the degree of hearing impairment was the best predictor of overall driving ability in their study sample. Their results also suggested that mild hearing impairment is not associated with poorer driving ability in the presence of distracters. This may have been a possible reason why a statistically signicant difference was not observed between the performances of the two groups in this study. Nearly half (8 out of 17) of the hearing loss group in this study had an impairment classied as mild, leaving relatively little data from those with a moderate (8 out of 17) and severe (1 out of 17) hearing impairment. Further research which examines how different levels of hearing loss affect UFOV performance in the presence of auditory distractors would therefore be of value.
An interesting trend identied in this study is the pattern of results for UFOV involving no simultaneous auditory task. A lower baseline score on subtest 3 of UFOV (incorporating visual, but no auditory distracters) was noted in the hearing impaired group compared to the normally hearing participants. This was unexpected, as it was hypothesised that extra attention to the auditory task would bring about a disturbance on UFOV in the hearing impaired sample. A lower score on a particular UFOV subtest in the absence of any auditory information cannot, therefore, be explained by this hypothesis. Interestingly, a similar phenomenon whereby hearing impaired individuals appear to be more distracted by visual information has been noted in past research (Thorslund et al. 2013a). Furthermore, the data presented by Hickson et al. (2010) suggests that visual distraction had a negative inuence equal to that of auditory distraction on overall measured driving performance in their hearing impaired participants. However, there is no forthcoming explanation as to why this may have been the case.
Although the groups in the current study were closely matched in terms of age, the number of years of formal education undertaken, driving experience, and a number of the cognitive tests undertaken, there was a signicant difference found between the two groups in the reading span section of KIPS. Importantly, this effect was found in the absence of any auditory information, thus suggesting that it is related to a general inability to process complex
information efciently, rather than it stemming from a distortion of auditory information at the periphery. This nding may explain the discrepancy in UFOV subtest 3 performance between the two groups in the absence of a simultaneous auditory task. It should be noted, however, that the cognitive test battery undertaken here was not entirely diagnostic with regard to visual attention breadth, auditory processing of language, or the production of language. Indeed, this was not a general cognitive test per se, though the reading span subtest is considered to predict performance on higher-order cognitive tasks (Engle 2002). It may be wise to incorporate alternative or supplementary measures of complex working memory span into future work (see e.g. Conway et al. 2005).
It should be noted that the results of this study do not imply that those with a hearing impairment are more at risk of vehicular crashes. However, this experiment has suggested that those with a hearing impairment may be slower to react to visual information whilst performing a cognitively demanding auditory task, or during periods where the visual scene is very cluttered. This has ramications for driving in terms of failures of visual attention under such circumstances. During periods of auditory task performance, hazard perception, for instance, may be suboptimal, leading to an increased risk of road trafc accidents, although this was not directly tested in our study. However, those who have issues with multi-tasking may well adapt their behaviour and withdraw from the auditory task in order to increase their road safety. Indeed, Thorslund et al. (2013a, b, 2014) have repeatedly observed, what they argue, is an adaptive driving style in their hearing impaired participants. Furthermore, there is a possibility that hearing impaired drivers may compensate for any negative effects of their hearing loss with more developed visual skills (e.g. Mitchell and Maslin 2007). This appears particularly pertinent for profoundly deaf individuals, as there has been some suggestion that their visual reactions, particularly in the periphery of vision, are quicker than those of normally hearing people (Bavelier et al. 2000; Loke and Song 1991). Studies of the real-world driving of hearing impaired individuals would, therefore, be of great value in determining whether these adaptations in behaviour are likely to be the case.
5 Conclusions
This study has shown that the simultaneous performance of a cognitively demanding auditory task and UFOV decreases performance on both respective tasks. These results are of great practical importance, as they may indicate that aspects of visual attention, related to driving performance, are compromised during periods of auditory engagement.
123
Cogn Tech Work (2016) 18:393402 401
The results may be more applicable to hearing impaired individuals, given that they recorded marginally worse performance across a number of UFOV subtests when undertaking a concurrent auditory task. It should also be noted that those with a hearing impairment showed a nonsignicant tendency to perform worse on subtest 3 of UFOV, even in the absence of any auditory information. This suggests that those with a hearing loss might be less able to perform complex visual tasks efciently, even when there is no inuence of sound present. Additionally, cognitive testing suggested a decrease in the working memory capabilities of hearing impaired individuals. These results bring to light interesting questions about the information processing abilities of those with a hearing impairment. Further research in this area is required in order to improve our understanding of the effect of hearing impairment on dual-task execution and its potential supplementary effect on driving competence.
Acknowledgments Support from the Higher Education Funding Council for Englands Higher Education Innovation Funding and the Economic and Social Research Council is gratefully acknowledged. The authors would also like to thank VTI, Linkping and Linkping University Hospital, Sweden, and in particular Mrs. Birgitta Thor-slund, for their help with the recruitment for, and running of, this experiment.
References
Arlinger S (2003) Negative consequences of uncorrected hearing lossa review. Int J Audiol 42(suppl 2):S17S20
Atchley P, Dressel J (2004) Conversation limits the functional eld of view. Hum Factors 46(4):664673
Ball K, Owsley C (1993) The useful eld of view test: a new technique for evaluating age-related declines in visual function. J Am Optom Assoc 64(1):7179
Ball K, Owsley C, Sloane ME, Roenker DL, Bruni JR (1993) Visual-attention problems as a predictor of vehicle crashes in older drivers. Invest Ophthalmol Vis Sci 34(11):31103123
Barreto SM, Swerdlow AJ, Smith PG, Higgins CD (1997) Risk of death from motor-vehicle injury in Brazilian steelworkers: a nested case-control study. Int J Epidemiol 26(4):814821
Bavelier D, Tomann A, Hutton C, Mitchell T, Corina D, Liu G,
Neville H (2000) Visual attention to the periphery is enhanced in congenitally deaf individuals. J Neurosci 20(17):RC93Bdard M, Weaver B, Darzins P, Porter MM (2008) Predicting driving performance in older adults: we are not there yet! Trafc Inj Prev 9(4):336341Borg E, Bergkvist C, Olsson IS, Wikstrom C, Borg B (2008)
Communication as an ecological system. Int J Audiol 47(suppl2):S131S138British Society of Audiology (2011) Recommended procedure: pure-tone air-conduction and bone-conduction threshold with and without masking. British Society of Audiology, Reading Classen S, Mccarthy DP, Shechtman O, Awadzi KD, Lanford DN,
Okun MS, Rodriguez RL, Romrell J, Bridges S, Kluger B, Fernandez HH (2009) Useful eld of view as a reliable screening measure of driving performance in people with
Parkinsons disease: results of a pilot study. Trafc Inj Prev 10(6):593598Clay OJ, Wadley VG, Edwards JD, Roth DL, Roenker DL, Ball KK
(2005) Cumulative meta-analysis of the relationship between useful eld of view and driving performance in older adults: current and future implications. Optom Vis Sci 82(8):724731 Conway ARA, Kane MJ, Bunting MF, Zach Hambrick D, Wilheim O,
Engle RW (2005) Working memory span tasks: a methodological review and users guide. Psychon Bull Rev 12(5):769786 Edwards JD, Vance DE, Wadley VG, Cissell GM, Roenker D, Ball
KK (2005) Reliability and validity of useful eld of view test scores as administered by personal computer. J Clin Exp Neuropsychol 27(5):529543Edwards JD, Ross LA, Wadley VG, Clay OJ, Crowe M, Roenker DL,
Ball KK (2006) The useful eld of view test: normative data for older adults. Arch Clin Neuropsychol 21(4):275286Engle RW (2002) Working memory capacity as executive attention.
Curr Dir Psychol Sci 11(1):1923Gilhotra JS, Mitchell P, Ivers R, Cumming RG (2001) Impaired vision and other factors associated with driving cessation in the elderly: the blue mountains eye study. Clin Experiment Ophthalmol 29(3):104107Green KA, McGwin G, Owsley C (2013) Associations between visual, hearing, and dual sensory impairments and history of motor vehicle collision involvement of older drivers. J Am Geriatr Soc 61(2):252257Hallgren M, Johansson M, Larsby B, Arlinger S (1998) Dichotic speech tests. Scand Audiol 27(suppl 49):3539Hallgren M, Larsby B, Lyxell B, Arlinger S (2001) Cognitive effects in dichotic speech testing in elderly persons. Ear Hear 22(2):120129Hick CB, Tharpe AM (2002) Listening effort and fatigue in school-age children with and without hearing loss. J Speech Lang Hear Res 45(3):573584Hickson L, Wood J, Chaparro A, Lacherez P, Marszalek R (2010)
Hearing impairment affects older peoples ability to drive in the presence of distracters. J Am Geriatr Soc 58(6):10971103Ho C, Tan HZ, Spence C (2006) The differential effect of vibrotactile and auditory cues on visual spatial attention. Ergonomics 49(7):724738Hogan CA, Turner CW (1998) High-frequency audibility: benets for hearing-impaired listeners. J Acoust Soc Am 104(1):432441 Ivers RQ, Mitchell P, Cumming RG (1999) Sensory impairment and driving: the blue mountains eye study. Am J Public Health 89(1):8587Loke WH, Song SR (1991) Central and peripheral visual processing in hearing and nonhearing individuals. Bull Psychon Soc 29(5):437440Lyxell B, Andersson J, Andersson U, Arlinger S, Bredberg G, Harder
H (1998) Phonological representation and speech understanding with cochlear implants in deafened adults. Scand J Psychol 39(3):175179McCloskey LW, Koepsell TD, Wolf ME, Buchner DM (1994) Motor-vehicle collision injuries and sensory impairments of older drivers. Age Ageing 23(4):267273Mccoy SL, Tun PA, Cox LC, Colangelo ML, Stewart RA, Wingeld
A (2005) Hearing loss and perceptual effort: downstream effects on older adults memory for speech. Q J Exp Psychol A 58(1):2233Mitchell TV, Maslin MT (2007) How vision matters for individuals with hearing loss. Int J Audiol 46(9):500511Owsley C (1994) Vision and driving in the elderly. Optom Vis Sci
71(12):727735Owsley C, Ball K, Mcgwin G, Sloane ME, Roenker DL, White MF,
Overley ET (1998) Visual processing impairment and risk of
123
402 Cogn Tech Work (2016) 18:393402
motor vehicle crash among older adults. JAMA, J Am Med Assoc 279(14):10831088Picard M, Girard SA, Courteau M, Leroux T, Larocque R, Turcotte F,
Lavoie M, Simard M (2008) Could driving safety be compromised by noise exposure at work and noise-induced hearing loss? Trafc Inj Prev 9(5):489499Pomplun M, Reingold EM, Shen JY (2001) Investigating the visual span in comparative search: the effects of task difculty and divided attention. Cognition 81(2):B57B67Rabbitt PMA (1968) Channel-capacity intelligibility and immediate memory. Q J Exp Psychol 20(3):241248Rabbitt PMA (1991) Mild hearing-loss can cause apparent memory failures which increase with age and reduce with IQ. Acta Otolaryngol 476:167176Rantanen EM, Goldberg JH (1999) The effect of mental workload on the visual eld size and shape. Ergonomics 42(6):816834 Shinn-Cunningham BG, Best V (2008) Selective attention in normal and impaired hearing. Trends Amplif 12(4):283299Sims RV, McGwin G Jr, Allman RM, Ball K, Owsley C (2000)
Exploratory study of incident vehicle crashes among older drivers. J Gerontol Biol Sci Med Sci 55(1):M22M27 Thorslund B, Peters B, Herbert N, Holmqvist K, Lidestam B, Black
A, Lyxell B (2013a) Hearing loss and a supportive tactile signal in a navigation system: effects on driving behavior and eye movements. J Eye Mov Res 5(6):19
Thorslund B, Peters B, Lidestam B, Lyxell B (2013b) Cognitive workload and driving behavior in persons with hearing loss. Transp Res Part F Trafc Psychol Behav 21:113121
Thorslund B, Ahlstrm C, Peters B, Eriksson O, Lidestam B, Lyxell B
(2014) Cognitive workload and visual behavior in elderly drivers with hearing loss. Eur Transp Res Rev 6:377385Unsworth CA, Wells Y, Browning C, Thomas SA, Kendig H (2007) To continue, modify or relinquish driving: ndings from a longitudinal study of healthy ageing. Gerontology 53(6):423431Williams LJ (1982) Cognitive load and the functional eld of view.
Hum Factors 24(6):683692Williams LJ, Lefton LA (1981) Processing of alphabetic information presented in the fovea or the peripheryfunctional visual-eld and cognitive load. Perception 10(6):645650Wingeld A, Tun PA, McCoy SL (2005) Hearing loss in older adulthood: what it is and how it interacts with cognitive performance. Curr Dir Psychol Sci 14(3):144148Wood J, Chaparro A, Hickson L, Thyer N, Carter P, Hancock J, Hoe
A, Le I, Sahetapy L, Ybarzabal F (2006) The effect of auditory and visual distracters on the useful eld of view: implications for the driving task. Invest Ophthalmol Vis Sci 47(10):46464650
123
Springer-Verlag London 2016