Key Points
Question
Do auditory reminder cues promote proactive scanning (making an early large scan to the blind side) by individuals with homonymous hemianopia when approaching intersections in a driving simulator?
Findings
In this post hoc analysis of data from 14 individuals with homonymous hemianopia who had participated in a cross-sectional study, the percentage of intersections at which an early large scan was made was higher in drives with vs without reminders (65% vs 45%). Responses to hazards were approximately 2 seconds faster when an early large scan was made.
Meaning
In this study, the auditory reminder cues promoted proactive scanning and faster responses to hazards.
This cross-sectional study of a driving simulation evaluates whether auditory reminder cues promoted proactive scanning on approach to intersections in individuals with homonymous hemianopia.
Abstract
Importance
Individuals with homonymous hemianopia (HH) are permitted to drive in some jurisdictions. They could compensate for their hemifield vision loss by scanning toward the blind side. However, some drivers with HH do not scan adequately well to the blind side when approaching an intersection, resulting in delayed responses to hazards.
Objective
To evaluate whether auditory reminder cues promoted proactive scanning on approach to intersections.
Design, Setting, and Participants
This cross-sectional, single-visit driving simulator study was conducted from October 2018 to May 2019 at a vision rehabilitation research laboratory. A volunteer sample of individuals with HH without visual neglect are included in this analysis. This post hoc analysis was completed in July and August 2020.
Main Outcomes and Measures
Participants completed drives with and without scanning reminder cues (a single tone from a speaker on the blind side). Scanning was quantified by the percentage of intersections at which an early large scan was made (a scan with a head movement of at least 20° made before 30 m from the intersection). Responses to motorcycle hazards at intersections were quantified by the time to the first fixation and the time to the horn-press response.
Results
Sixteen individuals were recruited and completed the study. Two were subsequently excluded from analyses. Thus, data from 14 participants (median [IQR] age, 54 [36-66] years; 13 men [93%]) were included. Stroke was the primary cause of the HH (10 participants [71%]). Six (43%) had right-sided HH. Participants were more likely to make an early large scan to the blind side in drives with vs without cues (65% vs 45%; difference, 20% [95% CI, 5%-37%]; P < .001). When participants made an early large scan to the blind side, they were faster to make their first fixation on blind-side motorcycles (mean [SD], 1.77 [1.34] vs 3.88 [1.17] seconds; difference, −2.11 [95% CI, −2.46 to −1.75] seconds; P < .001) and faster to press the horn (mean [SD], 2.54 [1.19] vs 4.54 [1.37] seconds; difference, −2.00 [95% CI, −2.38 to −1.62] seconds; P < .001) than when they did not make an early scan.
Conclusions and Relevance
This post hoc analysis suggests that auditory reminder cues may promote proactive scanning, which may be associated with faster responses to hazards. This hypothesis should be considered in future prospective studies.
Introduction
Drivers with homonymous hemianopia (HH) could compensate for their hemifield loss by scanning toward the blind hemifield using eye and head movements. However, prior research1,2,3,4,5 suggests that drivers with HH sometimes fail to scan to the blind side when approaching an intersection or do not scan far enough, resulting in impaired responses to hazards. A large gaze scan, up to 90°, composed of head as well as eye movements, may be needed to see hazards approaching from the blind side at a 4-way intersection. We report a post hoc analysis of data from a proof-of-concept study designed to evaluate auditory reminder cues as a method to alert individuals with HH when they fail to make a large enough scan to the blind side. The analysis investigated the potential of the reminders as a training tool, specifically, whether the reminders promoted proactive scanning, defined as making an early large scan to the blind side.
Methods
Between October 2018 and May 2019, 16 participants with HH without visual neglect completed the study (eMethods 1 in the Supplement). Two participants were excluded from analyses for excessive noise in the gaze data and an inability to follow experimental instructions. The study followed the tenets of the Declaration of Helsinki and was approved by the institutional review board of Massachusetts Eye and Ear. All participants provided written informed consent.
Driving Simulator Scenarios
Participants drove along predetermined routes in an urban environment in a LE-1500 driving simulator (FAAC Corp) (eMethods 2 and eFigure 1 in the Supplement). Head and eye movements were recorded with a remote, 6-camera tracking system (Smart Eye Pro version 6.1 [Smart Eye]). Each route included at least 33 intersections. Motorcycle hazards were programmed at 10 four-way intersections per route (eMethods 3 and eTable 1 in the Supplement). When the driver’s vehicle was 60 m from the intersection, a motorcycle was triggered to appear on the cross street, 5 times on the left and 5 times on the right (eccentricity of 51.7°), and move toward the intersection at a constant speed of 48 km/h (30 mph) (eFigure 2 in the Supplement).
Reminder Cues for Blind-Side Scanning
The reminder cues were a single tone (duration, 1 second6) played through a speaker at 45° on the side of the field loss, indicating which direction to scan. A cue was given only when participants failed to make an early large scan, defined as a scan with a head movement of at least 20° to the blind side between 60 m and 30 m to the intersection (based on a prior study,2 which found that a head scan of at least 20° was highly associated with blind-side hazard detection) (eTable 2 in the Supplement). The reminder was given at 30 m to allow sufficient time for participants to scan and respond to a hazard.
Driving Simulator Procedures
Participants followed normal traffic rules and pressed the horn to indicate detection of motorcycles. After simulator acclimation and practice drives, they completed drives for the 2 experimental conditions (with and without reminder cues) in counterbalanced order. Participants drove 2 routes for each condition (with the order counterbalanced). Participants practiced responding to the reminder cues before the first drive with cues (eMethods 4 in the Supplement).
Outcome Measures
Responses to reminder cues were quantified by the direction, timing, and magnitude of the first head movement after the cue onset (eFigure 3 in the Supplement). Scanning was quantified by early large scan rates (the percentage of intersections at which an early large scan was made), as well as the total number of head scans (eMethods 5 in the Supplement). Responses to motorcycles were quantified by the time to the first fixation on the motorcycle and the time to the horn-press response.
Statistical Analyses
This post hoc analysis was conducted in July and August 2020 using R version 1.1.456 (R Foundation for Statistical Computing). For continuous measures, we used a linear mixed-effects analysis.7 For noncontinuous measures, a generalized linear mixed-effect model was applied. All models included condition (with vs without cues) as a factor. Models for fixation and response times also included the side of the motorcycle (blind vs seeing) and early large scan (present vs absent) as factors. Intersection number and participant identifier were included as random factors. All P values were 2-sided; P values less than .05 were considered significant, and there were no adjustments for multiple analyses.
Results
Sample characteristics are summarized in Table 1. Briefly, the 14 participants included in analysis had a median (IQR) age of 54 (36-66) years, 13 (93%) were men, and 6 (43%) had right-sided visual field loss.
Table 1. Characteristics of the 14 Participants Included in Post Hoc Analyses.
| Characteristic | Median (IQR) |
|---|---|
| Age, y | 54 (36 to 66) |
| Male, No. (%) | 13 (93) |
| Binocular visual acuity, logMAR | −0.07 (−0.09 to 0.01) |
| Montreal Cognitive Assessment score | 26 (25 to 29)a |
| Right-sided homonymous hemianopia, No. (%) | 6 (43) |
| Hemianopia cause, No. (%) | |
| Stroke | 10 (71) |
| Surgery | 2 (15) |
| Trauma | 1 (7) |
| Brain cyst | 1 (7) |
| Age at onset, y | 37 (29 to 57) |
| Time since onset, y | 10 (3 to 16) |
| Current driver, No. (%) | 3 (21)b |
| Total driving experience, y | 21 (4 to 43) |
Data are missing for 2 participants.
More details are provided in eMethods 1 in the Supplement.
When a reminder cue was given, the most common response (total, 232 [85% of all reminder cue events]) was a head scan to the blind side of median (IQR) magnitude 26.7° (17.7°-36.6°), initiated a median (IQR) of 0.61 (0.45-0.81) seconds after the cue onset. Participants were more likely to make an early large scan to the blind side in the drives with cues (ie, a scan before a reminder cue) than in the drives without cues (mean, 65% vs 45%; difference, 20% [95% CI, 5%-37%]; P < .001; Figure). Overall, participants made more head scans to the blind side in drives with cues (mean [SD], 3.2 [1.5] vs 2.6 [1.6] scans; difference, 0.6 [95% CI, 0.42-0.82] scans), but the number of head scans to the seeing side did not differ between the 2 conditions (Table 2).
Figure. Percentage of Intersections at Which Participants Made an Early Large Scan to the Blind Side in Drives With vs Without Cues.
Each data point represents 1 participant.
Table 2. Results Summary for Number of Scans, Time to First Fixation, and Response Time.
| Measure | Mean (SD) | Difference between means (95% CI) | P valuea | Mean (SD) | Difference between means (95% CI) | P valuea | ||
|---|---|---|---|---|---|---|---|---|
| With cues | Without cues | Early large scan | No early large scan | |||||
| No. of scans | ||||||||
| Blind side | 3.2 (1.5) | 2.6 (1.6) | 0.6 (0.4 to 0.8) | .05 | NA | NA | NA | NA |
| Seeing side | 1.9 (1.5) | 1.7 (1.4) | 0.2 (0 to 0.4) | NA | NA | NA | ||
| Time to first fixation, s | ||||||||
| Blind-side motorcycles | 2.16 (1.55) | 2.56 (1.63) | −0.40 (−0.82 to 0.02) | .07 | 1.77 (1.34) | 3.88 (1.17) | −2.11 (−2.46 to −1.75) | <.001 |
| Seeing-side motorcycles | 2.52 (1.72) | 2.57 (1.64) | −0.05 (−0.53 to 0.44) | 2.61 (1.62) | 2.48 (1.74) | 0.13 (−0.35 to 0.61) | ||
| Response time, s | ||||||||
| Blind-side motorcycles | 2.88 (1.42) | 3.35 (1.62) | −0.46 (−0.86 to −0.07) | .05 | 2.54 (1.19) | 4.54 (1.37) | −2.00 (−2.38 to −1.62) | <.001 |
| Seeing-side motorcycles | 3.11 (1.42) | 3.20 (1.53) | −0.09 (−0.48 to 0.29) | 3.24 (1.49) | 3.07 (1.45) | 0.17 (−0.21 to 0.55) | ||
The P values for the interactions were obtained by likelihood ratio tests of the full linear mixed-effects model with the outcomes in question compared with the model without the outcomes.
Fixation times did not differ, but responses to blind-side motorcycles were faster in drives with vs without cues (mean [SD], 2.88 [1.42] vs 3.35 [1.62] seconds; difference, −0.46 [95% CI, −0.86 to −0.07] seconds; Table 2). In contrast, for seeing-side motorcycles, there was little difference between the 2 conditions (Table 2). When participants made an early large scan to the blind side, they were faster to make their first fixation on blind-side motorcycles (mean [SD], 1.77 (1.34) vs 3.88 (1.17) seconds; difference, −2.11 [95% CI, −2.46 to −1.75] seconds; Table 2) and faster to press the horn than when they did not make an early scan (mean [SD], 2.54 [1.19] vs 4.54 [1.37] seconds; difference, −2.00 [95% CI, −2.38 to −1.62] seconds; Table 2). However, making an early large scan to the blind-side had no association with fixation or response times to seeing-side motorcycles (Table 2).
Discussion
Participants initiated a head movement to the blind side approximately 0.6 seconds after receiving a cue, suggesting the cues worked well as reminders to scan. Participants made more head scans to the blind side in drives with cues, and their response times were faster. Post hoc analysis revealed that most participants modified their scanning behaviors when driving with the reminders (Figure). The possibility that they might receive a reminder to scan seemed to promote a more proactive scanning strategy; that is, they were more likely to make an early large scan to the blind side in drives with the reminders, because if they made an early scan, they did not receive a reminder. Making an early large scan had positive associations with hazard responses. For motorcycles from the blind side, fixation and response times were approximately 2 seconds faster when an early large scan was made. When there was an early large scan to the blind side, response times to motorcycles approaching on the blind side were faster than response times to motorcycles approaching on the seeing side. Importantly, the cues had no adverse associations with seeing-side scanning or responses.
Limitations
The sample was relatively small and heterogeneous, including currently licensed and unlicensed drivers, and the analysis was post hoc. Detection was evaluated in relatively simple motorcycle-hazard scenarios and response times were based on horn-press rather than braking responses.
Conclusions
This post hoc analysis suggests that auditory reminder cues may promote proactive scanning, which may be associated with faster responses to hazards. Thus, reminder cues may be useful in training aimed at improving scanning of drivers with HH. These hypotheses should be considered in future prospective studies.
eMethods 1. Participants
eMethods 2. Apparatus
eFigure 1. Driving simulator
eMethods 3. Driving simulator scenarios
eTable 1. Characteristics of the intersections included in the two routes
eFigure 2. Schematic diagram for a four-way stop-controlled intersection with a motorcycle event
eMethods 4. Procedures for practicing responses to reminder cues
eMethods 5. Quantification of head scanning
eTable 2. Minimum head movement threshold used to define a head scan
eFigure 3. Lateral head position on approach to an intersection when a cue was given for a participant with right hemianopia
References
- 1.Bowers AR, Tant M, Peli E. A pilot evaluation of on-road detection performance by drivers with hemianopia using oblique peripheral prisms. Stroke Res Treat. 2012;2012:176806. doi: 10.1155/2012/176806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Swan G, Savage SW, Zhang L, Bowers AR. Driving with hemianopia: VII, predicting hazard detection with gaze and head scanning magnitude. Transl Vis Sci Technol. 2021;10(1):20. doi: 10.1167/tvst.10.1.20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bowers AR, Ananyev E, Mandel AJ, Goldstein RB, Peli E. Driving with hemianopia: IV, head scanning and detection at intersections in a simulator. Invest Ophthalmol Vis Sci. 2014;55(3):1540-1548. doi: 10.1167/iovs.13-12748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bowers AR, Alberti CF, Hwang AD, Goldstein RB, Peli E. Pilot study of gaze scanning and intersection detection failures by drivers with hemianopia. Proceedings of the 8th International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design. Published 2015. Accessed October 19, 2021. https://drivingassessment.prod.drupal.uiowa.edu/sites/drivingassessment.uiowa.edu/files/wysiwyg_uploads/037.pdf
- 5.Papageorgiou E, Hardiess G, Mallot HA, Schiefer U. Gaze patterns predicting successful collision avoidance in patients with homonymous visual field defects. Vision Res. 2012;65:25-37. doi: 10.1016/j.visres.2012.06.004 [DOI] [PubMed] [Google Scholar]
- 6.Edworthy J, Loxley S, Dennis I. Improving auditory warning design: relationship between warning sound parameters and perceived urgency. Hum Factors. 1991;33(2):205-231. doi: 10.1177/001872089103300206 [DOI] [PubMed] [Google Scholar]
- 7.Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67(1):1-45. doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods 1. Participants
eMethods 2. Apparatus
eFigure 1. Driving simulator
eMethods 3. Driving simulator scenarios
eTable 1. Characteristics of the intersections included in the two routes
eFigure 2. Schematic diagram for a four-way stop-controlled intersection with a motorcycle event
eMethods 4. Procedures for practicing responses to reminder cues
eMethods 5. Quantification of head scanning
eTable 2. Minimum head movement threshold used to define a head scan
eFigure 3. Lateral head position on approach to an intersection when a cue was given for a participant with right hemianopia
