April 2010
Volume 51, Issue 13
ARVO Annual Meeting Abstract  |   April 2010
What Safety Margin Do Pedestrians Incorporate Into Their Street-Crossing Decision?
Author Affiliations & Notes
  • S. E. Hassan
    School of Optometry, Indiana University, Bloomington, Indiana
  • G. D. Barnett
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • A. K. H. Williamson
    School of Optometry, Indiana University, Bloomington, Indiana
  • R. W. Massof
    Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland
  • Footnotes
    Commercial Relationships  S.E. Hassan, None; G.D. Barnett, None; A.K.H. Williamson, None; R.W. Massof, None.
  • Footnotes
    Support  NIH Grant R03EY014874-05
Investigative Ophthalmology & Visual Science April 2010, Vol.51, 3626. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      S. E. Hassan, G. D. Barnett, A. K. H. Williamson, R. W. Massof; What Safety Margin Do Pedestrians Incorporate Into Their Street-Crossing Decision?. Invest. Ophthalmol. Vis. Sci. 2010;51(13):3626.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose: : The aim of this study was to determine how much of a safety margin normally-sighted, visually impaired and blind pedestrians incorporate into their street-crossing decision and whether or not the safety margin changes under different sensory conditions.

Methods: : Using a 5 point rating scale, safety ratings from 12 normally sighted, 10 visually impaired and 10 blind subjects were collected for 8 different vehicular gap times under three sensory conditions: (i) vision and hearing; (ii) vision only; and (iii) hearing only. Receiver Operating Characteristic (ROC) curves were fitted for all possible gap pairs and sensory conditions and the discriminability (d’) of the street-crossing safety decision variable for all gap pairs and sensory conditions was calculated as the area under the ROC curve. Using the dissimilarity matrix of the d’ values in a one -dimensional scaling model, we estimated the means of each distribution of the decision variable relative to a "center of gravity" (COG). The means of each of these distributions were plotted against gap time, and an ogive function was fitted to the data. The x-intercept (tCOG) of each ogive function, which represents the time (in seconds) at which subjects went from classifying gap times as being "safe" to unsafe" and vice versa was computed for each subject and sensory condition. Subjects’ safety margin under each sensory condition was then calculated as the difference between the subject’s actual street-crossing time and their tCOG.

Results: : We found no significant difference in the safety margins between the normally-sighted and visually impaired subjects. This was true for all sensory conditions (p>0.05). The safety margin of the blind subjects (-1.99 sec) however was significantly shorter than the safety margin of either the normally-sighted (-0.20 sec, p=0.018) or visually impaired subjects (-0.23 sec, p=0.019) under the hearing only sensory condition.

Conclusions: : Our data suggests that pedestrians do not allow enough time to cross the street when basing their street-crossing decision using only auditory information since all subjects had negative safety margins under the hearing only condition. Compared to either the normally-sighted or visually impaired pedestrians, blind pedestrians appear to exhibit the least safe street-crossing decision making behavior and thus may benefit from training that improves their detection ability and /or interpretation of vehicular gap times so that they can incorporate a longer safety margin into their street-crossing decision.

Keywords: low vision • vision and action 

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.