Lambert, Sharma, Ryckman Accident vulnerability

Published: 12 February 2019| Version 1 | DOI: 10.17632/7zkcbpc6cx.1
Tony Lambert,


These are the raw data for this study: Many accidents, such as those involving collisions or trips, appear to involve failures of vision; but the association between accident risk and vision as conventionally assessed, is weak or absent. We addressed this conundrum by embracing the distinction inspired by neuroscientific research, between vision for perception and vision for action. A dual-process perspective predicts that accident vulnerability will be associated more strongly with vision for action than vision for perception. Older and younger adults, with relatively high and relatively low self-reported accident vulnerability (Accident Proneness Questionnaire), completed three behavioural assessments targeting: vision for perception (Freiburg Visual Acuity Test); vision for action (Vision for Action Test - VAT); and the ability to perform physical actions involving balance, walking and standing (Short Physical Performance Battery). Accident vulnerability was not associated with visual acuity or with performance of physical actions; but was associated with VAT performance. VAT assesses the ability to link visual input with a specific action –launching a saccadic eye movement as rapidly as possible, in response to shapes presented in peripheral vision. The predictive relationship between VAT performance and accident vulnerability was independent of age, visual acuity and physical performance scores. Applied implications of these findings are considered.


Steps to reproduce

Questionnaire data: Bivariate correlations between APQ, CFQ and FES-I; plus comparison of older and younger participants with respect to these measures; plus Cronbach's alpha and mean inter-item correlation for APQ. Performance data: Partial correlation between APQ and threshold average (VAT threshold), controlling for age and visual acuity (and SPPB scores).


University of Auckland


Natural Sciences