Purchase this article with an account.
W. LiKamWa, C. S. Barnes, R. A. Schuchard; A New Visual Search Test for Finding Everyday Objects. Invest. Ophthalmol. Vis. Sci. 2009;50(13):4719.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Individuals with low vision often experience difficulties locating a single object in everyday tasks. Therefore, we evaluated a new visual search test as a tool to assess the ability to find everyday objects in real-world scenes.
Visual search performance, visual acuity, and contrast sensitivity were tested binocularly in 47 subjects, ages 22 to 87. The visual search test consisted of 16 cluttered scenes, each used for 1-3 trials, for a total of 39 trials (including 3 practice trials). These 16 scenes were either pages from the I SpyTM children’s picture book, used with permission, or photographs taken in a home or store. Single objects within the cluttered scenes were selected and cut out of the scene. During each trial of the visual search test, the subject was shown a single target object and asked to identify or describe it with one or two words. A cluttered scene of many objects would then appear on the screen, and the subject was asked to find the target object within the scene as quickly as possible. Accuracy and response times for the 36 trials were recorded for each subject.
Data were obtained for 10 "younger" subjects with normal vision, 15 older control subjects with age-normal vision, and 22 subjects who had been diagnosed with either age-related macular degeneration or retinitis pigmentosa. Subjects had median (range) visual acuity of 0.00 logMAR (-0.18 to 1.18), contrast sensitivity of 1.6 logContrast (0.15 to 1.95), visual search accuracy of 86.11% (30.56% to 100%), and visual search average response time of 11.15 seconds (2.95 to 51.79). Linear regression analysis showed that both visual acuity and contrast sensitivity had significant relationships (p<0.0001) with visual search performance. Approximately 79% of the within-subject variation in response times was explained by the regression of response time standard deviations on average response times. Object-by-object analysis clarified the variation in visual search performance within and between subjects. Visual search accuracy and response time had a moderately strong inverse correlation (r=-0.78). Two-sample t-tests indicated that older subjects with age-normal vision did not perform as well on the visual search test, compared to younger subjects with normal vision (p<0.001). Differences in visual search performance between these visually impaired subjects and age-matched subjects with normal vision were not found to be statistically significant (=0.05).
Our results suggest that this visual search test can be used to examine the ability to find a single object and reveals the relationship between visual function and searching for everyday objects in real-world scenes.
This PDF is available to Subscribers Only