The Ecological Validity of Measures of Visual Attention in Community-Dwelling Older Adults
View the poster >>
The relation between instrumental activities of daily living (IADLs) and neuropsychological test performance is unclear. Evidence suggests that neuropsychological measures more analogous to everyday tasks (i.e., measures with verisimilitude) predict variance in IADLs beyond traditional measures; however, few studies have included measures of visual attention despite the fact that this domain of cognitive function is involved in most IADLs. The purpose of the current study was to determine whether a neuropsychological measure of visual attention with verisimilitude, Driving Scenes from the Neuropsychological Assessment Battery (NAB), predicts variance in IADLs beyond that accounted for by a traditional, commonly used measure of visual attention, Spatial Span, and other demographic, global cognitive, and mood variables. Forty community-dwelling older adults (35 female) with mean (SD) age and education of 78.4 (7.5) and 11.9 (2.6) years, respectively, completed a battery of neuropsychological measures including Spatial Span and Driving Scenes. In Spatial Span examinees point to a sequence of specific blocks amongst an array of blocks. In Driving Scenes examinees are asked to identify differences between one image of a street from the perspective of an automobile driver and a second, similar image after the first image is removed. Participants also completed a performance based measure of IADLs, the Revised Observed Tasks of Daily Living (OTDL-R). The OTDL-R requires examinees to complete tasks involving cooking, medication management, and telephone use. A hierarchical regression was used to determine whether Driving Scenes predicted significant variance in OTDL-R performance beyond Spatial Span, demographic variables (age, education, estimated premorbid intellectual functioning), and measures of depression (Geriatric Depression Scale) and global cognitive functioning (Mini Mental State Exam). Results indicated that Driving Scenes had the highest correlation with the OTDL-R (r = .64, p = .001). The regression indicated that a total of 58.6% of the variance in OTDL-R performance was predicted, F(7,32) = 6.48, p = .001. When entered together on the first step, demographics, depression, and global cognitive function predicted 38.7% of the variance in OTDL-R performance, F(5,34) = 4.30, p = .004. On the second step, Spatial Span predicted an additional 10.6% of the variance, F(1,33) = 6.92, p = .013. On the third step, Driving Scenes predicted a significant 9.3% of additional variance in OTDL-R performance beyond the other measures, F(1,32) = 7.18, p = .012. Driving Scenes had the highest Beta weight in the regression, predicting significant independent variance in OTDL-R performance, t = 2.63, p = .013. These results are consistent with other studies suggesting that measures with verisimilitude predict variance in daily functioning that is not captured by traditional measures and support the inclusion of Driving Scenes in neuropsychological batteries used to predict daily functioning in older adults.