Abstract
Does chemotherapy impair selective attention? This is an important practical problem which demands input from basic vision science research. There is now substantial evidence that chemotherapy patients experience significant cognitive impairments. However, methodological problems make it difficult to determine which specific cognitive domains are impaired. The only point of consensus among the six existing meta-analytic reviews is that "attention" is not impaired. However, most neuropsychological "attention" measures have little or nothing to do with attention as vision scientists understand it; the majority are variations on memory span. Conversely, tests with some validity as attention tests (e.g., letter cancellation, Stroop) are sometimes excluded from the attention category. I re-analyzed data from one meta-analysis (Jim et al. Journal of Clinical Oncology, 2012), summarizing 17 studies of 807 breast cancer patients, tested at least 6 months after termination of chemotherapy. I extracted tests with some validity as selective attention tests (Stroop, test with a visual search component). I found a significant impairment for both Stroop (N studies = 5, effect size g = -0.214, p = .041) and search tests (N = 34, g = -0.080, p = .030). There was no search impairment in longitudinal comparisons (N = 9, g = -0.019, p = .074), or comparisons to healthy controls (N = 6, g = -0.043, p = .305); the effect was driven primarily by comparisons to cancer patients who did not receive chemotherapy (N = 19, g = -0.119, p = .020). However, in my view, none of these neuropsychological tests are properly designed to measure attentional function. There is a need for tests which are sensitive, designed to measure specific aspects of attention, and can be used to make direct connections to contemporary theories of attention and its neural underpinnings.
Meeting abstract presented at VSS 2014