The article by Catalan and colleagues1 provides an updated meta-analysis of neuropsychological impairment in individuals considered to be at clinical high risk for psychosis (CHR-P). The authors summarize the extensive literature addressing the CHR-P impairments compared with healthy controls, as well as the smaller literature contrasting impairments seen in individuals at CHR-P who will later develop full psychosis with those who do not. As suggested by earlier reviews,2 (1) individuals at CHR-P produce heterogeneous neuropsychological findings yet demonstrate impairments compared with healthy controls across nearly all measured abilities at broadly similar levels of impairment (around 0.5 SD); (2) people who later develop full psychosis evidence greater levels of impairment, again across nearly all measured abilities, compared with those who do not develop full psychosis (around 0.3 SD); and (3) the heterogeneity, nonspecificity, and small to medium effect sizes of neuropsychological findings in individuals at CHR-P appear to limit traditional neuropsychological testing to a supporting role both in the identification of the CHR-P state and in the critical task of predicting which individuals at CHR-P will actually progress to psychosis. Given the consistency of these findings, the publication of this comprehensive meta-analysis offers a moment to reflect on this body of work and its clinical utility and to consider alternative approaches to the use of cognitive measures that may enhance their value in CHR-P research.