Detecting Insufficient Effort on Computer-Based Concussion Assessment Programs: A Comparison of Two Batteries

J Int Neuropsychol Soc. 2009 Feb;


A. Capps, A., J. D. Gfeller, B. C.M., M. V. Oliveri and M. J. Ross.



Objective: Assessing the neuropsychological effects of concussion with computerized testing has increased in recent years as these batteries offer significant advantages over traditional neuropsychological tests. Factors such as fatigue or low motivation, however, can influence an individual’s test performance, resulting in insufficient effort. The present study examined this issue by assessing the accuracy of the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) and the Automated Neuropsychological Assessment Metrics Sports Medicine Battery (ASMB) to detect insufficient effort using an analog coached simulator paradigm. Participants and Methods: Seventy one undergraduate participants were randomly assigned to one of three conditions (optimal performance, simulated fatigue, or simulated concussion), provided with information on how to perform, and completed ImPACT and ASMB. The sample was 70% female, 76% Caucasian, with a mean age of 19.87 (SD = 1.11), and 13.22 (SD = 0.97) years of education. Results: Analyses indicated that the three conditions differed significantly from each other on most ImPACT composite scores and ASMB throughput scores. The simulated concussion group performed least well while the optimal performance group performed best. The performance of the simulated fatigue group typically fell between the other two groups. Classification of the optimal performance and simulated concussion groups, using ImPACT effort indicators, yielded an 87% classification rate (96% sensitivity, 78% specificity). A proposed ASMB effort indicator resulted in 79% correct classification (92% sensitivity, 65% specificity). Conclusions: Both test batteries detected insufficient effort relatively well. However, replication of the findings and additional research with clinical groups is recommended.

Links to full article: