Simulation of impairment: Commentary on Nickels, Rapp, and Kohnen (2015)

Nickels, L., Rapp, B., and Kohnen, S. (2015. Challenges in the use of treatment to investigate cognition. Cognitive Neuropsychology, 32, 91-103) argue that impairment and treatment may be used to test computational models of cognition. They state that, contrary to their view, the authors of the WEAVER++ model of spoken word production have explicitly rejected simulation of impairment [i.e., Levelt, W. J. M., Roelofs, A., & Meyer, A. S. (1999b). Multiple perspectives on word production. Behavioral and Brain Sciences, 22, 61-69]. Here, I argue that this incorrectly characterizes the position of Levelt et al. Moreover, I further clarify this position, which holds that simulation of impairment requires both a theory of the intact system and assumptions about the underlying deficit, which is a widely accepted view. To demonstrate this position, I outline the approach taken in WEAVER++ simulations of aphasic performance reported in Roelofs, A. (2014. A dorsal-pathway account of aphasic language production: The WEAVER++/ARC model. Cortex, 59, 33-48). These simulations not only prove that the developers of WEAVER++ endorse simulation of impairment, but also highlight the importance of integrating psycholinguistic, functional neuroimaging, and tractographic evidence in testing models of impaired performance.