This proposal addresses the cognitive mechanisms underlying schizophrenic performance deficits of attention and language processing. Deficits have been observed in a variety of tasks related to each of these domains. However, there is little understanding of how these deficits relate to one another, or the information processing mechanisms that are responsible for them. In this proposal we describe a research program that uses theoretical and empirical methods from cognitive psychology to address these issues. We focus on deficits in three types of tasks: the Stroop task (a measure of selective attention), the continuous performance test (also an attentional measure), and lexical disambiguation tasks (which tests language processing). We present a set of computer models - based on the mechanisms of parallel distributed processing, or """"""""connectionism"""""""" - that simulate the performance of normal and schizophrenic subjects in each of these tasks. The same disturbance is introduced into all three models to induce schizophrenic patterns of performance. As such, the models account for schizophrenic behavior in these tasks in terms of a single, specific disturbance in information processing: an impairment in the ability to maintain context for the selection of appropriate action. Prospectively, the models identify two dimensions along which behavioral tasks can be varied (the relative strength of competing response tendencies, and the demand placed on memory for context) that are predicted to have direct and specific influence on schizophrenic performance. We outline a research plan designed to test these predictions, and the hypothesis that a single, underlying disturbance can account for schizophrenic deficits in measures of attention and language. This involves a study, in which schizophrenics and appropriate controls are tested in the three tasks referred to above, as well as in a task that controls for short term memory function. All subjects will be tested in all tasks, so that cross-task correlations of performance can be studied within subjects. Finally, hypotheses concerning variables of clinical and biological relevance will be tested. While this research focuses on a particular subset of the psychopathological phenomena associated with schizophrenia, it promises to advance our understanding of the basic cognitive mechanisms underlying these phenomena, and to organize a set of performance deficits that have occupied schizophrenia researchers for three decades. Finally, this work has the potential, over the long term, to relate behavioral deficits to underlying biological mechanisms. Our theoretical models have been developed within a framework that seeks to explain cognitive and behavioral phenomena in terms of information processing mechanisms that are plausible at the biological level. While this proposal focuses primarily on the behavioral component of this research, validation of our models in the behavioral studies proposed here will lead directly to a theoretically motivated program of research concerning the biological mechanisms underlying cognitive deficits.
| Forster, Sarah E; Carter, Cameron S; Cohen, Jonathan D et al. (2011) Parametric manipulation of the conflict signal and control-state adaptation. J Cogn Neurosci 23:923-35 |
| Cho, Raymond Y; Orr, Joseph M; Cohen, Jonathan D et al. (2009) Generalized signaling for control: evidence from postconflict and posterror performance adjustments. J Exp Psychol Hum Percept Perform 35:1161-77 |
