N400 as an index of semantic expectancies: Differential effects of alcohol and cocaine dependence Natalie A. Ceballos a, * , Rebecca J. Houston b , Natosha D. Smith d , Lance O. Bauer c , Robert E. Taylor d a Department of Behavioral Sciences, 1035 University Drive, 236 Medical School Building, University of Minnesota School of Medicine, Duluth, MN 55812-3031, United States b Research Institute on Addictions, State University of New York at Buffalo, Buffalo, NY 14203, United States c Alcohol Research Center, Department of Psychiatry, University of Connecticut School of Medicine, 263 Farmington Avenue, MC-2103, Farmington, CT 06030-2103, United States d Collaborative Alcohol Research Center, Department of Pharmacology, Howard University, College of Medicine, Washington, DC 20059, United States Accepted 18 April 2005 Available online 20 June 2005 Abstract Background: Chronic substance abuse has been associated with decrements in the processing and expression of language. The present study utilized the N400 event-related electroencephalographic potential to index semantic processing in 133 adults with (n = 49) or without (n = 84) a history of alcohol and/or cocaine dependence. The contributions of age, gender, and comorbid marijuana and nicotine dependence, and antisocial symptomology to N400 decrements were either covaried or controlled. Methods: A continuous series of 300 stimuli was presented for 150 ms each (interstimulus interval = 1475 ms) on a computer screen. The series was arranged such that a word (approximately 17% of stimuli) immediately preceded presentations of its antonym (primed condition; approximately 17% of stimuli), or a semantically unrelated word (unprimed condition; approximately 17% of stimuli). The remaining 50% of stimuli consisted of unpronounceable letter combinations (non-word condition). EEG responses to the antonyms, unrelated words, and letter jumbles were retained for analysis. Throughout the task, the subject pressed response keys to discriminate words from non-words. Results: Analyses revealed a detrimental effect of alcohol dependence on N400 amplitude and no significant main or interactive effects of cocaine dependence. Conclusion: The present findings suggest that alcohol-dependent individuals may exhibit verbal processing decrements. These findings also challenge hypotheses suggesting that the combined use of cocaine and alcohol is more deleterious to brain function than alcohol use alone. D 2005 Elsevier Inc. All rights reserved. Keywords: Alcohol; Cocaine; Evoked potentials; N400 1. Introduction Language comprehension can be studied via the N400 component of the event-related electroencephalographic potential. In the classic study of N400 by Kutas and Hillyard (1980), sentences were presented visually word-by- word. For half of the sentences, the concluding word of the sentence was congruous with its context. For other sentences, the concluding word was incongruous or odd, and did not logically complete the sentence. Incongruous sentence endings evoked a large scalp negativity 0278-5846/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2005.04.036 Abbreviations: ALCÀ/COCÀ, Alcohol and Cocaine Dependence Negative; ALC+/COCÀ, Alcohol Dependence Positive and Cocaine Dependence Negative; ALCÀ/COC+, Alcohol Dependence Negative and Cocaine Dependence Positive; ALC+/COC+, Alcohol and Cocaine Dependence Positive; ERP, Event-Related Potential; COGA, Collaborative Study on the Genetics of Alcoholism; PCA, Principal Components Analysis; SSAGA, Semi-Structured Assessment for the Genetics of Alcoholism. * Corresponding author. Tel.: +1 218 726 8425; fax: +1 218 726 7559. E-mail address: nceballo@d.umn.edu (N.A. Ceballos). Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 936 – 943 www.elsevier.com/locate/pnpbp