Behavioral Traits Predicting Alcohol Drinking in Outbred Rats: An Investigation of Anxiety, Novelty Seeking, and Cognitive Flexibility


Reprint requests: Mary C. Olmstead, Department of Psychology, Queen’s University, Kingston, ON K7L 3N6, Canada; Tel.: 613-533-6208; Fax: 613-533-2499; E-mail:


Background:  Most adults in Western society consume alcohol regularly without negative consequences. For a small subpopulation, however, drinking can quickly progress to excessive and chronic intake. Given the dangers associated with alcohol abuse, it is critical to identify traits that may place an individual at risk for developing these behaviors. To that end, we used a rat model to determine whether anxiety-related behaviors, novelty seeking, or cognitive flexibility predict excessive alcohol drinking under both limited and continuous access conditions.

Methods:  Adult male rats were assessed in a series of behavioral tasks (elevated plus maze [EPM], locomotor activity, and discrimination/reversal learning in a Y-maze) followed by 6 weeks of daily, 1-hour access to alcohol in a free-choice, 2-bottle paradigm (10% alcohol vs. tap water). Next, subjects were given the opportunity to consume alcohol for 72 hours in drinking chambers that permit separate measures of each drinking bout. Half of the animals experienced a 2-week deprivation period between the limited and continuous access sessions.

Results:  Time spent on the open arms of the EPM, but not novelty seeking or discrimination/reversal learning, predicted alcohol consumption during limited, 1-h/d access sessions to alcohol. Anxiety-related behavior also predicted the escalation of intake when animals were given 72 hours of continuous access to alcohol. Bout size, but not frequency, was responsible for the increased consumption by high-anxiety subjects during this period. Finally, intake during limited access sessions predicted intake during continuous access, but only in subjects with low intake during limited access.

Conclusions:  These findings confirm that preexisting anxiety-related behavior predicts alcohol intake under several schedules of alcohol access. Moreover, when access is unlimited, the high-anxiety-related group exhibited an increase in bout size, but not frequency, of drinking. In addition, we show that modest intake when alcohol is restricted may or may not progress to excessive intake when the drug is freely available.