Human studies have indicated that the use of anabolic androgenic steroids may be associated with the abuse of alcohol and other drugs. Also, experimental animal research has indicated that chronic nandrolone administration subsequently increases voluntary alcohol drinking. The aim of our study was to test our hypothesis that alcohol-induced testosterone elevation, especially associated with stress conditions derived by nandrolone treatment, could be the underlying factor in causing increased alcohol drinking. Male alcohol-preferring AA and low drinking Wistar rats were randomly divided into control and nandrolone decanoate treated (15mg/kg for 14days) groups. Basal serum testosterone and corticosterone were determined before the first nandrolone treatment, after 7days of treatment, and after an additional (7-day) washout period, during which also the acute effect of alcohol (1.5g/kg) on steroid hormones was determined. Hereafter followed a (5-week) voluntary alcohol consumption period, during the last 2weeks of which the rats were treated again with nandrolone. Both normal and reversed dark- vs. light-cycle experimental designs were used. Contrary to our hypothesis, nandrolone treatment decreased voluntary alcohol consumption in both AA and Wistar rats. Also, instead of stress causation, elevated basal testosterone and lowered basal corticosterone levels were observed after nandrolone treatment in both AA rats and Wistars. During acute alcohol intoxication the frequency of testosterone decreases was higher in the nandrolone-treated groups compared with control AA and Wistar rats. Present data support the hypothesis that nandrolone-treatment mediated attenuation of alcohol intake in both AA and Wistar rats may be the result of negative reinforcement caused by alcohol-mediated testosterone reduction.