The hypothetical involvement of central noradrenergic projections in the manifestation of "anxiety" and the "anti-anxiety" effects of alcohol and benzodiazepines was tested in an operant conflict situation by examining the effects of destruction of this noradrenergic system on response rates. Rats were trained on a Geller-Seifter conflict test  modified for incremental shock , in which two food-reinforced lever press components were separated by a time-out. Responses during the "food-alone" component (RI) were reinforced on a random interval schedule-30 sec; responses during the "time-out" component were never reinforced; and responses during the conflict component (CONFLICT) were continuously reinforced with both food and shock. This shock strength increased incrementally with each successive shock during the conflict period. Each session consisted of two cycles of a 5 min RI period, a two min time out, and a two min CONFLICT period presented in succession. Both ethanol (0.5 to 1.0 g/kg) and chlordiazepoxide (5 and 10 mg/kg) produced a significant increase in punished responding during the CONFLICT component. Ethanol, but not chlordiazepoxide (CDP) also significantly decreased responding during the RI component. Virtual total destruction of the dorsal noradrenergic projection with 6-hydroxydopamine (6-OHDA), or destruction of the locus coeruleus itself, failed to significantly alter baseline responding or the release of punished responding to ethanol or CDP. These results do not support the hypothesis that the locus coeruleus projections have an essential role in "anxiety" or in the "anxiety-reducing" properties of ethanol or the benzodiazepines.