Forty adult overweight female volunteers were randomly assigned to one of four conditions: (1) two self-control instruction conditions; (2) an automatic immunization instruction group; and (3) a no treatment control group. In the self-control groups, treatment was presented as an aid to gaining self-control over behavior, while in the automatic immunization group, it was explained in classical conditioning terminology. The major finding was that covert sensitization by itself has relatively little effect as a treatment for obesity. Relaxation level, visualization ability and felt discomfort were uncorrelated with weight loss. Methodological considerations are also discussed. (Author/TL)