Experimental investigations examining the effectiveness of social-skills-training approaches to the treatment of heterosexual-social anxiety in humans were critically reviewed. In general the review studies were suuportive of the skills-training treatment approach. These results should be interpreted with a certain degree of cautious optimism because many of the reviewed studies suffered from several methodological and procedural problems. The major methodological and procedural problems were concerned with subject selection and screening, inadequate and invalid assessment procedures, lack of transfer and follow-up measurement, and inattention to the interaction between treatment procedures and subject characteristics. These errors need to be rectified for investigators in this research area to make significant contributions to the various theories (conditioned anxiety, faulty cognitive-evaluative appraisal and skills deficit) regarding the etiology and maintainers of heterosexual-social anxiety and to the clinical treatment of such disorders.