In extending the implications of our earlier research, we found that a measure of impulsivity developed by Barratt (1965) differentiated recovering, male pathological gamblers (N = 12; mean age = 48.9 years) from male control subjects (N = 15; mean age = 43.3 years). Among the gamblers themselves, however, this measure of impulsivity did not correlate with an index of the social and familial disruption engendered by past gambling. In contrast, a measure of one facet of the gamblers' cognitive style (the TF subscale of the Myers-Briggs Inventory) did correlate with this index of gambling-induced disruption but did not differentiate gamblers from controls. These results, as well as other findings, are discussed in the context of previous research and with regard to the inferential limits imposed by studies of this kind.