The Gambling Task (GT) has demonstrated sensitivity to a type of decision-making that differentiates individuals manifesting substance use disorders from those without such disorders. However, studies have not yet compared the GT performance of "heavy smokers" to the performance of never-smokers. In the present study, the GT performance of "heavy smokers" (n = 39) and never-smokers (n = 32) recruited from the community was compared in an experimental design. Analysis of covariance showed that "heavy smokers" performance on the GT was significantly worse than that of never-smokers (p < .01). Implications, the study's limitations and future directions are discussed.