This study had two aims: (a) to test precisely the degree to which subjects can learn to discriminate their own heartbeats and (b) to pursue preliminary data suggesting that males and females differ in their ability to learn such discriminations. A new methodology, based on the theory of signal detection, was employed to evaluate objectively and quantitatively the performance of nine male and nine female subjects. The results confirmed the validity of the new methodology for assessing heartbeat detection and also confirmed earlier observations that males are able to learn to detect their own heartbeats but females are not. The results are discussed with particular reference to implications for viscerally based theories of emotion.