ABSTRACTTwo experiments are reported which investigate the efficacy of Ashton, White, and Hodgson's (1979) two‐alterative‐forced‐choice discrimination technique for training accuracy in perception of heart rate. In the first study, subjects were assigned to either a knowledge of results (KR) or no‐knowledge of results condition (noKR). In the second study a third condition, augmented feedback (AFB), was added. The two studies differed essentially in the timing of the commencement of training, and in the number of training sessions employed. Transfer of training was assessed in both studies by testing all subjects on a heartbeat discrimination task modified from Brener and Jones (1974). The pattern of results was consistent in indicating that providing subjects with information about the accuracy of their decisions is an effective means of training cardiac awareness, and that transfer of the trained awareness to a more commonly used exteroceptive discrimination procedure occurs. The data also point to the possibility that allowing subjects to “practice” prior to any training may be a procedural measure best avoided in discrimination training