The present work compares the efficiency of two training techniques as aids to learning selected aspects of a sequentially ordered action such as that of rowing. Subjects in one group were trained with a conventional learning technique (CLT) while those in a second group were trained by an augmented feedback technique referred to as external feedback (EFB). Progress was recorded on learning curves. Rowing athletes with limited experience and psychophysiology students were used for the study. The tasks consisted of learning movement timing (rhythmicity of action and coordination of body parts) and movement intensity (force and electromyogram development), in four separate experiments. The learning curves for EFB subjects were found to have significantly and consistently higher slopes than those for CLT subjects. Optimal criteria were reached by EBF subjects, after a continuous increase in performance levels and a concomitant decrease in standard deviations evaluated from periodicity, movement accuracy and force. Subjects, who after 8 to 10 sessions of CLT learning had not reached optimal level, were exposed to EFB. Their performances then showed a marked improvement and attained the required criterion in 2 to 4 sessions. This further demonstrates the efficacy of EFB as compared with CLT, as an aid to learning a complex sensorimotor action. The efficacy of EFB as a learning technique is discussed in relation to the internal model of the task to be executed and to sensory motor control and motor programmes.