The effectiveness of motor imagery training on cognitive performance was examined and the physiological mechanisms involved in the contribution of mental practice to motor learning were considered. The subject's mental effort during motor imagery was assessed by using psychophysiological measures and particularly eye blink activity as an "indirect" measurement of subjects' attention. An electronic flight simulation program (Multiple Attribute Task Battery--MATB) was used to assess performance. Twenty healthy volunteers participated in the study divided in two groups: the control group and the imagery-training group. The subjects of the imagery group were asked for additional imagery training. The subjects of the actual performing group were asked additionally to passively observe the task in order to have equal time of exposure to the task. Performance scores and physiological parameters such as heart rate, respiratory rate, eye blinking activity and muscular activity were recorded during all sessions. The results revealed significantly higher performance level of the imagery-training group than the control group. Heart rate and respiratory rate significantly increased during imagery sessions compared to rest. A slight electromyographic activity was observed during the imagination of movement. Our findings support the notion that mental practice improves motor performance in a task where spatiotemporal or dynamic control of the action is highly required. The effects of mental practice on motor performance could be explained by the existence of a top-down mechanism based on the activation of a central representation of the movements, since the vegetative activation during motor imagery seems to be centrally controlled.