Purpose: The ability to perform motor imagery (MI) might be impaired by the physical fatigue elicited during training. Interestingly, there is also theoretical support for a more limited influence of fatigue in the existing literature. Method: We evaluated MI ability before and after two exercise protocols: (i) a continuous exercise of 20 min performed on a cycle ergometer at 80% of the secondary ventilatory threshold (Continuous exercise), and (ii) an intermittent exercise of 20 min involving sprints at maximal intensity performed with regular intervals (Intermittent exercise). MI ability evaluations were performed using validated behavioral (mental chronometry) and psychometric (subjective reports) methods. MI ability evaluations included mental rehearsal of a motor sequence which involved the main effectors of the exercise protocols (walking), and mental rehearsal of a motor task which did not involve the main somatic effectors of the exercise protocols (pointing movements with the upper limbs). Results: Mental chronometry showed that MI ability was degraded only after Intermittent exercise, while self-report measures of MI vividness revealed that MI ability was primarily impaired during MI of the walking task. Conclusions: Present results suggest that Intermittent exercise engaging anaerobic processes of energy expenditure, but not Continuous exercise engaging aerobic processes of energy expenditure, impaired MI ability. Findings are discussed in relation to the internal models theory of motor simulation, specifically changes in current state of the motor system under the fatigued state-affecting motor predictions. Present findings may contribute to successful applications of MI training in sports and rehabilitation.