Aim: To investigate the feasibility of syncopated (anti-phase) gait, particularly at different walking cadences. We expect it to be more successful at preferred speed because of central pattern generators (CPGs), subcortical neural networks which contribute to gait’s automaticity. Study Design: Research protocol Place and Duration of Study: Faculty of Engineering, University of Bologna, Cesena, Italy, May 2011 to March 2012. Methodology: Ten healthy subjects walked to an auditory metronome beat set to 85%, 100% and 115% of their preferred cadence in a series of one-minute trials. They were instructed to walk in time to the beat for the in-phase condition, so that their heel strike coincided with the beat. For the anti-phase condition, they were instructed to time their heel strike to occur midway between beats. Anterior-posterior acceleration signals, used to calculate heel strike, were measured by an accelerometer worn on the back. The variables cadence variability, jerk (the derivative of acceleration) and harmonic ratio (HRa measure of gait symmetry and consistency) assessed gait perturbation; phase shift, in degrees, quantified the difference between beats and heel strikes for each step cycle (360o). Results: Anti-phase gait is possible (average timing of inand anti-phase heel strikes was 350.3±44.8o and 173.4±65.7o, respectively; P<.001). Cadence variability was lowest Short Research Article Mayberry and Chiari; JSRR, Article no. JSRR.2014.009 1118 at the preferred speed for in-phase gait (slow 12.32±1.16%, preferred 7.54±0.49%, fast 10.85±1.20%, P<.001), but no different for anti-phase gait (slow 12.75±1.04%, preferred 13.84±0.98%, fast 13.26±1.07%, P=.75). Conclusion: Decreased variability at preferred speed is an indication of CPG facilitation; this was observed during in-phase but not anti-phase gait. The results provide no evidence that CPGs operate preferentially at preferred speed to contribute to locomotor control during anti-phase gait.