Correcting of the lack of regularity in steps is a key component of gait rehabilitation in Parkinson's disease. The proposal is to introduce adaptive spatial auditory cueing (ASAC) based on verbal instruction "lengthen the step" automatically delivered when the stride length decreased below a predetermined threshold. The present study compared the effect of usual rhythmic auditory cueing versus ASAC used during a walking training in Parkinson's disease.

Correcting of the lack of regularity in steps is a key component of gait rehabilitation in Parkinson's disease. The proposal is to introduce adaptive spatial auditory cueing (ASAC) based on verbal instruction "lengthen the step" automatically delivered when the stride length decreased below a predetermined threshold. The present study compared the effect of usual rhythmic auditory cueing versus ASAC used during a walking training in Parkinson's disease. Fifteen patients with Parkinson's disease performed both interventions in randomized order, one week apart: a 20-minute walking training with rhythmic auditory cueing, in form of a metronome adjusted on 110% of the patient's own cadence, or ASAC delivered when the stride length is less than 110% of the patient's own stride length. Assessment criteria were walking distance covered during the intervention, speed, step length, cadence, coefficients of variation of step length and step duration, and indexes of spatial and temporal asymmetry during a walking test before and just after the intervention.

Inclusion Criteria:

• the diagnosis of idiopathic PD based on the United Kingdom-Parkinson's Disease Society Brain Bank criteria;
• stage 2 or 3 on the Hoehn and Yahr scale;
• comfortable walking speed over 10 meters ≤ 1 m/s;
• ability to walk over 20 minutes without aids or antiparkinsonian medications;
• a stable antiparkinsonian medication regime;
• cognitive abilities to understand the verbal instructions for a walking test according to the investigator's judgment;
• written consent for the participation.

Exclusion Criteria:

• intercurrent disease other than PD, affecting gait;
• any intercurrent medical condition preventing them from participating in two consecutive gait training one week apart;
• medical diagnosis of hearing loss;
• non-affiliation to the social security regime.

A biomechanics engineer leads the gait assessment, being blinded to the allocation schedule, and participants were specifically asked to not discuss the intervention.