Gilles de la Tourette Syndrome (GTS) is a developmental disorder. Empirical studies and an emerging cognitive framework on GTS suggest that GTS is a disorder of abnormally strong ‘perception-action binding’. Theoretical considerations imply that the effectiveness of long-established behavioral interventions might be related to a normalization of increased binding in GTS. This has not been tested yet. We examined the effect of a standardized Comprehensive Behavior Intervention for Tics (CBIT) in N = 21 adolescent GTS patients and N = 21 healthy controls on perception-action binding in an inhibitory control paradigm. Prior to CBIT, GTS patients showed compromised performance compared to controls, specifically when inhibitory control was triggered by uni-modal visual compared to bi-modal stimuli. After CBIT intervention, GTS patient’s performance was at the same level as healthy controls. This is supported by a Bayesian data analysis. CBIT specifically affected inhibitory control in a condition where reconfigurations of perception-action bindings are necessary to perform inhibitory control. A power of 95% was evident for these effects. CBIT reduces increased ‘binding’ between perception and action in GTS and thereby increases the ability to perform response inhibition. The results are the first to provide insights as to why CBIT is effective by relating elements of this intervention to overarching cognitive theoretical frameworks on perception-action bindings.