Dystonia is a neurological disorder characterized by involuntary repetitive movements and abnormal postures. Animal models have played a pivotal role in studying the pathophysiology of dystonia. However, many genetic models, e.g., the Tor1a+/ΔE (DYT1) mouse, lack an overt motor phenotype, despite significant underlying neuronal abnormalities within the striatum and other motor control regions. Because the striatum is implicated in action sequencing, it is possible that the behavioral defect arises as a disruption in the frequency and temporal ordering of behaviors, rather than execution, which cannot be captured using traditional behavioral assays, thus limiting drug discovery efforts. To address this challenge, we used MoSeq, an unsupervised behavioral segmentation framework, to compare the continuous free behavior of control Tor1a+/+ mice and knockin Tor1a+/ΔE mutant mice in response to the anti-dystonia drug trihexyphenidyl. Although minimal baseline differences in behavioral organization were detected, both genotypes exhibited robust and consistent shifts in behavioral space structure after treatment with trihexyphenidyl. Further, we demonstrate differences in the behavioral space structure of male vs. female mice after trihexyphenidyl challenge. The distinct behavioral signatures evoked by trihexyphenidyl and biological sex, a known risk factor for dystonia, suggest that the analysis of the temporal structure of continuous free behavior provides a sensitive and novel approach to the discovery of therapeutics for the treatment of dystonia.