HABITUATION AS A BEHAVIORAL PARAMETER TO MODELING SEX DIFFERENCES IN PARKINSON’S DISEASE PROGRESSION AND RECOVERY USING 6-OHDA

Habituation, the progressive reduction in response to repeated stimuli, offers a window into underlying neural adaptability. We aimed to quantify habituation across multiple behavioral domains in a Parkinson’s disease (PD) rat model and assess whether decay patterns predict sex or injury severity. Young adult male and female Wistar rats underwent neurotoxic lesions in the substantia nigra with 6-OHDA of varying severity and were evaluated across four sessions (baseline, post-injury, pre- and post-amphetamine) in an open field test. Minute-by-minute data were collected for locomotion, rearing, scanning, and turning. For each domain and session, we modeled behavior using an exponential decay function (y = a·e⁻ᵏᵗ), extracting intercept (a) and decay constant (k) for each rat. These parameters showed high model fit and strong internal consistency. Principal Component Analysis (PCA) followed by k-means clustering on decay parameters revealed distinct behavioral subtypes, yet clusters did not align with injury severity. Notably, logistic regression and random forest classifiers showed that decay profiles accurately predicted biological sex (accuracy 100%) but poorly predicted lesion group (accuracy 20%). Locomotion decay constant (k) emerged as the top predictor of sex. These findings suggest that behavioral habituation profiles are shaped more strongly by biological sex than lesion severity, and that sex-specific neuroadaptive strategies may underlie exploratory behavior following injury. Our framework provides a compact, interpretable representation of longitudinal behavior suitable for clustering and supervised learning. Habituation modeling may thus serve as a sensitive phenotypic marker to disentangle sex differences in PD progression and recovery.