Assessing Self- and Social-grooming Dynamics as Potential Behavioral Biomarkers of Estrus in Precision-managed Dairy Herds

Abstract Understanding how affiliative or maintenance behaviors shift during reproductive activity can enhance behavioral monitoring of estrus expression in precision-managed dairy herds. The objective of this study was to evaluate whether self-grooming (automated brush use) and social grooming (allogrooming) behaviors are associated with estrus expression in dairy cows. Behavioral data were collected from ten lactating Holstein cows (n = 10) housed in a single freestall. Prior to trial initiation, cows were synchronized using a single injection of Lutalyse (Dinoprost tromethamine 5mg/mL; Zoetis). Continuous video was recorded for seven days post-injection using a four-camera CCTV system, yielding 168 hours of footage for behavioral analysis. Individual brush use and allogrooming events were annotated using the BORIS behavioral analysis software and an ethogram-based framework that documented start and stop times, anatomical modifiers (e.g., head, shoulder), and directional components of social grooming (providing = PRV; receiving = REC). Behavioral estrus was concurrently scored from daily observations of mounting, standing to be mounted, chin-resting, sniffing, and restlessness. A daily estrus activity score was calculated as the sum of these behaviors for each cow and merged with brush and allogrooming metrics. All behavioral metrics were scored by a single observer. Total durations and frequency counts were aggregated per cow per day and analyzed to assess behavioral associations with estrus activity. Significant correlations (r = 0.86–0.90; p 0.001) were observed among classical estrus behaviors, including standing to be mounted, being mounted, chin-resting, and tailside mounting, confirming the reliability of the behavioral scoring system. Brush use demonstrated a mild but positive correlation with estrus activity (r ≈ 0.33; p 0.01), suggesting that cows may intermittently engage in self-grooming during periods of increased social and locomotor activity. In contrast, both providing and receiving allogrooming behaviors were not correlated with estrus scores (r = −0.06 to 0.06; p 0.50), indicating that social grooming frequency and duration were not reliable indicators of reproductive status. Directional differences in allogrooming (PRV vs. REC) did not influence this outcome, and anatomical modifiers revealed no consistent trends. Overall, these findings suggest that while automated brush use may slightly fluctuate with general behavioral activation around estrus, social grooming behaviors remain largely independent of reproductive state. Monitoring classical estrus behaviors remains the most effective approach, whereas brush or allogrooming metrics may better reflect welfare or social stability versus reproductive readiness.