Evaluating enrichment use in group-housed rhesus macaques (Macaca mulatta): A machine learning approach

Abstract

Environmental enrichment programmes are widely used to improve welfare of captive and laboratory animals, especially non-human primates. Monitoring enrichment use over time is crucial, as animals may habituate and reduce their interaction with it. In this study we aimed to monitor the interaction with enrichment items in groups of rhesus macaques (Macaca mulatta), each consisting of an average of ten individuals, living in a breeding colony. To streamline the time-intensive task of assessing enrichment programmes we automated the evaluation process by using machine learning technologies. We built two computer vision-based pipelines to evaluate monkeys' interactions with different enrichment items: a white drum containing raisins and a non-food-based puzzle. The first pipeline analyses the usage of enrichment items in nine groups, both when it contains food and when it is empty. The second pipeline counts the number of monkeys interacting with a puzzle across twelve groups. The data derived from the two pipelines reveal that the macaques consistently express interest in the food-based white drum enrichment, even several months after its introduction. The puzzle enrichment was monitored for one month, showing a gradual decline in interaction over time. These pipelines are valuable for assessing enrichment by minimising the time spent on animal observation and data analysis; this study demonstrates that automated methods can consistently monitor macaque engagement with enrichments, systematically tracking habituation responses and long-term effectiveness. Such advancements have significant implications for enhancing animal welfare, enabling the discontinuation of ineffective enrichments and the adaptation of enrichment plans to meet the animals' needs.

Keywords: Animal welfare; computer vision; deep learning; enrichment; laboratory animals; macaques.

Figure 1.

Types of enrichment used for the studies showing (A) white drum used for food enrichment, (B) puzzle enrichment with blue monkey toy and (C) revised puzzle enrichment without blue monkey toy.

Figure 2.

Automated pipelines used to detect enrichment usage by groups of macaques. These figures illustrate the steps involved in the automated detection of white drum enrichment usage (A to C) and puzzle enrichment usage (D to F). More specifically these show (A) the input frame from video for white drum enrichment, (B) segmentation mask and bounding box (bbox) for the white drum (circle shows the centre-point [CP] of the bounding box, (C) example of x co-ordinate from the bounding box with a threshold (dotted line), (D) input frame from video for puzzle enrichment, (E) segmentation masks for monkeys (M and unlabelled white objects) and front platform (FP) together with region of interest (ROI) and (F) number of monkeys interacting with the puzzle enrichment.

Figure 3.

White drum enrichment use across nine macaques groups, each with a mean (± SD) group size of 10 (± 2.9) individuals. Boxplots showing the number of frames where the drum moved, caused by macaques’ integration with it, on Mondays and Thursdays for each interval after food was given. Notably, raisins were only placed into the drum on Mondays, while feeding time occurred in the morning every day.

Figure 4.

Interaction with puzzle enrichment by twelve groups of macaques each with a mean (± SD) group size of 10 (± 2.8) individuals showing (A) average number of interactions with the puzzle enrichment per group per hour plotted against days since added (solid line shows fit from the mixed effect model output), (B) average number of interactions with the enrichment per group per hour when the blue monkey toy was present or absent and (C) average number of interactions with the enrichment per group per hour relative to time of day (solid line shows fit from the mixed effect model).