Does Measuring Social Attention Lead to Changes in Behavior? A Preliminary Investigation into the Implications of...

A welfare assessment tool in development must satisfy several criteria before it is considered ready for general use. Some tools that meet many of these criteria have been criticized for their negative effect on welfare. We conducted a preliminary assessment of the impact of attention bias (AB) trials using threat-neutral conspecific face pairs followed by presumed neutral-positive filler stimuli on the behavior of 21 rhesus macaques (Macaca mulatta; 15 female). Behavioral observations were conducted following AB trials and repeated two weeks later when no AB trials had occurred (no trial: NT). The association between observation period and behavior was assessed using linear mixed-effects models in R. Trials did not impact any observed behavior except for fear, which was displayed by five monkeys over six trials (four NT). For this sample, there was a significant reduction in fear behavior following AB trials. We, therefore, found no evidence suggesting that AB trials negatively affect behaviour. AB protocols may be suitable for continued development for primate welfare assessment and we encourage researchers to include assessing test impact on welfare in their AB protocols.