Inspiring systematic inclusion of individual animal states to enhance the quality of research

Janire Castellano Bueno , Vittoria Elliott

Studies on animals continue to attract criticism over data quality, reproducibility and generality of findings, yet one source of variation remains rarely addressed: differences in individuals’ affective states. In this paper, we suggest that evaluating affect should be considered standard good practice in ecological and behavioural research with wild animals, alongside familiar variables such as sex, age or dominance rank. Affective states, viewed here as integrated products of animals’ past and current experiences, influence how individuals perceive and respond to experimental treatments, environmental conditions and human activities, and can systematically alter behavioural, physiological and ecological outcomes.
We first place affective states within a review of existing discussions about replication, bias and rigour in animal research, and argue that unaccounted for differences in affect can generate unexplained variation, apparent outliers, selective exclusion of individuals and difficulties in reproducing results across sites or studies. By reviewing relevant animal welfare science research, we show how behavioural, somatic, physiological and cognitive indicators can be used to evaluate welfare and infer affective states in both captive and free-ranging animals, increasingly with minimally invasive or remote methods. Building on this, we outline practical ways that differences in affect can be inferred by incorporating welfare assessments into wildlife studies. By systematically collecting welfare-relevant data throughout a project, affective status can be used as an explanatory or control variable in statistical models, to detect sampling biases, and to interpret behavioural patterns.
Finally, we discuss ethical and logistical implications. Integrating welfare assessments has the potential to support the 3Rs by reducing unnecessary repetition, increasing the amount learned from each animal and helping to identify welfare problems earlier. Despite challenges such as limited species-specific indicators and added analytical complexity, routinely accounting for affective states is a feasible, important step toward more robust, informative, and ethically defensible research with wild animals.