To change or not to change experimenters: caveats for repeated behavioural and physiological measures in Montagu's harrier

Animals facing anthropogenic disturbances may exhibit different anti-predator responses depending on their perception of the risk. Experimental measures of behavioural and physiological traits may be impacted by the disturbance due to experimenter handlings and activities. In this study, we assessed the behavioural and physiological responses of Montagu's harrier chicks Circus pygargus from 14 nests, visited four times from hatching to fledgling by either the same (group I) or different (group NI) experimenters. Escape, aggressiveness and stress-induced corticosterone concentration increased for both groups over time but chicks from group I showed a higher increase in their behavioural and physiological responses than chicks from group NI. This increase occurred after the second visit when experimenters took the first blood sampling on nestlings. Handling by the same experimenters throughout the study did not result in habituation but in sensitization. Our results suggest that Montagu's harrier nestlings may discriminate between the people who visit their nest and respond to experimenters' identity. Therefore, we urge researchers working on behaviour and/or the physiology of stress including repeated measurements (such as animal personality studies for instance) to consider whether the experiments should or should not be conducted by the same experimenters (as usually done to minimize bias) as it can influence the outcomes of the experiments.