Reading books to shelter animals combining auditory enrichment with human presence is increasingly used although its effects on animal welfare have not yet been investigated. This study compared the behaviour of single-housed shelter dogs and cats during a prerecorded reading condition in the absence or presence of an unfamiliar human (without direct physical contact). Fourteen dogs and twenty-one cats were observed in their enclosure in the two conditions in a counterbalanced order. Behaviours such as scratching the door, gaze direction and location in relation to the audio source/human were analysed from video recording for 10 min per condition. Dogs spent more time in their bed (p < 0.047) and looking at the auditory source (p < 0.004) when a human was present. Cats showed door scratching and rubbing when a human was present (p < 0.043), whereas they tended to spend more time in the vertical dimension (p = 0.051), where the hiding boxes were located, during auditory stimulation without a human present. These results show that the presence of a human induces greater interest compared to just audio stimulation in shelter dogs and cats but may induce frustration likely due to not being able to physically interact in some animals.

Pet dogs are promising candidates to study attachment-related and potentially jealousy-like behaviours in non-human animals, as they form a strong and stable bond with their human caregivers who often engage in affiliative interactions with diverse social partners. Nevertheless, it is still debated whether non-human animals are capable of experiencing such complex emotions. Even though caregivers frequently report observations of jealousy-like behaviours in dogs, behavioural studies in dogs have thus far led to contradictory results. Adding to this complexity, dogs appear extraordinarily skilled in understanding humans’ communicative behaviour and can flexibly and diversely interact with them in social contexts. Here, we aimed at investigating (1) whether dogs indeed respond in a jealousy-consistent manner when seeing their caregiver interact in an affiliative way with a remotely controlled, realistic-looking fake dog, or (2) whether they would rather synchronize their reaction to the fake dog with the caregiver’s behaviour, or (3) whether they respond directly to the caregiver without paying much attention to the third party. To address what drives the dogs’ behaviours in this triadic situation, we compared four groups of dogs who first observed and then joined the interaction of either the caregiver or a stranger greeting or medically examining the fake dog. We found that the dogs initially responded negatively or neutrally when the fake dog entered the room but changed to more positive reactions when the caregiver approached the fake dog, especially if initiating a positive interaction. When being released, more dogs showed friendly behaviours towards the fake dog when the caregiver—rather than the stranger—was interacting with it. At the same time, however, the dogs tried to block the interaction of the caregiver with the fake dog more often than the one of the stranger. In conclusion, we did not find clear evidence for jealousy-like behaviours in dogs during the human–fake dog interactions, but we observed indicators of behavioural synchronization with the caregivers, suggesting that the caregivers’ affiliative behaviours directed at a third party may more often facilitate positive than negative interactions in dogs.