Why do we see different things in the same dog?

Home » Latest news » Why do we see different things in the same dog?

Why do people watching the same dog often reach very different conclusions? Recent research suggests that the way we interpret behaviour may shape what happens next, sometimes before we understand what the animal is experiencing.

Spend a few minutes reading the comments beneath a social media video of a dog growling towards a visitor, and you will soon see how differently people interpret the same behaviour. One person is certain the dog is frightened, another describes them as angry, while others blame poor training, aggression or an irresponsible owner. Everyone has watched the same behaviour, yet they have reached very different conclusions.

Building explanations from limited information is a distinctly human trait, especially when a situation feels emotionally charged, such as when a dog growls towards a visitor or a cat swipes at another animal or person. People naturally reach for labels: a dog becomes jealous, stubborn or anxious; a cat is aloof, unfriendly or spiteful. We tend to use these descriptions to make sense of behaviour that can feel unpredictable or worrying, yet problems arise when an interpretation becomes accepted as fact. Once we settle on an explanation, confirmation bias may lead us to notice evidence that supports our existing interpretation while paying less attention to information that points in another direction (Nickerson, 1998).

A recent study by Holloway and Kisley (2026) examined whether the emotions people attributed to cats and dogs were associated with the responses they said they would consider. The researchers recruited 194 adults living in the United States and presented each participant with two written scenarios, one involving a recently adopted dog and another involving a recently adopted cat.

What the new research found

In the lower-intensity dog scenario, an unfamiliar visitor approached and reached towards the dog, who stiffened, pulled back, growled and moved away. The higher-intensity version described the dog baring their teeth, pinning back their ears, staying close and barking as well as growling. Equivalent descriptions were created for cats. Participants rated fear and anger independently before judging how likely they would be to move the animal to another room, consult a trainer, consider medication, physically punish them, privately rehome them or relinquish them to a shelter.

Fear received higher ratings overall, although ratings of anger rose as the defensive behaviour became more intense. Higher fear ratings were associated with a greater reported likelihood of moving either species away from the situation. Higher anger ratings were associated with a greater reported likelihood of choosing physical punishment towards cats and relinquishment or rehoming for dogs. Participants who gave greater moral weight to animal emotions were less likely to endorse several disruptive or aversive responses (Holloway and Kisley, 2026).

These findings describe associations between people’s ratings and their stated intentions; they cannot show that interpreting an animal as angry caused somebody to choose a harsher response. Participants were also considering hypothetical written scenarios, so it remains unknown whether those intentions would translate into real decisions involving their own animal in a similar context. While the vignettes allowed the researchers to standardise the information presented, they could not provide the fuller behavioural sequence, medical and learning histories, relationships, or wider environmental context that would form part of a typical clinical animal behaviour assessment.

Observation and interpretation are different skills

One of the first things I explain to clients is that observation and explanation are different skills. Noticing that a dog stiffened, turned their head away or growled is observation. Deciding why they did so is interpretation. Many owners I work with are highly observant and may spot subtle changes in posture, expression or routine. The challenge comes when observation turns into explanation, because that explanation can shape every decision that follows, including how the animal is approached, handled, trained or managed.

Professor Daniel Mills has also drawn attention to the difference between knowing canine body-language signals and developing the practical skill of reading dogs. During my recent conversation with him and Paul Davis on the Everything’s Psychology podcast, he explained that this skill develops through repeated observation and immersion, rather than by memorising isolated body-language signals. Recent experimental work also supports a cautious approach: people can confuse canine emotional expressions, while the relative contribution of facial and bodily cues differs according to the emotion being judged (Guo, Correia-Caeiro and Mills, 2024).

Then there is frustration, which adds another layer to how we interpret the behaviour. Holloway and Kisley focused on fear and anger because both are commonly proposed motivations for defensive behaviour, yet frustration was outside their study. McPeake and colleagues describe frustration as a negative emotional response following the violation of an expectation. Their Canine Frustration Questionnaire identified tendencies associated with barriers, restraint, unmet expectations, reduced control and difficulty coping when access to something valued is blocked (McPeake et al., 2019). Later testing found associations between CFQ scores, selected behavioural responses and salivary cortisol measures, providing further evidence for the questionnaire’s validity (McPeake et al., 2021). Even so, no single display can establish that frustration is the underlying process.

Consider a dog barking and lunging at the end of a lead. Fear may be involved, frustration may contribute, or the dog may be experiencing pain, conflict, social motivation or several processes at once. Handling, previous reinforcement, early experience, daily routines, health status and predisposing characteristics can each alter the behaviour we see. That is why one of the most useful questions behaviour and training practitioners can ask is also one of the simplest: what else could explain this?

Holloway and Kisley’s study tells us far more about human judgement than it does about animal anger. Behaviour professionals spend much of their time helping people separate what they observed from what they inferred. That distinction may appear subtle, yet it can change the direction of a consultation and the way an animal is treated. Before deciding what an animal is feeling, pause and ask yourself a simple question: what else could explain this?

If you are struggling with your pet’s behaviour, please get in touch at [email protected], or find a qualified practitioner near you through the Animal Behaviour & Training Council’s Practitioner Directory.

By Hanne Grice, Clinical Animal Behaviourist, Certified Animal Trainer and Animal Sciences Lecturer

References

• Guo, K., Correia-Caeiro, C. and Mills, D.S. (2024) ‘Category-dependent contribution of dog facial and bodily cues in human perception of dog emotions’, Applied Animal Behaviour Science, 280, 106427. Available at: https://doi.org/10.1016/j.applanim.2024.106427
• Holloway, A.E. and Kisley, M.A. (2026) ‘Reading between the whines: Human perceptions and beliefs about animal emotions predict how people would intervene with cats and dogs showing challenging behaviors’, Frontiers in Psychology, 17, 1857104. Available at: https://doi.org/10.3389/fpsyg.2026.1857104
• McPeake, K.J., Collins, L.M., Zulch, H. and Mills, D.S. (2019) ‘The Canine Frustration Questionnaire: Development of a new psychometric tool for measuring frustration in domestic dogs (Canis familiaris)’, Frontiers in Veterinary Science, 6, 152. Available at: https://doi.org/10.3389/fvets.2019.00152
• McPeake, K.J., Collins, L.M., Zulch, H. and Mills, D.S. (2021) ‘Behavioural and physiological correlates of the Canine Frustration Questionnaire’, Animals, 11(12), 3346. Available at: https://doi.org/10.3390/ani11123346
• Nickerson, R. (1998) Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2, 175-220. Available at: https://doi10.1037/1089-2680.2.2.175.