This month in canine science – August 2026

Every month I go through what’s been published in canine science, looking for studies that might shift how we understand our dogs.

August brought something I find exciting: researchers are getting better at looking inside the canine mind.

Here are the studies that caught my attention this month.

What does your dog’s brain see when they look at your face?

We already know that dogs are remarkably good at reading us. They respond to our voices, body language, facial expressions and even our scent.

But what happens inside a dog’s brain when they look at an emotional human face?

Researchers from the University of Vienna, together with colleagues in Hungary and Mexico, used fMRI brain scans and machine learning to investigate exactly that question. Twelve awake, trained family dogs were shown photographs of people displaying four different emotions: happiness, anger, fear and sadness.

The researchers first looked specifically at happy versus neutral faces. Happy human faces produced increased activity in the dogs’ temporal cortex, extending into the caudate nucleus, a region involved in reward processing. That suggests that seeing a happy human face may have particular emotional significance for dogs.

When the researchers looked at the dogs’ whole-brain activity patterns, they found that the dogs could distinguish not only between positive and negative expressions, but also between different negative emotions. The brain activity patterns allowed the researchers to distinguish fear from anger and sadness.

This doesn’t mean that dogs experience fear, anger or sadness in exactly the same way humans do. And the study used still photographs of human faces, whereas real-life communication involves many other signals: voice, posture, movement and smell.

But I find method fascinating: Instead of watching a dog react and trying to interpret what that reaction means, researchers were able to look directly at patterns of activity across the dog’s brain.

Source:

Hernández-Pérez, Raúl et al. “Dog brain representations of human facial expressions: Encoding happiness and differentiating specific negative expressions.” iScience 29 8 (2026): 116900 .

But… what happens if a face says one thing and a voice says another?

Dogs can tell when your face and voice don’t match

Another August study took the human-dog communication question one step further.

Researchers used non-invasive EEG to investigate how dogs respond when their owner’s facial expression and voice send conflicting emotional signals. For example, a happy face paired with an angry voice.

Only three dogs took part, so this is very much a pilot study. But the researchers found that emotionally incongruent audiovisual combinations produced larger late positive EEG responses than congruent combinations. In other words, the dogs’ brains responded differently when the emotional information from the owner’s face and voice didn’t agree.

The researchers are careful not to label the response, the precise functional interpretation of these EEG signals in dogs remains uncertain. Still, the basic finding is fascinating.

We all do this. You call the dog back in a bright, sing-song voice while your face and shoulders say “I am so done with you right now”. And they pick up both messages.

Source:

Koike, H., Kubo, T., Maruno, Y. et al. Event-related potentials to emotional incongruence in dogs using non-invasive EEG. Sci Rep (2026). https://doi.org/10.1038/s41598-026-66970-8

And what if there’s no expression to read at all?

Your dog may notice more about your emotional state than you realize

We already know dogs are remarkably good at reading human emotional cues. But what happens when those signals are much more subtle?

Researchers from several Japanese institutions, including Azabu University and the Nara Institute of Science and Technology, tried to find out. Owners were placed in either a stress or relaxation condition. The researchers then collected facial video, voice and odour samples from the owners while they performed a standardised reading task.

The dogs were subsequently exposed to these different combinations of cues while researchers recorded both their behaviour and heart-rate variability.

The results suggest that dogs responded differently depending on the context and combination of signals they received.

Interestingly, the researchers did not find clear differences in the dogs’ heart-rate variability between conditions. They did, however, observe a tendency toward greater physiological synchrony between owners and dogs when dogs were exposed to stress-derived stimuli.

Dogs may be responding to subtle combinations of information about their owner’s physiological and emotional state, even when there isn’t an obvious facial or behavioural expression to react to.

So your dog might not need you to look visibly upset to pick up that something has changed.

Source:

Ikeda R, Saito M, Iwata Y, Ozono T, Alemayoh TT, Kubo T, Izawa E-I, Fujiwara K, Yamakawa T, Ohno K, Kikusui T and Nagasawa M (2026) Dogs display context-dependent behavioral responses to subtle owner-derived multimodal stimuli. Front. Vet. Sci. 13:1916007. doi: 10.3389/fvets.2026.1916007

And then there’s the channel we still seem to underestimate.

The nose

If there is one sensory system we probably underestimate when thinking about canine cognition, it’s the obvious one: the nose.

A new study published in Scientific Reports adapted the human “Sniffin’ Sticks” approach for dogs, creating a standardised behavioural olfactory test. The researchers tested 80 healthy dogs and established reference values for their ability to detect the odour eugenol.

The results were remarkable. Dogs detected concentrations of eugenol below 84 parts per quadrillion in the empirical testing. That’s an almost absurdly tiny amount. The researchers found that olfactory performance differed with age.

Adult dogs performed better than senior dogs, suggesting that olfactory ability may follow an inverted U-shaped trajectory: improving from juvenile to adulthood, then declining in old age.

Could changes in smell become a useful indicator of neurological or age-related changes?

The study itself does not establish this as a diagnostic test for dementia. The dogs tested were healthy, and dogs with cognitive dysfunction were excluded. But the researchers specifically point toward the potential clinical value of standardised olfactory testing.

And that makes this particularly interesting in the context of canine cognitive aging. Perhaps one day, assessing an older dog’s brain health won’t only involve watching how they behave.

Source:

Hagemann, L.H., Volk, H.A., Prielipp, M. et al. Canine “Sniffin’ Sticks” enable behavioural olfactometry to complement neurological examination in dogs. Sci Rep 16, 25894 (2026). https://doi.org/10.1038/s41598-026-60466-1

Can we measure cognitive control in a dog’s brain?

Imagine asking a dog to wait for something they really want. For us, that’s a familiar test of self-control. But what if, instead of simply observing whether the dog manages to wait, we could see what happens in his brain while he does so?

Researchers from the HUN-REN Research Centre for Natural Sciences and ELTE Eötvös Loránd University in Hungary did just that. In a study published in Animal Cognition, they recorded EEG activity from 14 family dogs during two conditions: relaxed wakefulness and a delayed-gratification challenge.

The researchers found an increase in neural activity in the 5–7 Hz theta range, particularly over the frontal midline, during the self-control task. This resembles an EEG signature associated with cognitive control in humans. They also observed a reduction in broadband power and a flattening of the EEG spectrum, findings that the authors interpret as potentially involving frontoparietal connections.

There is an important caveat here: the researchers are not saying they have found a canine equivalent of human conscious self-control. The study was small, and the EEG markers cannot distinguish between different facets of cognitive control.

But we are getting closer to being able to study canine cognition without having to infer everything from behaviour alone.

Source:

Iotchev, I.B., Kis, A. & Gácsi, M. A human-like EEG signature of cognitive control in the domestic dog. Anim Cogn (2026). https://doi.org/10.1007/s10071-026-02096-8

Dogs protect seniors from dementia

A large study from Japan followed 11,224 older adults (average age 74) for about seven and a half years and found that dog owners were far less likely to develop disabling dementia — the kind severe enough to leave someone needing daily help. Current dog owners had roughly half the risk, about 47% lower, compared with people who had never owned a dog. The researchers suspect the everyday demands of dog ownership do the work: regular walks add moderate exercise, a dog pulls its owner into social contact, and there may be biological effects too, from oxytocin to changes in gut bacteria from more time spent outdoors. They also ran the numbers on cost and found that, from a public payer’s perspective, dog ownership was actually cost-saving — it lowered medical and long-term-care spending while adding healthy years of life. Two things worth keeping in mind, though: the clear protection showed up in people who currently live with a dog, while former owners showed at most a weak, non-significant trend for dementia (they did, however, have a modestly lower risk of death). And the authors stress this is an association — dogs are linked to lower dementia risk, but the study can’t prove they cause it.

Source:

Taniguchi, Y.; Saito, Y.; Seino, S.; Hata, T.; Mori, H.; Kobayasi, E. Investigating the Potential Protective Effect of Dog Ownership on Incident Disabling Dementia and Its Cost-Effectiveness. Int. J. Environ. Res. Public Health (2026), 23, 938. https://doi.org/10.3390/ijerph23070938

Your dog is afraid of new things? Keep calm and carry on may be your best response

When your dog meets something unfamiliar and unsettling, the instinct is to step in — pet them, talk softly, offer a treat, or coax them toward the scary thing. A new study put those instincts to the test. Researchers watched 48 dog–owner pairs as each dog was introduced to a mildly stressful novel object — a remote-controlled car — for one minute. Owners were assigned one of four responses: petting with a soothing voice, encouraging the dog to approach the object, offering treats, or simply staying passive and doing nothing. The result was humbling for well-meaning owners: none of the active strategies reduced the dogs’ stress responses any more than doing nothing at all. If anything, dogs whose owners urged them to approach looked more stressed at first. The interventions mostly changed how close the dog stayed and how it behaved socially — not how stressed it actually felt. The authors are careful to note this was a brief, mild scare, so repeated or longer exposures might tell a different story. But for a one-off startle, staying calm and unbothered yourself may be as good as anything you can do.

Source:

Miller, J., Cavalli, C., Azadian, A. et al. Immediate behavioural outcomes of owner-delivered support during novel object exposure in dogs. BMC Vet Res (2026). https://doi.org/10.1186/s12917-026-05837-6

For a long time, canine cognition research was dominated by behavioural experiments. Researchers watched what dogs did and tried to infer what was happening inside their minds. But we’re increasingly getting additional windows into the dog:

The science of canine cognition is becoming increasingly measurable.

Which of these surprised you most?

For me it is that dogs can pick up the mismatch in the emotions we show and it forces me to rethink how I approach difficult moments.