Summary: A new study demonstrated that dogs possess specialized neural networks for processing human facial expressions. The team used functional magnetic resonance imaging (fMRI) and machine learning to analyze canine brain activity while pet dogs viewed images of human faces depicting happiness, anger, fear, and sadness.
The researchers discovered that happy human faces selectively activate a neural network encompassing the temporal cortex (involved in higher cognitive and sensory processing) and the caudate nucleus (a key reward center). Whole-brain analyses revealed that canine brains do not merely categorize human expressions into positive or negative valences; dogs generate distinct, unique brain activity patterns that differentiate fear from anger and sadness.
The findings provide the first direct neural evidence that thousands of years of domestication have wired the canine brain to decode subtle human emotional signals.
Key Facts
- Reward Center Activation for Smiles: Viewing happy human faces triggered significant activation in the canine caudate nucleus, the brain’s primary reward processing center, suggesting human smiles hold intrinsic emotional and motivational value for dogs.
- Discrimination Between Negative Emotions: Using machine learning pattern analysis, researchers proved that dogs do not simply lump all bad moods together; their brains display distinct, distinguishable neural signatures when viewing human fear, anger, and sadness.
- Convergent Evolutionary Adaptation: Unlike primates, who share a recent common ancestor with humans, dogs developed sophisticated human-decoding social skills independently through thousands of years of convergent evolution and artificial selection during domestication.
- Temporal Cortex Specialization: The temporal cortex in dogs plays a central role in visual face processing and higher-order social cognition, forming a functional circuit with the caudate nucleus specifically tuned to human facial cues.
- Multimodal Real-World Integration: While static photographs successfully elicited distinct brain responses, the authors emphasize that dogs in real-world settings integrate facial expressions with body posture, vocal tone, and olfactory signals.
Source: Cell Press
Scientists—and dog lovers—have long known that dogs are remarkably good at interpreting our laughs, frowns, and tears. But the biology behind this ability has remained largely unexplored until now.
In a study published in the Cell Press journal iScience on August 7, researchers peered inside the brains of dogs as they viewed images of various human expressions. The brain scans revealed that dogs process smiles in a unique brain region and, for the first time, provided evidence that dogs can distinguish between negative facial expressions, including anger, sadness, and fear.
“Humans are really good at picking up small details on the face, and so are dogs,” says first author Raúl Hernández-Pérez from the University of Vienna. “This makes dogs very interesting. By studying their brains, we can understand what adaptations have developed to support their remarkable social and emotional understanding of humans.”
The researchers first focused on happiness, an expression that previous studies suggest holds particular significance for dogs. They scanned eight dogs while showing them images of strangers with either happy or neutral expressions. The happy faces lit up the temporal cortex and caudate nucleus in the dogs’ brains—regions associated with higher cognitive and reward processing, respectively.
“The activity in the caudate nucleus is particularly interesting,” says Laura Cuaya, also of the University of Vienna, who began the research at the National Autonomous University of Mexico and later expanded it through a collaboration at Eötvös Loránd University in Hungary. “It suggests that happy human faces may have emotional significance for dogs.”
The team then asked whether the same brain regions respond to other emotions. They showed images of people displaying happiness, anger, fear, and sadness to 12 dogs and analyzed their brain activity using machine learning. They found that the dogs did not react in the same to these negative emotions; the temporal cortex-caudate nucleus network responded uniquely to happiness, allowing researchers to easily distinguish brain reactions to smiles from all three negative expressions.
But the canine brain was not merely sorting the world into good and bad, the team found. A whole-brain analysis revealed distinct patterns of activity that separated fear from anger and sadness, providing the first evidence that dogs can distinguish between specific negative facial expressions.
The researchers say that this ability does not mean that dogs experience emotions exactly as humans do. Also, a still photo of a human face doesn’t tell the whole story; in real life, dogs watch human body language, listen to nuances in our voices, and detect information from our scent.
“When we compare humans with primates, we are looking at abilities that may have come from a common ancestor,” says senior author Laura Cuaya of the University of Vienna. “With dogs, the story is totally different. They followed a very different evolutionary path, yet through domestication, they developed the social abilities needed to understand and cooperate with us.”
In this study, the canine participants were all pets from loving families, notes the team. Free-ranging dogs, such as those that live on the street, navigate a very different social world and may process human cues differently.
Next, the researchers plan to adopt a more dog-centric perspective to understand how dogs integrate these brain signals to make sense of the humans around them. Their future work will also compare how human and dog brains process the same emotional signals.
“I would like people to see dogs as emotional beings,” says Cuaya. “They have evolved with us to understand our expressions and cooperate with us, and recognizing that can change how we communicate with and care for them.”
Funding:
This work was supported by funding from the Austrian Science Fund, the European Research Council, the Hungarian Academy of Sciences, and the National Brain Programme 3.0.
Key Questions Answered:
A: The dogs involved in these studies are beloved family pets trained using positive reinforcement to lie completely still inside an fMRI scanner. They wear ear protection and voluntarily place their heads on custom chin rests while viewing screens, allowing researchers to capture real-time brain activity without stress or anesthesia.
A: The caudate nucleus is a key structure in the brain’s reward system, involved in processing positive expectations and pleasure. The fact that seeing a smiling human face activates the caudate nucleus indicates that human smiles carry genuine emotional significance and positive reward value for dogs.
A: No. The study demonstrates that a dog’s brain can distinguish between different human facial expressions (such as fear, anger, and sadness), but it does not prove that dogs experience or conceptualize those emotions identically to humans. It shows that domestication has equipped dogs with the neural machinery required to decode and respond to human emotional states.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this animal psychology and emotion research news
Author: Julia Grimmett
Source: Cell Press
Contact: Julia Grimmett – Cell Press
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Dog Brain Representations of Human Facial Expressions: Encoding Happiness and Differentiating Specific Negative Expressions” by Karl Kristjan Kaup, Javier Hidalgo Jiménez, Jaan Aru. iScience
DOI:10.1016/j.isci.2026.116900
Abstract
Dog Brain Representations of Human Facial Expressions: Encoding Happiness and Differentiating Specific Negative Expressions
Dogs can distinguish human facial expressions, particularly happiness, yet brain processes remain unclear. Using fMRI, we conducted two experiments in awake pet dogs.
In Experiment 1 (n = 8), happy faces elicited a stronger response than neutral faces in a right temporal cluster extending to the caudate nucleus, including the rostral Sylvian gyrus. In Experiment 2 (n = 12), dogs viewed faces expressing happiness, anger, fear, or sadness.
Using the Experiment 1 cluster as a region of interest, a machine-learning classifier distinguished happiness from each negative facial expression, but not between negative pairs, showing differential BOLD responsiveness to happy faces.
Whole-brain representational similarity analyses revealed activity patterns differentiating angry vs. fearful faces (right mid ectosylvian and left splenial gyri) and sad vs. fearful faces (right rostral suprasylvian gyrus) but not angry vs. sad faces. This provides direct evidence that dog brains can distinguish between two negative facial expressions.
