Failing to recognize a perturbed feline can end in serious scratch marks. (Credit: Anna Hoychuk on Shutterstock)

In A Nutshell

• Nearly 1 in 4 people can’t recognize when a cat is showing obvious distress signals like hissing or aggressive postures, and accuracy drops to coin-flip levels for subtle stress cues.
• Even when people correctly identify a distressed cat, 20% still choose to engage rather than give the animal space, rising to 44% for cats showing subtle discomfort.
• Cat bites account for 75% of infection-causing mammalian bites, with 30% of cat bites becoming infected and potentially causing serious complications like bone infections or septic arthritis.
• A brief educational video improved recognition of obvious cat behaviors but paradoxically made people worse at spotting early warning signs, the exact signals needed to prevent escalation.

A cat hissing with ears flattened back and fur on end. You’d know to back away, right? Apparently not. Research from the University of Adelaide shows nearly one in four people misidentify cats displaying clear signs of distress. Moreover, when researchers tried to fix this problem with an educational video, they accidentally made people worse at recognizing the subtle feline warning signs.

Researchers tested 368 adults in Australia, showing them short video clips of people interacting with cats. Participants had to categorize each cat’s state as positive or negative. The results, published in Frontiers in Ethology, were eye-opening. When cats showed obvious distress signals like hissing, spitting, or aggressive postures, 23.3% of people still got it wrong. When the signals were subtle (slight tension, whisker changes, gradual stiffening), accuracy dropped to essentially a coin flip at 48.7%.

Among people who correctly identified that a cat was displaying obvious negative behaviors, nearly 20% still said they’d engage with the animal rather than retreat. For cats showing subtle distress, that number jumped to 44%. The encouraging news: when cats showed obvious distress, most people (91%) did choose to walk away at least once.

Still, some interaction choices were particularly dangerous. Close to half (42%) said they’d rub the belly of a cat already showing subtle signs of distress. That seemingly innocent gesture can trigger defensive responses involving all four paws, claws extended, plus teeth.

Why Cat Bites Are More Dangerous Than You Think

Cat bites can be painful, and they can also be medically serious. Three-quarters of all mammalian bites that result in infection come from cats. Their long, thin teeth create deep puncture wounds that seal quickly on the surface while bacteria multiply inside. About 30% of cat bites become infected, and the complications can be severe: deep abscesses, bone infections, septic arthritis, and in rare cases, conditions serious enough to cause long-term disability or death.

Cat scratches carry their own risks beyond corneal lacerations that can damage eyesight. Scratches transmit cat scratch disease (bartonellosis), which causes chronic lymph node swelling and has been linked to depression and schizophrenia. Scratches contaminated with saliva can transmit rabies, which remains fatal once symptoms appear.

For cats showing subtle negative signals, 73.6% of study participants said they’d stroke the cat. Playing with bare hands was chosen by 30.4%. Both actions frequently trigger defensive responses. Hands and arms account for roughly half of all cat-inflicted injuries, many occurring during play that crossed a line the human didn’t see.

A training video actually made people worse at recognizing the early warning signs of feline distress. (Credit: Magui RF on Shutterstock)

The Belly Rub Trap

Many cats don’t enjoy belly rubs, even when they roll over and expose their stomach. Rolling over can signal trust and an invitation to interact, but usually not through touching that vulnerable area. For many cats, the belly is easily overstimulated, prompting a swift defensive reaction. Yet 42% of participants viewing cats with subtle negative behaviors said they’d rub the animal’s belly anyway.

When Education Backfires

Researchers created a 2.5-minute training video to teach people about cat play signals and body language. Half the participants watched it; the other half saw a control video about general cat care. The training was supposed to help.

The video did produce small improvements for recognizing obvious behaviors. People who watched it showed slightly better accuracy for clear positive signals (a 1.9 percentage point increase) and clear negative signals (1.4 percentage points). So far, so good.

But here’s the problem. The training video made people worse at recognizing subtle negative behaviors, the very signals that matter most for preventing injuries and cat stress. Accuracy dropped by 18.8 percentage points for these early warning signs. People who didn’t watch the training video were actually better at spotting when a cat was beginning to feel uncomfortable.

Several factors might explain this. The brief video may have emphasized dramatic behaviors at the expense of subtle ones, essentially teaching people to wait for a crisis instead of catching problems early. People likely remembered the striking images of hissing and spitting cats while glossing over quieter signals of tension. The video also didn’t directly contrast similar-looking behaviors that mean different things. For example, a playful paw swipe versus a warning swat.

Perhaps most troubling, the training video increased participant confidence. People who watched it rated themselves as better at reading cat body language, even though their actual performance on subtle cues had gotten worse. This combination of increased confidence and decreased accuracy creates a dangerous situation where people feel more certain they understand their cat while actually understanding less.

Experience Doesn’t Help As Much As You’d Think

Having a cat at home didn’t guarantee better recognition skills. General cat ownership was associated with only an 11 percentage point improvement in identifying overall negative behaviors, and didn’t help much with subtle cues at all.

Professional experience made more of a difference. Veterinarians, vet techs, shelter workers, and others with vocational cat experience showed an 8.1 percentage point improvement in recognizing subtle negative behaviors, the category where most people struggled. But even with professional training, the improvement was modest.

The Disconnect Between Recognition and Response

Participants were asked to identify what emotion each cat was experiencing. For obvious negative behaviors, most correctly selected “frustrated/annoyed” (40.2%). For obvious positive behaviors, “playful” dominated at 79.8%.

Subtle behaviors generated confused responses. When cats displayed subtle negative signals, “playful” (20.9%) and “frustrated/annoyed” (19.8%) were chosen at nearly equal rates. People also frequently selected “relaxed” and “happy” for cats showing early distress. This confusion offers an explanation as to why unwanted interactions continue to happen. People appear to genuinely mistake mild discomfort for contentment.

For subtle positive behaviors, responses split between “relaxed” (36.4%) and “bored” (31.1%), showing difficulty distinguishing a calm, content cat from a disengaged one.

Cat owners didn’t fare much better at recognizing when to give their furry friend some space. (credit: Photo by Tran Mau Tri Tam ✪ on Unsplash)

The disconnect matters. Most cat bites happen at home from the victim’s own cat during play or petting. Play-related and petting-related aggression are the leading causes of cats attacking people. About 38% of study participants reported worrying about injuries when playing with cats. Yet many chose interactions likely to cause those injuries. Recognizing the problem isn’t enough.

What This Means for You and Your Cat

These findings reveal a troubling gap in how people interact with cats. The inability to recognize subtle warning signs means people continue interactions that may cause cat stress and possible human injury. When a cat is showing even mild signs of discomfort, that’s your cue to stop.

If your cat’s body stiffens, their ears flatten even slightly, their tail starts twitching, or they push your hand away, that’s a “no thank you.” Rolling over doesn’t always mean “rub my belly.” In fact, for most cats, it means the opposite. Playing with bare hands puts you at higher risk for bites and scratches. Use a toy instead, and watch for those subtle signals that playtime is over.

The good news? Now that you know what to look for, you can start paying closer attention. The early warning signs are there. You just need to watch for them, and more importantly, respect them when you see them.

---

Disclaimer: This article is intended for general informational purposes and should not be considered professional veterinary or behavioral advice. If you have concerns about your cat’s behavior or health, consult with a qualified veterinarian or certified animal behaviorist.

---

Paper Summary

Methodology

The study used a randomized controlled trial design with 368 adult participants recruited online in Australia between May and June 2025. Participants viewed short video clips of human-cat interactions sourced from YouTube. An expert panel of five feline behavior specialists first categorized 25 videos as showing positive or negative cat valence, and 16 videos with at least 80% expert agreement were selected for the study. These videos were further classified as “overt” (displaying immediately obvious behaviors like hissing or playful pouncing) or “subtle” (showing small-scale movements like whisker changes or gradual posture shifts).

Participants were randomly assigned to view 10 videos total: four before any intervention, then six after viewing either a 2.5-minute training video on cat play cues or a control video on general cat care. For each video, participants indicated whether they thought the cat’s overall state was positive or negative, what emotions the cat was experiencing, and how they would interact with the cat. Accuracy was analyzed using a hierarchical summary receiver operating characteristic (HSROC) framework, a novel application of this method that accounts for individual participant variability.

Results

Overall accuracy for identifying positive cat behaviors was 77.6%, while accuracy for negative behaviors was 67.4%. However, these numbers masked dramatic differences based on behavior obviousness. For overt behaviors, participants achieved 78.1% accuracy for positive signals and 76.7% for negative signals. For subtle behaviors, accuracy dropped to 67.2% for positive signals and just 48.7% for negative signals, barely better than chance.

Past vocational experience with cats was associated with an 8.1 percentage point improvement in recognizing subtle negative behaviors. Having lived with cats improved recognition of overall negative behaviors by 11.0 percentage points. Self-rated confidence showed a small positive effect of 1.7 percentage points per 10-point increase in confidence.

The training video produced mixed results. Participants in the training group showed small improvements in recognizing both overt positive (1.9 percentage points) and overt negative behaviors (1.4 percentage points). However, they performed substantially worse on subtle negative behaviors, with an 18.8 percentage point decrease in accuracy compared to baseline.

When asked how they would interact with cats, 90.8% of participants chose to walk away from at least one overtly negative video, but 13.9% chose to stroke a cat showing obvious distress. For subtle negative videos, 73.6% chose stroking and 42.1% chose belly rubbing at least once. Among those who correctly identified overt negative states, 19.4% still selected engaging interactions. For subtle negative states, 44.4% who correctly identified the cat’s distress still chose to engage.

Limitations

The study’s sample was not representative of the general population. Most participants (87.5%) identified as female, 81.7% currently lived with cats, and 38.6% had professional cat experience. This likely overestimates the general public’s ability to recognize cat cues. The study excluded children, who are at higher risk for facial injuries from cats. The training video was brief (2.5 minutes) and may have lacked sufficient emphasis on subtle cues. Videos were sourced from YouTube with varying quality and context, which may have influenced participant responses. The binary classification of “positive” or “negative” simplified the complex spectrum of cat emotional states. Participant responses about intended interactions may not reflect actual behavior. The study was conducted online, which prevented assessment of real-world interaction skills.

Funding and Disclosures

The authors declared no financial support for the research or publication. No conflicts of interest were reported.

Publication Details

Henning JSL, Nielsen T, Hazel S, Atkinson PJ. (2025). “Do you speak cat? Assessing the impact of a training video on human recognition of cat emotions and behaviours during play interactions,” published in Frontiers in Ethology, 4:1675587. doi:10.3389/fetho.2025.1675587

About StudyFinds Analysis

Called “brilliant,” “fantastic,” and “spot on” by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

Sophia Naughton

Associate Editor