Humans are terrible at understanding aggressive characteristics in dogs — and unfortunately, in other humans.
Being able to correctly interpret social interactions is an important skill that allows humans to react appropriately in different situations.
But, research from the Dog Studies Research group at the Max Planck Institute for Geoanthropology in Germany shows while we’re generally clued in to distinguishing the nature of social interactions between children, dogs, and monkeys, we suck at identifying negative behaviour in dogs and humans.
Researchers presented 96 adults with short video clips of aggressive, neutral, and playful interactions between two individuals in three different species: human children, dogs, and macaques.
The clips included clues, such as body postures and facial expressions, but ended before the interaction took place.
Half the study participants were asked to categorise the interaction as aggressive, neutral, or playful, while the other half were asked to predict the outcome from three sentences prepared by the experimenters describing the three possible outcomes.
The researchers found that people performed better than chance for both tasks, even without prior experiences with the non-human species.
They selected the correct choice of the three outcomes in 50–80% of interactions and were most accurate in categorising playful interactions, which they correctly identified 70% of the time.
This wasn’t the case for predicting aggressive outcomes in humans and in pooches in particular. Participants rated aggressive contexts among dogs at chance level, and predicted outcomes below chance level.
“It is possible that we are biased to assume good intentions from other humans and from ‘man’s best friend’,” says first author Dr Theresa Epperlein the DogStudies Research group in the Max Planck Institute for Geoanthropology, Germany.
“Perhaps this bias prevents us from recognising aggressive situations in these species.”
“Our results underscore the fact that social interactions can often be ambiguous and suggest that accurately predicting outcomes may be more advantageous than categorising emotional contexts,” adds senior author Dr Juliane Bräuer, DogStudies Research Group Leader.
The research has been published in the journal PLOS ONE.