Humans are complicated, emotional creatures — and to date, scientists haven’t been able to define the full spectrum of their feelings. A recent study took a crack at the problem and mapped out at least 27 distinct emotions using statistical models and thousands of video clips.
Researchers from the University of California, Berkeley asked study participants to watch videos from their pool of 2,185 clips found online — gorgeous landscapes, explosions, cute animals, weddings, near-accidents, and embarrassing situations from subreddits like /r/cringe. Then they asked participants to report the emotions they experienced while watching.
“These days, there’s a much wider range of extremely emotional stimuli at our fingertips and our disposal than we’ve ever had in history because of the Internet,” said Alan Cowen, the study’s first author and a researcher at the Berkeley Social Interaction Lab. “The aim here was to explore how many different kinds of emotional experiences there are, and how to best conceptualize them.”
Cowen and his colleagues found that study participants reliably reported 27 different emotions: admiration, adoration, aesthetic appreciation, amusement, anger, anxiety, awe, awkwardness, boredom, calmness, confusion, craving, disgust, empathic pain, entrancement, excitement, fear, horror, interest, joy, nostalgia, relief, romance, sadness, satisfaction, sexual desire and surprise.
“We do see that there are multiple sets of words you could use to conceptualize these states,” Cowen said. “The states that lie between awe and interest, for example, can be pretty accurately conceptualized as ‘epiphany’ or ‘shock.’”
The list of distinct emotions, however, is far from complete. People encounter deep emotions during their lives — complex grief, shame, feelings of failure — and short videos might not be enough to induce them.
Test subjects listening to music, for instance, would rarely exhibit disgust, said Kevin LaBar, professor and head of the Cognition and Cognitive Neuroscience program at Duke University. “People may not like the song, but you certainly won’t see disgust.”
The next step is to study the emotions individually and how they’re produced by the brain.
“It could be useful in understanding disorders in social interaction and dysfunction, like autism or autism spectrum disorders,” Cowen said.
It could also improve our transition into an increasingly automated world, where companies are racing to bring robot maids to our homes and humanoid nurses to hospitals.
“It’s important if you want to build machines that are interacting with people and trying to understand how they’re feeling, which is a huge aim for big tech companies who are trying to build emotion recognition algorithms,” Cowen said. “There’s a lot of practical reasons for why it’s very important to understand how many different emotions there are to study.”
At the very least, we now know a portion of the feelings hidden away in our minds.
Kelly Kasulis is a freelance journalist. Follow her on Twitter @KasulisK.