JUL 29, 2019 9:29 AM PDT

Novel Neural-Network Can Determine What Emotion An Image Evokes

WRITTEN BY: Nouran Amin

Ever wondered if a computer may someday differentiate between joyful images from a depressing one? Can it someday tell if a film is a romantic comedy or a horror film?

Well, the answer to that is yes—at least according to the newest interdisciplinary neuroscience and computer systems studies by the University of Colorado Boulder researchers.

"Machine learning technology is getting really good at recognizing the content of images -- of deciphering what kind of object it is," stated senior author Tor Wager, who worked on the study while a professor of psychology and neuroscience at CU Boulder. "We wanted to ask: Could it do the same with emotions? The answer is yes."

The study, published in the journal Science Advances, explains how a unique computer system known as EmoNet uses the human brain as an inspiration for its model. The model integrates part machine-learning innovation and part human brain-imaging study advancing the application of "neural networks" to the study of emotion. Furthermore, the study sheds light on our very own brains to how and where images are being represented.

"A lot of people assume that humans evaluate their environment in a certain way and emotions follow from specific, ancestrally older brain systems like the limbic system," said lead author Philip Kragel, a postdoctoral research associate at the Institute of Cognitive Science. "We found that the visual cortex itself also plays an important role in the processing and perception of emotion."

University of Colorado-Boulder: Researchers use fMRI brain imaging technology at CU Boulder. (Credit: Glenn Asakawa/CU Boulder)

The new neural network, EmoNet, uses 25,000 images of various depictions and categorizes them such as horror, awe and surprise. For example, images of puppies evoked amusement.

"We found a correspondence between patterns of brain activity in the occipital lobe and units in EmoNet that code for specific emotions. This means that EmoNet learned to represent emotions in a way that is biologically plausible, even though we did not explicitly train it to do so," said Kragel. "What you see and what your surroundings are can make a big difference in your emotional life."

Watch a test of EmoNet:

A simple color can even elicit an emotion with EmoNet—black is associated with anxiety and red produced craving. Later, researchers showed EmoNet brief movie clips and requested to catgerize them--romantic comedies, action films or horror movies. The system was correct ¾ of the time.

"This shows that emotions are not just add-ons that happen later in different areas of the brain," stated Wager, now a professor at Dartmouth College. "Our brains are recognizing them, categorizing them and responding to them very early on."

Source: University of Colorado-Boulder

 

About the Author
  • Nouran earned her BS and MS in Biology at IUPUI and currently shares her love of science by teaching. She enjoys writing on various topics as well including science & medicine, global health, and conservation biology. She hopes through her writing she can make science more engaging and communicable to the general public.
You May Also Like
JUN 26, 2020
Neuroscience
Artificial and Natural Neurons Communicate Using Dopamine
JUN 26, 2020
Artificial and Natural Neurons Communicate Using Dopamine
Researchers from Stanford University have shown that artificial neurons can communicate with biological neurons with dop ...
JUL 05, 2020
Technology
Computational Model Predicts Speech
JUL 05, 2020
Computational Model Predicts Speech
Our brains analyze spoken language by predicting syllables. As such, this inspired researchers at the University of Gene ...
JUL 31, 2020
Chemistry & Physics
Advancing Nuclear Fusion by Taming the "Chirping" Plasma
JUL 31, 2020
Advancing Nuclear Fusion by Taming the "Chirping" Plasma
2020 turns out to a year of breakthrough for nuclear fusion. The world's largest fusion project ITER (International ...
AUG 21, 2020
Chemistry & Physics
MIT Scientists Found the Causes Behind Dull Blades
AUG 21, 2020
MIT Scientists Found the Causes Behind Dull Blades
The reason why we replace our shaving blades so often is because no matter how sharp they start as they eventually get d ...
SEP 07, 2020
Clinical & Molecular DX
Spot the Robodog Measures COVID Patients' Vitals
SEP 07, 2020
Spot the Robodog Measures COVID Patients' Vitals
Hospital staff caring for COVID-19 patients face the greatest risk of exposure to the virus. A new dog-like robot design ...
OCT 02, 2020
Clinical & Molecular DX
Detecting Dystonia in the Blink of an AI
OCT 02, 2020
Detecting Dystonia in the Blink of an AI
A team of scientists have created a diagnostic tool, powered by artificial intelligence (AI), that can pick up on the su ...
Loading Comments...