To help with processing faces, the STS relies on two distinguished neural patterns that were uncovered by researchers at Ohio State University. One pattern picks up on details such as brow movement, and the other looks for movement of the lips. Using the functional MRI signals of their test subjects, the researchers were able to develop a machine-based algorithm determining what their subjects were looking at based entirely on key muscle movements in their faces.
“Humans use a very large number of facial expressions to convey emotion, other nonverbal communication signals and language,” says Aleix Martinez, a cognitive scientist at Ohio State who is also a professor of computer engineering. “Yet, when we see someone make a face, we recognize it instantly, seemingly without conscious awareness. In computational terms, a facial expression can encode information, and we’ve long wondered how the brain is able to decode this information so efficiently.”
Faces can be deceptively complex — we may not appreciate it at the time, but we’re being offered a great deal of information each time we look at one. Not only are they an indicator of that person’s mood, but also of what they are looking at. Other faces also clue us in on how to respond. If a person is fearful — we may look for signs of imminent danger and retreat. It’s also the reason that we tend to laugh more (and louder) in group settings. We may even look to each other to determine if laughter is acceptable at a given moment.
So far, Martinez and his team of researchers have been able to build a machine that can predict emotional human responses with an accuracy rate of 60 percent, since we all tend to use similar expressions for experiencing extreme feelings of joy or sadness.
During the functional MRI session, 10 college students served as test subjects. They were given a series of faces with mixed emotion — a total of 1,000 separate images: “happily surprised,” “angrily surprised,” and “fearfully surprised” — emotions that, for example, may consist of raised eyebrows, though other muscles throughout the face may differ. Regardless of their reactions, the nine subjects all showed increased brain activity in the STS during the exercise. An algorithm that effectively models human facial recognition of emotion may soon be on the way.
The study consisted of participants with typical neural functioning, but the study’s co-author, Julie Colomb, director of the Vision and Cognitive Neuroscience Lab at Ohio State, believes that their data may be helpful when it comes to treating other disorders, particularly spectrum disorders in which patients have difficulty with learning and processing emotions. “This work could have a variety of applications, helping us not only understand how the brain processes facial expressions, but ultimately how this process may differ in people with autism, for example,” she said.
What remains to be seen is whether the STS is made up of a series of fine-grained surfaces that carry out its functions, or if the length of this so-called social river links all of its commands in one linear sequence. The upper bank of the STS is more sensitive to voices than other sounds, but this trait is often deficient in people with spectrum disorders. Specialized patches of cortex within the STS are what allow us to see where another person is looking, to share in their gaze.
Other less pronounced areas of the STS recognize physical gestures and overlap with areas that process words — perhaps a reason why we can quickly adapt to sign language. The posterior STS, a cluster at the back of the groove, tends to be less developed in people with spectrum disorder.
“There are several socially relevant functions that we know are occurring there that appear to be different in autism,” said Kami Koldewyn, a lecturer in psychology at Bangor University of Gwynedd, Wales. As small as the STS structure may be, unstable connections here can offset connectivity in the rest of the brain. The path of brain signals quickly becomes like a tangle of stored-away Christmas lights with signals overlapping in the wrong places as time goes by. Fortunately, it’s also not impossible to “untangle,” and neuroscience is already looking at new solutions.
The hormone oxytocin has shown some effectiveness in treating certain cases of autism — a chemical released by the posterior pituitary. It’s the same chemical released into the bloodstream during social bonding, and increases activity in the STS. Neurofeedback training may be yet another option — although it remains controversial — in which patients use video images to take control of their own brain waves.
Now, researchers are looking at using transcranial magnetic stimulation to target the STS region, in which magnetic coils are placed near the head and produce small electrical currents to stimulate the brain. Already, the coils have shown some success in helping people who report neuropathic pain from spinal injuries and multiple sclerosis, and may offer some promise for the future.
This article was first published in Brain World Magazine’s Spring 2018 issue.
More From Brain World
- A Look at the Write Brain
- The Heart and the Brain: Our Heartbeat at Its Best Performance
- I Can’t Decide! Why An Increase in Choices Decreases Our Happiness
- Know Your Brain: The Motor Cortex — What Moves You
- The Neuroscience Behind Rationalizing Our Mistakes
- The Transposed Mind: How Self-Aware Are We?
