Eyesight is the portal through which researchers at Mississippi State University are gaining deeper insights into how the brain processes information — scientific explorations that could pave the way to next-generation computer technologies and medical breakthroughs.


 

 

 

 



MSU Cognitive Psychologist Seeks to Uncover Mysteries of Human Vision

Consider for a moment the importance of vision—sight is the primary conduit through which humanity experiences the world. Whether moving safely from one location to another or connecting with a friend in a crowd, a person’s ability to see shapes his or her life in important ways.

However, “eyesight is just the tip of the iceberg,” says Michael S. Pratte, assistant professor of cognitive psychology at Mississippi State University. “Vision really happens in the brain.” According to Pratte, more than half of the brain is devoted to processing the visual information streaming in through eyes.


 


 

Despite decades of research, scientists who study the brain don’t yet have a firm understanding of exactly how the brain analyzes and makes sense of the information it receives from the eyes. From his research lab in Magruder Hall, Pratte and his team of undergraduate and graduate students study human behavior and brain activity in hopes of advancing vision science to the point they understand how the brain accomplishes the ability to see while applying that knowledge to areas spanning from medical science to military applications.

Several experimental methods are employed to study the human visual system, including electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), in collaboration with the MSU Institute for Imaging Technologies.

“EEG electrodes measure the tiny electrical signals that occur on the scalp when the brain is active and lets us measure exactly when something happens in the brain,” says Pratte, “but measurements on the outside of the head cannot tell us precisely where in the brain the neural activity is occurring.”

That’s where fMRI comes into play—an fMRI scanner measures the small changes in blood flow that occur inside the brain.

“When you use a specific part of your brain, the body sends blood to that area,” Pratte says. “An fMRI allows us to measure this blood flow, revealing what parts of a person’s brain are being used while they perform various tasks.”

Thanks in part to fMRI research, scientists know that different parts of the brain have different jobs. However, more recent use of the technology has advanced beyond simply understanding which brain areas are active.

 


“By using modern mathematics, we can now use this measured brain activity to determine what you are seeing and even what you are thinking,” Pratte says. “Take color, for example. With the data from an fMRI scan, we can figure out what color you’re seeing from the patterns of brain activity in brain areas that process color. And even if your eyes are closed, we can identify what color you’re thinking about by measuring activity from those same areas.”


 

Pratte and collaborators have demonstrated that almost every part of the brain involved in vision is affected by “what you are attending.”

“A major goal in psychological science is to understand how the brain does this attentional filtering,” he says. “How does the filter work? Where does it happen in the brain? And what happens to all of that information that was sent to your brain but never reached awareness?

“If an engineer was asked to build a computer-based brain that had something like attention, they would probably put a filter somewhere in the middle that only lets the most important information through,” Pratte says. “But that’s not at all how the brain works—almost every part of your brain changes how it processes information in complex ways depending on what you’re attending.”

Differences between how the brain works and how computers currently work is “one of the most exciting aspects” of making new discoveries about the brain.

 



 

 



“If we can figure out how the brain does things like read a sentence or drive a car, these same processes could be built into computers,” Pratte says. “Although computers have advanced tremendously in recent years, even our best technology is nowhere near what our brain is capable of doing.”

For example, technologies like Siri and Alexa have become good at recognizing words, but they still can fail miserably in noisy rooms where someone would have no trouble following a conversation. And these computer programs certainly don’t understand those words in the way humans do, building them into concepts, stories and ideas.

Pratte also studies very specific tasks where computers currently fail but humans excel, with the hope that “computers might eventually make us better at these critical jobs.”

“There are several military applications, such as our amazing ability to break the visual camouflage of an enemy in combat, where computers can be easily fooled,” Pratte says. “Similar problems exist in the field of medicine.

“When you get a scan of a broken bone, or a sample of tissue is taken to test for a disease such as cancer, these images are sent to experts who look at them to determine whether you have a disease and what type,” Pratte says. “It takes many years to train a person to be effective at this job, and there is a lot of variability in how good any individual might be. To date, computers have been terrible at these tasks. But if we can understand how a radiologist sees that a bright spot is in fact cancer, or a pathologist knows that something about a particular sample is abnormal, then we could use computers to have more reliable diagnoses, and ones that are available across the world.

“Can we build computers that do that?” he asks. “Can we build computers that can listen and talk or identify cancer? Probably not until we have a much better understanding of how our brains are able to do all of these things so well.”

 


Want to learn more about Mississippi State?