By David F. Salisbury
Dec. 14, 2000

Have you ever wondered what’s going on in your head when you say, “Oops!”

Neuroscientists at Vanderbilt University have come up with an answer. They have shown that a set of neurons in a specific region of the brain reacts when you realize that you have made a mistake.

The finding, reported in the Dec. 14 issue of the journal Nature, was made by post-doctoral fellows Veit Stuphorn and Tracy L. Taylor — now an assistant professor at Dalhousie University — and Professor of Psychology Jeffrey D. Schall.

The researchers propose that this region is part of an “executive system” that has evolved within the brain in order to control its own activity as it makes decisions, corrects errors and overrides habitual responses. Although cognitive psychologists generally agree that such a supervisory system must exist, this is one of the first studies to reveal its workings at such a fundamental level.

“The work is very important because it shows the cellular basis of self-control,” says Sohee Park, associate professor of psychology at Vanderbilt. “It gets at really basic questions of psychology and philosophy like the origin of thought and free will.” It also has important implications for the understanding of schizophrenia, obsessive-compulsive disorder and psychopathic behavior, she adds.

Schall’s group specializes in the study of the brain’s control of eye movement. The study they report is based on an elegantly simple task: deciding whether or not to shift one’s gaze. The researchers sat macaque monkeys in front of a computer screen. An eye-tracking system monitored where they were looking. A spot appeared in the center of the screen. When the monkey’s gaze was fixed on the spot, the spot disappeared and another spot appeared on the periphery of its vision. If the monkey shifted its gaze to the new spot, it was rewarded with a drink of juice. During some of the trials, the central spot reappeared during the time the monkey was preparing to shift its gaze to the peripheral spot. In these cases, the monkey was rewarded when it cancelled the eye movement it was planning and kept its vision fixed on the central target.

As the monkeys were performing these tasks, the researchers were monitoring the activity of neurons in part of the macaque’s brain called the supplementary eye field. This structure is located in the frontal lobe of the brain and is part of the supplementary motor area that was discovered in the 1940s by neurosurgeons exploring the brains of epileptic patients. Previous research by Schall and others had shown that an area nearby, called the frontal eye field, exercised direct control over eye movements. The researchers knew the supplementary eye field also had some involvement in the control of eye movements and they were attempting to discover the role that it plays.

Schall and his colleagues found that the supplementary eye field exhibited a much different pattern of neuron activity than the frontal eye field. “It appears that the neurons in the secondary eye field are monitoring eye movement, not controlling it,” Schall summarizes. He and his colleagues report finding three distinct types of neurons in the area. One type acts when the monkey realizes that it has made the correct decision and will be rewarded. Similar “reward” or reinforcement neurons have been reported in other parts of the brain. The second type, which they have dubbed the “oops” or error neurons, reacts when the monkey realizes that it has made a mistake and will not receive a reward. The third type responds when the brain has received two conflicting instructions.

These findings shed new light on an ongoing debate over the interpretation of similar research performed with human subjects using electroencephalograms (EEG) and fMRI, a remote sensing technique that measures levels of brain activity.

Michael Coles and coworkers at the University of Illinois discovered an EEG signal that occurred when human subjects made errors. They called this the “blunder blip” and attributed it to the brain’s error-recognition response. Then Jonathan Cohen at Princeton University conducted a series of fMRI experiments that mapped brain activity when human subjects were put in situations where they are likely to make mistakes. When they realize that they have made an error, Cohen found that the supplementary motor area and an adjacent area called the anterior cingulate cortex both become active. But Cohen’s group also recorded activity in these areas when the person judged correctly. So he concluded that this activity can’t just be about errors and has proposed that it signals when the brain is coping with conflicting impulses.

“Our results suggest that both interpretations are partially right,” Schall says. Different groups of neurons are responding to both errors and conflicts.”

Gordon Logan developed the “countermanding paradigm” that provides the basis for Schall’s study. “I was interested in impulse control. To what extent are impulses automatic and how well can people control them,” says Logan, who recently joined the psychology department at Vanderbilt as a Centennial professor.

But there is a basic difficulty in studying this subject. When a person is asked to stop a behavior and they do stop, there is no behavior to measure. To get around this difficulty, Logan decided to study what happens when people are asked to start then stop doing something in rapid succession. In this situation he reasoned that two neurological processes—“go” and “stop”—must be racing with each other. Based on this conception, he developed a mathematical model that estimates the probability that a subject will stop a behavior in a given circumstance.

“I’m amazed at how successfully Schall has used this method and the quality of the data that he has gotten from it,” says Logan.

There is an interesting parallel between Schall’s findings and a study of children with Attention Deficit Hyperactivity Disorder (ADHD) that he was involved with in Toronto, Logan points out: “We found that the children with ADHD were slower to respond to stop signals than normal children. Interestingly, Ritalin, one of the drugs used to treat this condition, also improved their stopping ability.”

Park, who studies schizophrenia, describes an even more striking clinical connection. In eye-tracking experiments, she has found that 80 percent of schizophrenia patients and about half of their healthy, first-degree relatives have difficulties in the executive control of eye movements.

Park, Logan and Schall plan to collaborate with Herbert Meltzer, professor of psychiatry and pharmacology at Vanderbilt and an expert on the treatment of schizophrenia, on a series of parallel studies with monkeys and people to test the efficacy of anti-schizophrenia drugs like clozapine.

The research was supported by the National Institute of Mental Health and the National Sciences and Engineering Research Council of Canada.