Newswise — As director of the McGovern Institute (web.mit.edu/mcgovern), Robert Desimone divides his time between his administrative responsibilities and his own research on the control of attention. His career has reflected the evolution of the field, as basic research has begun to provide new insights into disease mechanisms.
Robert Desimone, Ph.D., director of the McGovern Institute and first incumbent of MIT's Doris and Don Berkey Professorship in neuroscience, entered Macalester College in 1970 determined to become a psychotherapist. He majored in psychology and worked in a halfway house for the mentally ill, dealing with nighttime emergencies and suicide attempts. After graduation, he spent a year as a mental health counselor.
"These experiences convinced me that therapy was not my strong suit," he reflects. "I wanted to help people directly, but my talent was in basic research."
With redirected goals, he earned a Ph.D. at Princeton University under the guidance of neuroscientist Charles Gross. He then accepted a position at the National Institute of Mental Health (NIMH), where he ran his own research lab. He rose through the ranks to become director of Intramural Research, overseeing both basic and clinical research. He revitalized the NIMH's clinical research agenda, creating programs on the genetics of psychosis and on mood and anxiety disorders. He also strengthened the translational programs that connect the research lab to the clinic. The emphasis on translational research continues at the McGovern Institute, which under Desimone's leadership has established a major new center for psychiatric disease research.
Desimone's own research focuses on attention, which researchers now understand is critical in education and learning, and is disturbed in many psychiatric conditions. "Attention is a cognitive function that is impaired in depression, schizophrenia, attention deficit disorder (ADD) and other forms of brain disorders," he says.
"When I first entered the field, basic neuroscience research appeared quite removed from patients suffering from mental illnesses," Desimone continues. "But to my satisfaction, neuroscience has grown much more relevant to psychiatry. In my administrative capacity, at NIH and now at the McGovern Institute, I have been involved in translational research with clinical applications. My own research has also come full circle and I'm now seeing potential applications to mental illness that I could never have imagined when I started my career."
The challenge of attention
Our brains are constantly flooded with sensory information, most of which is usually irrelevant—the feel of our shoes, traffic sounds, patterns on the wall, and so on. The brain's attentional control system allows us to filter out these irrelevant distractions and focus on the task at hand.
The ability to ignore distractions and stay on task is an essential skill that most of us take for granted. But some people find this very challenging, and attentional difficulties can be profoundly disruptive for many aspects of everyday life.
Grandmother cells
Desimone came to the study of attention indirectly, through his interest in visual perception. His early research focused on the first stages of visual processing, as information enters the brain from the retina. He and his colleagues set out to map the connections between different parts of the visual cortex, and to understand how visual information is transformed as it passes from one area to the next.
Neurons in the early visual areas respond to edges, corners and other simple visual features, but as researchers looked deeper into the brain, they found neurons with preferences for more complex patterns. In the early 1980s, while recording neural activity in the relatively uncharted temporal lobe of the monkey, Desimone and Gross made the remarkable discovery that some neurons respond exclusively to faces.
The response was overwhelmingly skeptical, he recalls. "Even though we knew that damage to the temporal lobe could impair face recognition in people, a condition called prosopagnosia, scientists didn't believe face cells were possible. They said we must be misinterpreting. One NIH reviewer told me, 'Even if it is true, you should drop it because no one will believe you.'"
Indeed, neuroscientists had mocked the notion that cells could have such a precise sensitivity by invoking the so-called 'grandmother cell'—a hypothetical neuron that could respond only to one's grandmother. The grandmother cell was intended as a straw man, but today it is clear that exquisitely specialized cells do in fact exist within the brain—indeed a recent study reported two different neurons that responded selectively to the actresses Jennifer Aniston or Halle Berry. Modern neuroimaging techniques, not yet developed when Desimone made his discovery, have confirmed that, far from being an isolated curiosity, this specialization is a general organizational principle within the brain. For example, Desimone's McGovern colleague Nancy Kanwisher has identified distinct brain regions that respond to faces and to body parts.
The wheel of attention
Desimone might have continued to study object recognition, but a serendipitous discovery led him to shift his research to the problem of attention. While recording neural activity in the monkey visual system, he found that some neurons behaved inconsistently, sometimes responding strongly to a visual stimulus and other times showing little response. Could the difference depend on whether the monkeys were paying attention to visual stimulus?
As with most vision studies at the time, Desimone's monkeys only needed to look passively at the screen before their eyes. No one could tell when a monkey was paying attention to the objects on the screen, or whether its mind was wandering. "It was as if a steering wheel was spinning out of control," recalls Desimone. "If I could get control of that wheel, I thought I could get more consistent results."
To gain that control, he trained monkeys to perform tasks requiring them to attend to a particular stimulus—for instance, tracking a red dot while ignoring a green one nearby. He realized that attention was something worth studying in its own right. "We saw a dramatic difference. If the monkey was attending to the stimulus, there was a big response. If the monkey was not attending, the neuron had no response, as if the unattended information was deleted. We could actually show where unattended things were taken out of the visual pathway."
Biased competition
What guides the hand on the steering wheel? When Desimone first entered the field, most scientists thought that attention was related to the control of eye movements— that an attentional 'spotlight' illuminates a point in the visual field and direct the eyes to move there. But he realized that attention is not necessarily directed to any particular location; it can also be driven by higher level cognitive processes, such as memories, instructions or internal plans, that are independent of a stimulus's location. For example, attention allows us to search for a specific person in a crowd even if we don't know where to look. In his most cited paper, a 1995 review article, Desimone and his coauthor John Duncan proposed a new view of attention which they termed 'biased competition'. The brain cannot process all the sensory signals that it receives, so these signals must compete for neural resources. A winner emerges either because of its inherent salience (the brightest object, the loudest noise) or because it is selected through a top-down biasing influence (the person we were looking for). Desimone argued that a likely origin of these biasing signals was the prefrontal cortex, the seat of higher cognitive functions and of working memory.
Again, after initial skepticism, this view of attentional bias is now generally accepted. Desimone and others are now mapping out the underlying brain pathways to understand how attention normally functions and how it malfunctions in brain disorders.
All together now
In recent years Desimone's research has moved from studying individual cells to surveying the synchronous activity of many neurons distributed across multiple brain structures. When a monkey pays attention to a stimulus, the neurons responding to that stimulus do not merely become more active, they also coordinate their activity.
By firing in synchrony with each other, they gain a competitive edge, increasing the likelihood that other brain regions will detect their message.
"It's like being in a crowd of people who are all talking at once," explains Desimone. "An individual voice gets lost, and if people simply raise their voices it just increases the noise. But if a group starts chanting in unison, their voices rise above the background and get noticed."
Desimone is now studying the same phenomenon in humans, using a non-invasive brain imaging method called magnetoencephalography (MEG). This technology uses sensitive detectors outside the head to monitor tiny magnetic fluctuations that originate from underlying brain activity. When large populations of neurons synchronize their activity, they produce high-frequency oscillations in the magnetic field that can reveal large-scale patterns of activity thought to underlie attention and other cognitive processes. Desimone's MEG work is currently done in collaboration with colleagues at NIMH, but he hopes in the future to acquire a state-of-the-art MEG scanner at MIT.
Applications to human disorders
"We want to know whether these basic mechanisms that we are learning about in monkeys also apply to humans, and we also want to look for disease effects," says Desimone. "There's growing evidence that neural synchrony may be impaired in disease conditions like schizophrenia, epilepsy, autism and Alzheimer's and Parkinson's disease. If this proves to be true, we could use MEG to help identify the dysfunctional circuits, and if we find them, it may be possible to restore synchrony therapeutically."
This intervention might come in the form of new drugs, or through direct manipulation using new optical and genetic technologies that Desimone's colleagues Ed Boyden and Xue Han have helped develop. Desimone is exploring that possibility with Han and Boyden, an associate member of the McGovern Institute.
Desimone is enthusiastic that neuroscience can now explicitly study issues related to psychology and human behavior. "When I entered the field, the opportunities for neuroscience research to impact mental illness were far behind where we are today. The trajectory of my own research parallels that of the entire field. It is extremely satisfying that 38 years into my career, I am right in the middle of what I originally set out to do."
RSS