2019: GOLDBERG

Discovering how the brain coordinates perception, attention, and action

The eyes may well be the most mobile organs of the body, moving rapidly several times per second. With each movement, the visual image on the retina shifts. Despite the constantly shifting input, our visual world remains remarkably stable. How does the brain compensate for eye movements to generate a continuously accurate representation of visual space?

One possibility, independently proposed by German physiologists Wilhelm Wundt and Ewald Hering in the 19 th century, is that sensors in eye muscles produce the sense of eye position. But according to German physician and physicist Hermann von Helmholtz, the muscle sense was too inaccurate to use for calibrating vision. Instead, he proposed that the brain determines eye position using what is now called corollary discharge or efference copy – a copy of the neuronal signal driving the eye movement. This copy of the eye-movement command is sent to visual brain regions to inform them of the impending eye movement.

This intriguing topic continued to puzzle visual neuroscientists until a seminal study published in Science in 1992. That’s when Michael (Mickey) Goldberg, then a senior investigator in the Laboratory of Sensorimotor Research at the National Eye Institute, and now the David Mahoney Professor of Brain and Behavior at Columbia University Medical Center, and colleagues discovered that single neurons in lateral intraparietal area (LIP) – a subregion of the parietal lobes – use information about intended eye movement to update the representation of visual space.

At the time a rapid eye movement called a saccade is planned, the parietal representation of the visual world undergoes a shift analogous to the shift of the image on the retina. But unlike the retinal shift that follows a saccade, the parietal shift precedes the eye movement and predicts the location of visual input. In other words, these parietal cortex neurons anticipate the onset of each saccade, and this anticipation could compensate for the visual disruption produced by the saccade.

The discovery of this anticipatory activity provided the first neuronal clue as to how the computations underlying visual stability in humans could operate. “This was the first proof of Helmholtz’s postulate that the brain creates a stable representation of the world, despite a constantly moving retina, by feeding back the dynamics of an impending eye movement to update the visual information in the brain,” Goldberg said.

For his pivotal contributions to the field of visual neuroscience, Goldberg has been named the recipient of the 2019 Golden Brain Award from the Berkeley, California-based Minerva Foundation. The award, now in its 35th year, recognizes outstanding contributions in vision and brain research. Goldberg was honored in a private ceremony held remotely on June 23rd.

“Michael Goldberg is one of the world leaders in cognitive neuroscience, which results from a lifetime of innovative experiments,” said former Golden Brain Award recipient Robert Wurtz, a senior investigator (scientist emeritus) at the National Eye Institute, and Goldberg’s former postdoctoral adviser. “He brings to these experimental achievements an incomparable ability for connecting neuronal activity in the brain to the highest levels of brain function.”

Shaping the future direction of research

Despite the discovery reported by Goldberg and colleagues nearly 30 years ago in Science, it turns out that Wundt and Hering were not wrong about the sense of eye position arising from sensors in eye muscles. In a study published in Nature Neuroscience in 2007, Goldberg and colleagues shed light on where this process unfolds in the brain. Specifically, they discovered the proprioceptive representation of eye position in primary somatosensory cortex, which plays a critical role in processing touch and body-position information. These neurons are located near cells with a tactile representation of the eyebrow.

According to Goldberg, the proprioceptive eye-position signals could be critical for the long-term calibration of corollary discharge. Although corollary discharge may support immediate action and other functions that require high precision in time and space, it is not 100% accurate. That is, the copy of the neuronal signal driving the eye movement is not always identical to the actual eye movement. So both slower proprioceptive eye-position signals and the faster process of corollary discharge could work together to support critical visual and motor functions.

“Goldberg’s experiments and insights resolved the historic controversy between giants in the field: Wundt and Hering on the one hand and von Helmholtz on the other. Both sides were right,” Wurtz said. “His work has stimulated a multitude of studies of anticipatory activity throughout the brain and generated a new field in human psychophysics devoted to exploring the behavioral consequences of saccadic anticipation.”

Beyond stimulating basic science, Goldberg’s research could have clinical implications. For example, the mechanics of the eyes change with development and injury, requiring calibration that would be aided through feedback about eye position. Finding the eye-position signal in the brain might facilitate the treatment of eye disorders and traumas. Moreover, understanding how the parietal lobe processes space illuminates why patients with lesions in this part of the brain behave as they do.

As a neurologist, Goldberg has focused on research topics with clinical relevance. On top of his research and teaching responsibilities, he is a senior attending neurologist at New York Presbyterian Hospital, Columbia University Medical Center.

“It is astonishing that Dr. Goldberg has been able to combine his outstanding research output – both in terms of the quantity and the quality of his discoveries – with his significant clinical duties,” Wurtz said. “This combined experience, as an active clinician and world-leading scientist, lent Dr. Goldberg the credibility, and put him in the unique position, as SfN [Society for Neuroscience] president, to mount an initiative to convey the value of basic research as an equally worthy endeavor to clinical research, to Congress and to the public at large.”

Goldberg has received widespread recognition for his accomplishments. He is a member of the National Academy of Sciences, and a fellow of the American Academy of Arts and Sciences, and the American Association for the Advancement of Science. In 2011, he was awarded the Patricia Goldman-Rakic Award for Cognitive Neuroscience by the Brain and Behavior Research Foundation. He served as president of SfN, the world’s largest neuroscience society, from 2009 to 2010. In 2006, he received the Lewis P. Rowland Teaching Award from the Department of Neurology at Columbia University.

These honors are well deserved, according to former Golden Brain Award recipient Okihide Hikosaka, chief of the Neuronal Networks Section at the National Eye Institute. “Discoveries by Dr. Michael Goldberg and his colleagues over more than 40 years have been critical for progress in understanding the mechanisms of our vision and the mechanisms that regulate our visually controlled movements,” Hikosaka said. “His discoveries are centered on aspects of neuroscience that have shaped the future direction of research, including how the brain controls perception, attention, and action. These topics are among the most prominent issues in neuroscience, and research on many are largely based on the original studies of Dr. Goldberg.”

Turning into a scientist

Goldberg received his A.B. in Biochemical Sciences from Harvard College in 1963. He then spent one year as a graduate fellow at the Rockefeller Institute for Medical Research (now Rockefeller University), where he studied mammalian molecular biology. He received his M.D. from Harvard Medical School in 1968. “I fell in love with neuroscience during the last six weeks of the first year of medical school,” Goldberg said.

After completing a medical internship at the Peter Bent Brigham Hospital, he became a staff associate at the National Institute of Mental Health. “By a huge stroke of luck, I ended up in Bob Wurtz’s lab as his first postdoc,” Goldberg said. “Bob Wurtz was my most important mentor. Before arriving in his lab, I had never heard the word ‘saccade.’ I had never taken psychology in college, so I didn’t know what attention was, or what corollary discharge was. I certainly had no idea how to think experimentally, or how to struggle with hardware, or even to analyze data statistically. I learned all of this in Bob’s lab. Bob turned me into a scientist, and fifty years later still helps me.”

The duo made a lasting mark in the field with a collection of influential studies published in the Journal of Neurophysiology in 1972. They discovered that attention enhances the activity of many neurons in the superior colliculus – a brainstem structure involved in generating eye movements. Prior to this groundbreaking research, the properties of neurons had been studied without regard to the behavioral significance of the visual stimuli. “This was the first demonstration of the effects of attention on visual responses,” Goldberg said.

These experiments set the standard for the physiological study of cognitive problems, and demonstrated that sensory responses could be modulated by behavior. “In the years following their discoveries, many neuroscientists have studied the neuronal mechanisms of attention in many brain areas,” Hikosaka said. “This subsequent expansion of research on visual attention was one of the most important advances in neuroscience research.”

Shifting to cerebral cortex

After completing his residency at Harvard Longwood Neurology, Goldberg joined the Laboratory of Sensorimotor Research at the National Eye Institute as a founding member. As a principal investigator, he examined how the responses of neurons in parietal cortex are modulated by attention and the behavioral relevance of objects.

In a study published in the Journal of Neurophysiology in 1981, Goldberg and colleagues showed that attention enhances the activity of many visually responsive neurons in area 7 of posterior parietal cortex, even in the absence of an eye or hand movement used to respond to the visual stimulus. By contrast, additional experiments revealed that visually responsive neurons in the frontal eye fields only show enhanced activity prior to eye movements, similar to cells in the superior colliculus.

In all previous studies of area 7, eye movements and attention were intertwined, and it was not possible to determine whether the neuronal activity was related more to movement or attention. Through clever experimental design, Goldberg and his colleagues dissociated these two processes.

“They showed that neurons in posterior parietal cortex had a visual attentional signal that could not be explained as simply a motor command,” Wurtz said. “Goldberg and his collaborators were among the first investigators to investigate the neuronal basis for the clinical observations that parietal cortex lesions altered visual attention. His contributions have stood the test of time and have stimulated lines of experiments that are being pursued in many laboratories today.”

In a similar vein, Goldberg and colleagues went on to show that LIP neurons are responsive to the behavioral significance of visual stimuli, independent of the planning of saccadic eye movements. As reported in a study published in Nature in 1998, only the most salient or behaviorally relevant objects are represented in LIP, and the representation of visual salience in this brain region may subserve a wide range of behaviors including, but not limited to, saccadic eye movements.

The influence of spatial attention on neuronal activity in LIP was further elucidated in a study published in Science in 2003. The researchers found that LIP neurons tracked visual attention on a moment-by-moment basis. The results provided strong evidence that LIP contains an attentional priority map, which continuously represents the importance or salience of every location in the visual field. Spatial attention is deployed to locations in the order of their priority. Once visited, the priority at that location is canceled to prevent revisiting recently attended locations through a process called inhibition of return.

After Goldberg joined the faculty at Columbia University Medical Center, his experiments showed that the activity of neurons in LIP correlates with the ability to ignore visually salient stimuli. As reported in a study published in Nature Neuroscience in 2006, the findings suggest that activity in this area represents a top-down mechanism that allows us to ignore salient but behaviorally irrelevant distractions based on prior knowledge, intentions and goals.

“These observations led to the innovative idea that the activity in parietal cortex represents a priority map where bottom-up (stimulus) factors combine with top-down factors (such as attention) that produce a weighting of a variety of simultaneous stimuli,” Wurtz said. “These observations and ideas are now widely recognized as one of the organizing theories of posterior parietal cortex.”

Taking a reductionist chunk

In related experiments, Goldberg likewise examined how top-down neuronal signals constrain behavior, but with a focus on frontal cortex. As reported in Neuron in 2004, he and his colleagues showed that neurons in prefrontal cortex are involved in suppressing specific saccadic eye movements.

Prior to that study, the prefrontal cortex had been implicated in the suppression of unwanted behavior, but there had been little direct neurophysiological evidence for the underlying mechanism. These results revealed a frontal signal that is related to the active suppression of one action while the subject performs another.

The findings also suggest that there are separate neural mechanisms in the frontal cortex for response suppression and response planning. The parallel representations of “look” and “don’t look” in these prefrontal neurons may be important for the flexibility of context-dependent behaviors, allowing the brain not only to plan an appropriate movement, but also to simultaneously inhibit an inappropriate movement.

“Dr. Goldberg’s group has discovered the neuronal mechanisms of functions that are critical for our everyday vision,” Hikosaka said. “These mechanisms work flawlessly with eye movements that occur several times per second and that are critical for perception, attention, and action. More importantly, these discoveries raised many more important questions, especially about the interactions and integrations between brain areas. Dr. Goldberg’s group has solved some of these questions, but many questions raised by these studies remain, but are likely to be solved by neuroscience research in the near future, much of that research stimulated by Dr. Goldberg’s work.”

Despite his impressive achievements, Goldberg is not resting on his laurels. Currently, his lab is investigating how eye-position proprioception affects long-term visual memory, the role of the neurotransmitter acetylcholine in arousal and modulating activity in parietal cortex, and the function of the cerebellum in visuomotor learning.

As always, Goldberg remains driven by a deep sense of curiosity. “Understanding how the brain works, and what is the physiology that underlies human behavior, is one of the great questions in science,” Goldberg said. “I love taking a reductionist chunk of the question, and when I am lucky, adding to our understanding, discovering something that no one else has ever known.”