Do you remember where you were when the first man landed on the moon? Can you recall the day your first child was born? And what did you eat for breakfast last Monday? Many people can still answer the first questions in detail, even when those events happened close to 40 years ago, but they have trouble answering the last. This phenomenon is well known to science. “The emotional context affects memory and learning,” says Dr. Rony Paz, who recently joined the Weizmann Institute’s Neurobiology Department. “Our memories are more easily recalled, and are more vivid, when they are tied to strong emotions. Unfortunately, this means memories connected to traumatic events may be especially powerful.”
Paz researches what happens in the brain when emotions and motivations meet cognitive functions. How do our emotions, whether positive or negative, reinforce our memories? How does the expectation of reward affect learning? How do emotional factors affect “rational” decisions, and vice versa: How do we use our rational thinking to control emotions? Disruptions in the delicate balance between feeling and rational thought may be involved in many psychological phenomena – from post-traumatic stress syndrome and anxiety disorders to autism and schizophrenia. Understanding how that balance is achieved might lead to a better understanding of these problems, as well as suggesting better means of treating them.
An evolutionarily primitive, almond-shaped structure deep in the brain – the amygdala – is a main neural circuit for processing emotions. Incoming sensory information, on the other hand, is processed in a number of brain areas, including the neocortex – the wrinkled gray matter forming the thin outer layer of the brain – and a small structure found on either side of the brain called the hippocampus. The neocortex and the hippocampus engage in a sort of dialogue, with signals traveling from the neocortex to the hippocampus and back. But when that information is tied to an emotion, the amygdala jumps into the conversation and affects the transfer of signals. “To understand exactly what role the amygdala plays in this discussion, we need to use unconventional methods to observe a whole array of brain activity at once,” says Paz. Using a combi
nation of neurophysiological and behavioral techniques as well as computational and statistical analyses, Paz’s studies capture the activities of both solitary neurons and large, multidimensional networks of brain cells, all at the same time.
These findings appeared in Nature Neuroscience, which described them as a “tour de force of neurophysiological behavioral research.” Paz developed his systematic approach while conducting postdoctoral research at Rutgers University, New Jersey, where he began to uncover the role of the amygdala in reinforcing memories with emotional content. By measuring electrical activity of neurons in a number of brain regions simultaneously, he discovered that the amygdala intercepts the signals, intensifying them and realigning them as they’re sent from the neocortex, so they arrive at the hippocampus strong and clear.
Existing memories can also be extinguished. Paz has investigated this phenomenon, as well. “Memory extinction isn’t forgetting,” he says, “but rather new learning that alters the original memory. Essentially, we learn to ‘silence’ the response we learned earlier.” The best model to date points to a specific part of the neocortex that modulates the activities of the amygdala by depressing the emotional response, though exactly how it does this is not clear. This subduing of one part of the brain by another takes place, for instance, when we try to make a “rational” decision: The cognitive, information-processing outer layer suppresses the “gut feelings” of the emotion center. In episodes of post-traumatic stress or panic attacks, the neocortex fails to properly suppress emotion-laden memories, and details of the traumatic event surface uncontrollably.
Those who suffer from post-traumatic stress may also have trouble maintaining a separation between one specific event and the general class of similar events. Generalization is a normal part of the learning process – we learn early on to lump things into categories, so that even if we’ve never seen a particular cup before, we still know we can drink from it. In other words, generalization allows us to apply past experience to unfamiliar situations. But we also learn to remember specific details (e.g., coffee tastes better in the red mug). People who overgeneralize may have difficulty separating a specific incident such as a traffic accident from the broad activity of driving, and they may therefore be more prone to developing a fear of driving after an accident. Paz is now focusing his research on the neurobiological bases of this sort of generalization when emotions or rewards are involved. This research may not only aid in understanding why some people seem more susceptible to post-traumatic stress than others; it may also provide valuable insight into how we manage to achieve a mental balance between specific details and sweeping generalizations. Understanding the mechanisms underlying this balance might also aid in the creation of machines that think and learn like humans.
Dr. Rony Paz’s research is supported by the Estelle Funk Foundation; the estate of Florence Cuevas, Mineola, NY; Mr. and Mrs. Gary Leff, Calabasas, CA; and Ms. Lois Rosen, Los Angeles, CA.
The Math of Neurons
Tel-Aviv born and raised, Dr. Rony Paz originally wanted to be a doctor. But after beginning studies at the Hebrew University Medical School, Paz switched to a double major in mathematics and philosophy, a decision that entailed going back and forth between the university’s two campuses on Mount Scopus and Givat Ram. Searching for graduate studies that would combine his two subjects, Paz learned of an interdisciplinary program in computational neuroscience, and went on to receive an M.Sc. and Ph.D. in the field. Paz, who had served in the army as head of a programming unit, held senior R&D positions in several high-tech companies while studying, designing and implementing machine-learning algorithms. He joined the Weizmann Institute’s Neurobiology Department as a senior scientist in 2007.
Paz is married to Netta and is father to Iddo, aged two, and Abigail, aged two months.