May 29, 2015
Kay Tye, Ph.D.

By examining the neural circuits involved in emotional processing, scientists are gaining a clearer picture of how the brain learns to associate certain environmental cues with positive or negative emotions. Such research could help guide the development of better treatments for disorders in which emotional processing is impaired, such as anxiety, addiction, and depression.  

In work published April 29th in the journal Nature, a team led by 2013 NARSAD Young Investigator grantee Kay Tye, Ph.D., at the Massachusetts Institute of Technology, focused on two groups of neurons that connect the amygdala (the brain’s emotional processing hub) to circuits involved in either fear or reward responses. In experiments with mice, the scientists found that the strength of these cells’ connections with other cells in the amygdala changes as an animal learns to either fear a certain stimulus or to associate a stimulus with a reward. They also showed that these cells are critically involved in both fear and reward learning.

Tye and her colleagues tested how fear and reward conditioning affect these two groups of cells, which intermingle in a part of the amygdala known as its basolateral complex, and extend from there to one of two other parts of the brain––either the centromedial amygdala, which is important for fear responses, or to a part of the brain’s reward system called the nucleus accumbens.

When they trained animals to fear a sound, or tone, by pairing it with a mild shock, they found that cells that connect the basolateral amygdala to the fear-processing circuits developed stronger connections to the cells from which they received input. The same experience weakened such connections in cells that linked the basolateral amygdala to reward-processing circuits. Reward conditioning, in which the animals learned to associate a tone with a sip of sugar water, had the opposite effect on both groups of cells.

The team was able to enhance fear learning in the mice by artificially activating the link between the basolateral amygdala and the centromedial amygdala. When they switched those cells off, fear learning was impaired, but reward learning was enhanced. They observed the opposite effects when they manipulated the cells linking the basolateral amygdala to the nucleus accumbens.
The team’s findings help explain how changes to neural circuits in the amygdala contribute to the learning processes that underlie both reward-seeking and fear-related behaviors.