Where Fear Lives

Emotions have a powerful effect on the mind, and fear-considered a strong survival instinct-is particularly potent. Now, with the help of modern technology and functional Magnetic Resonance Imaging, scientists can observe what is occurring in the brain as humans experience and learn fear. Elizabeth Phelps, Ph.D., director of the Phelps Lab at New York University's psychology department, discusses her research on the cognitive neuroscience of learning and memory in the amygdala and how it is changed by emotion.

Susan Fiske [SF]: What role does the amygdala play?

Elizabeth Phelps [EP]: It has been called the seat of emotion. Most of what we know, we learned from studies on fear conditioning. Patients with amygdala damage report experiencing fear and anxiety; it's not that they don't think they're feeling this, it's that they don't show an amygdala response to learned fear.

SF: How did you recognize this discrepancy?

EP: By studying startle reflex, which is a nonverbal physiological response to a stimulus that is modulated by an emotional state. In patients with amygdala damage, there were deficits in their startle response to emotional stimuli. At the same time, they responded normally when asked to rate how emotional a stimulus was.

Other research has examined the differentiation between what people say about various racial groups and implicit measures of their attitudes.

SF: Race is a good example because people are often unaware of their prejudices. Will neuroscientists be measuring things in the brain that people don't want measured?

EP: Sure, but you can measure these things behaviorally, as well. If I don't see something behaviorally, it's hard to understand the corresponding neural response. In one of my race studies, we monitored the amygdalas of white Americans while they were shown pictures of both black and white faces. Then we paired some of the faces with a loud noise and measured the subjects' startle response. Those who showed a greater startle response also showed more amygdala activity.

SF: Does this suggest that you can learn to associate a black face with danger; it's not a wired-in response?

EP: People have this idea that everything in the brain is hardwired. The truth is that our brains change every time we learn. Much of what drives these responses is familiarity. Simply showing subjects something they should be afraid of also triggers an amygdala response, so the fear doesn't have to be experienced; it can be imagined.

SF: The link between cognitive expectations and emotion is profound.

EP: Yes, because the amygdala can get information about the emotional significance of an event prior to our awareness of it. If you subliminally present faces expressing fear to subjects, the subjects aren't actually aware of seeing the faces, but the amygdala responds. This makes some sense evolutionarily, because if something is dangerous, you want to have an early-warning system.

SF: The amygdala acts before the slower conscious system?

EP: It can make you more likely to be aware of something emotional. For instance, when we showed subjects a series of words in rapid succession and asked them to remember the two words that appeared in green, we found that people were good at ignoring everything except those two words-unless the second green word occurred soon after the first one. There's an attentional refractory period. However, if that second green word elicited an emotional response, there wasn't as much of a deficit in reporting. It's almost as if the emotional aspect of the word pushed it through to awareness.

SF: It overrides the attentional deficit.

EP: Right. But if you have damage to the amygdala you don't have the ability to override that refractory period. Basically, it's easier to be aware of something that's emotional in a situation where your attention is limited, and that seems to depend on the amygdala.

SF: Does amygdala damage impair day-to-day functioning?

EP: No, because we have other ways of representing things. Patients who are shown a blue square paired with the receiving of a shock can later associate the two. They might not show a fast amygdala response, but they can still function in everyday life.

SF: What about social responses?

EP: Patients do have odd social interactions at times. It may be that they aren't picking up on subtle social cues. Social interactions are the most variable of human behaviors, so perhaps this is where the amygdala is more critical in life. Brain imaging shows amygdala activation during different types of tasks, but the question remaining is how that translates into behavior.