During fear conditioning in which a phasic explicit cue (e.g., a light) is repeatedly associated with an aversive unconditioned stimulus (e.g., a shock), the organism develops fear to the explicit cue as well as to the environmental context in which the experiment took place. Experimental evidence suggests that cued fear and contextual fear model different aspects of anxiety. Studies in patients indicated that contextual fear may model an aspect that is especially relevant to anxiety disorders (Grillon et al.,1994, 1998a,b; 1999). However, the neural basis for the expression of contextual fear has not previously been elucidated in human imaging studies.

One important determinant of contextual fear is predictability: contextual fear increases when a treat (e.g., electric shock) is unpredictable, as opposed to when the threat is predictable. The aim of this study is to compare the neural substrates underlying fear evoked by predictable versus unpredictable shocks. Animal studies have indicated that conditioned responses to predictably cued threat and to less explicit threat are separate processes mediated by distinct brain structures. Psychophysiological data suggest that the proposed procedure can differentiate between these two responses. Hence, we anticipate that this procedure will allow us to compare brain correlates of these responses in humans.

Another objective is to study effects of threat of shock on processing of threat cues in the amygdala and the visual system. This will be investigated by means of event-related magneto-encephalography (MEG), electro-encephalography, and fMRI measurements.