Intersubject variability in the processing of fearful faces in healthy adults

Threatening stimuli have a privileged status in the brain, meaning they receive priority in the processing stream. This makes sense from an evolutionary perspective where quickly and accurately identifying threat is necessary for survival. However, how much priority the brain gives threatening stimuli has been shown to vary among individuals, even healthy adults, in the laboratory setting. This has been shown repeatedly with threatening face stimuli, but researchers to date have used limited approaches to study intersubject variability. Some studies have correlated personality measures with brain activation, thereby ignoring behavioral variability. Other studies have looked at behavior but have only looked at behavior in pre-selected groups of subjects (namely high and low anxiety).

This thesis investigates intersubject variability in fearful face processing by using variability in behavior to guide analyses and find links to personality traits and neural activation.

In Study I, we explored behavioral variability on a novel fearful face categorization task where faces were presented at varying expectation levels. Faster reaction times to fearful faces compared to neutral faces correlated with higher levels of cautiousness, as measured by a harm avoidance personality assessment. Faster reaction times to fearful faces also correlated with increased activation to unexpected fearful faces in a prefrontal-striatal network.

In Study II, we explored behavioral variability on a fearful face repetition task where fearful and neutral faces were presented repeatedly and subjects were asked to categorize the face as fearful or neutral. A decrease in reaction time for repeated fearful faces was associated with less state and trait anxiety in our healthy subjects. Additionally the lack of behavioral advantage to categorize repeated fearful faces was also associated with more striatal and early visual activation to the first fearful face. The opposite was found for repeated neutral faces, where subjects who were faster to respond to the repeated neutral face had less activation to the repeated face.

In Study III, we explored behavioral variability on a fearful face detection paradigm where subjects were asked to detect a briefly presented (33ms) fearful face that was directly followed by a neutral face mask. Fearful face detection sensitivity correlated with trait anxiety and other personality measures related to trait anxiety, but it did not correlate with state anxiety.

In conclusion, our results demonstrate large amounts of behavioral variability on three different fearful face processing tasks in healthy adults. It is not the case that all healthy individuals showed a processing advantage for fearful faces, in fact, some healthy adults actually showed a processing disadvantage for fearful faces. In Study I, neither state nor trait anxiety correlated with the behavioral advantage for fearful faces, but harm avoidance did. While state and trait anxiety directly correlated with a lack of behavioral advantage for repeated fearful faces in Study II, trait anxiety directly correlated with a behavioral advantage for detecting briefly presented fearful faces in Study III. These results underscore that personality interacts with fearful face processing differently in different contexts. Behavioral variability was also associated with varying neural activation patterns in Studies I and II. These activation patterns were distinctly different from the activation patterns evoked by neutral faces demonstrating that behavioral advantages for fearful faces are uniquely neurally encoded. These results argue that there is no “normal” when it comes to fearful face processing and future studies should avoid lumping healthy individuals together on such tasks.

List of scientific papers

I. The Interaction between expectation and sensitivity to threatening stimuli in healthy adults: An FMRI study of intersubject variability. [Manuscript]

II. Anxiety prevents fearful face repetition advantage. [Manuscript]

III. Fearful face detection sensitivity in healthy adults correlates with anxiety related traits. [Manuscript]