Assess whether there is a change in stimulus evoked activity after detecting an error when faces are angry versus happy.
Previous work in stimulus discrimination research has found that error detection reduces stimulus-evoked activity for the next trial dependent of task conditions (Buzzell et al., 2017). We were interested in testing this finding in the context of face encoding with varying emotional expressions (i.e., anger versus happiness), while keeping our future goals in mind of studying how this mechanism may differ across individuals with social anxiety. We quantify stimulus-evoked activity through EEG signals with source localization from MRI.
Stimuli are created using FaReT, a recent open source toolbox to create realistic computerized 3D face models (Hays et al., 2020).
Prior to beginning the study, participants fill out a series of questionnaires to assess anxiety levels (both generalized and social). Participants then perform a two-choice (counterbalanced) perceptual decision-making task, in which they are instructed to determine if the second face presented contained a higher intensity emotion than the first face. Faces presented vary in identity (also modifying gender and race to prevent adaptation and an own-race-bias) and display either a happy or angry facial expression. The first stimulus is a pedestal of either 67.4% angry or 57.7% happy, corresponding to d’=2.5 for each emotion found in a pilot study. The first 100 trials are titrated to ensure an accuracy rate of ~80% using an accelerated stochastic approximation staircase procedure (without EEG). The stimulus level value after titration is used to create the second face for half of the trials in the main task, with the pedestal shown for the other half of the trials.
Pilot Study Launched January 2023
Data Release 1: Anticipated Q3 2023
| Name | Role |
|---|---|
| Emily R. Martin | Project Administration |
| George A. Buzzell | Supervision |
| Fabian A. Soto | Supervision |
Learn more about us here.
If you are interested in contributing, please read our CONTRIBUTING.md file.