96 Chapter 4 Figure 4.3 Average levels of gaze toward the eye region of targets (i.e., parent, unfamiliar peer, unfamiliar adult, self) in either gaze direction (i.e., direct and averted gaze) in HC (n = 42) and DEP adolescents (n = 11). Gaze was operationalized as the percentage of dwell time within the eye region of targets relative to the total video duration per video. The right and left eye AOIs were combined into a single AOI of the eye region (A). HC adolescents gazed more toward the eye region of targets during direct versus averted gaze videos (B=3.09, SE=1.22, t(611)=2.54, p=.011, d=0.20). Gaze responses of HC adolescents were independent of the identity of the targets in the videos. DEP and HC adolescents did not differ in their eye gaze patterns, indicating that they made equivalent eye contact with the targets (B). Neuroimaging findings For all ROIs we extracted BOLD-responses to direct and averted gaze of the targets. We examined the effects of gaze direction, target, and their interaction, while controlling for multiple comparisons (p <.05/5) in five separate generalized linear mixed regression model analyses. Our analyses in HC adolescents revealed a significant main effect of target in left TPJ (χ2(3) = 36.99, p <.001), right IFG (χ2(3) = 43.48, p <.001), and right FG (χ2(3) = 121.54, p <.001). Posthoc (Bonferroni corrected) pairwise comparisons revealed decreased deactivation in BOLDresponse in left TPJ (parent: p <.001, d = 0.56; unfamiliar peer: p <.001, d = 0.60; unfamiliar adult: p <.001, d = 0.73) and decreased activation in BOLD-response in right IFG (parent: p <.001, d = 0.68; unfamiliar peer: p <.001, d = 0.76; unfamiliar adult: p <.001, d = 0.65), and in right FG (parent: p <.001, d = 1.11; unfamiliar peer: p <.001, d = 1.20; unfamiliar adult: p <.001, d = 1.20) in response to ‘others’ versus the self (Figure 4.4, left panels). There were significant main effects of target (χ2(3) = 8.40, p = .038) and gaze direction (χ2(3) = 6.22, p = .013) in dmPFC, but these effects did not survive correction for multiple comparisons (p <.05/5). Analyses in other ROIs did not reveal additional main effects of gaze direction, nor there were significant interactions between gaze direction and target on adolescents’ neural responses in the eye contact task. To test whether task-related BOLD-activation in brain regions outside the ROIs was found in HC adolescents in response to direct and averted gaze of all targets, we performed a complementary whole-brain analysis. These analyses revealed a main effect of gaze direction in left superior
RkJQdWJsaXNoZXIy MjY0ODMw