Affective and neural reactivity to criticism in individuals high and low on perceived criticism

Abstract

People who have remitted from depression are at increased risk for relapse if they rate their relatives as being critical of them on a simple self-report measure of Perceived Criticism (PC). To explore neural mechanisms associated with this we used functional magnetic resonance imaging (fMRI) to examine how people with different levels of PC responded to hearing criticism from their own mothers. To maximize variability in affective reactivity, depressed, recovered depressed, and healthy control participants (n = 33) were classified as high or low in PC based on a median split. They were then exposed to personally-relevant critical and praising comments from their mothers. Perceived Criticism levels were unrelated to depression status and to negative mood change after hearing criticism. However, compared to low PC participants, those who scored high on PC showed differential activation in a network of regions associated with emotion reactivity and regulation, including increased amygdala activity and decreased reactions in prefrontal regulatory regions when they heard criticism. This was not the case for praise. Criticism may be a risk factor for relapse because it helps to "train" pathways characteristic of depressive information processing. The Perceived Criticism measure may help identify people who are more susceptible to this vulnerability.

Figure 1. Empirically determined regions displaying condition (criticism/rest) × scan interactions, p<.005, 30 voxels contiguity.

Figure 2. Empirically determined regions within a priori specified structures displaying condition (criticism/rest)

× PC × scan interactions, p<.005, 12 voxels contiguity. Shown for (A) criticism (B) criticism in the a priori temporal period of 21–30 seconds following the onset of criticism, and (C) praise.

Figure 3. Brain regions associated with diagnosis × condition × scan (green) and PC × condition × scan (orange) interactions and their overlap (red) in (A) amygdala and (B) DLPFC.

As shown in the figure, these effects were largely divergent. Time-series showing replication of the Hooley et al. effects in recovered depressed participants from areas significant in the diagnosis × condition × scan analysis are shown for amygdala (C) and DLPFC (D).