Brain mechanisms of anxiety's effects on cognitive control in major depressive disorder

Abstract

Background: Adults with major depressive disorder (MDD) demonstrate increased susceptibility to interfering effects of anxiety on cognitive control; although under certain conditions adults with MDD are able to compensate for these effects. The brain mechanisms that may facilitate the ability to compensate for anxiety either via the recruitment of additional cognitive resources or via the regulation of interference from anxiety remain largely unknown. To clarify these mechanisms, we examined the effects of anxiety on brain activity and amygdala-prefrontal functional connectivity in adults diagnosed with MDD.

Method: A total of 22 unmedicated adults with MDD and 18 healthy controls (HCs) performed the Tower of London task under conditions designed to induce anxiety, while undergoing a functional magnetic resonance imaging assessment.

Results: During the easy condition, the MDD group demonstrated equivalent planning accuracy, longer planning times, elevated amygdala activity and left rostrolateral prefrontal cortex (RLPFC) hyperactivity relative to HCs. Anxiety mediated observed group differences in planning times, as well as differences in amygdala activation, which subsequently mediated observed differences in RLPFC activation. During the easy condition, the MDD group also demonstrated increased negative amygdala-dorsolateral prefrontal cortex (DLPFC) connectivity which correlated with improved planning accuracy. During the hard condition, HCs demonstrated greater DLPFC activation and stronger negative amygdala-DLPFC connectivity, which was unrelated to planning accuracy.

Conclusions: Our results suggest that persons with MDD compensate for anxiety-related limbic activation during low-load cognitive tasks by recruiting additional RLPFC activation and through increased inhibitory amygdala-DLPFC communication. Targeting these neural mechanisms directly may improve cognitive functioning in MDD.

Keywords: Anxiety; cognitive control; depression; norepinephrine; stress.

Figure 1

Panel A. Bar graphs indicate group differences and changes in anxiety across task blocks; bars with differing letters are significantly different from one another. Panel B. Non-linear mediation of group differences in easy planning time by baseline anxiety.

Figure 2

Region of interest time series analyses. Panel A shows group differences in the time series of left amygdala activity for easy and hard problems. Panel B shows group differences in the time series of left rostrolateral prefrontal cortex activity (RLPFC) for easy and hard problems. Panel C shows group differences in the time series of left dorsolateral prefrontal cortex (DLPFC) activation for easy and hard problems. The squares under each time-series indicate where the groups differ at white=p<.050, black=p<.100. Black underlined segments indicate regions that are statistically significant after contiguity thresholding to control for multiple comparisons at p<.050.

Figure 3

Path analyses demonstrating the mediation of group differences in brain activity. Panel A shows mediation of group differences in amygdala activation by baseline anxiety. Panel B shows statistical mediation of group differences in rostrolateral prefrontal cortex (RLPFC) activation by amygdala activity. Panel C shows serial mediation of group differences in RLPFC activation via anxiety and amygdala activation. Solid lines reflect mediation paths.

Figure 4

Voxel-wise results: Left amygdala psychophysiological interaction (PPI) analysis. Increased connectivity, (p<.050 cluster corrected) with the left amygdala seed during the easy versus hard trials was observed in: Panel A the dorsolateral prefrontal cortex (DLPFC) extending into the rostrolateral prefrontal cortex (RLPFC) and Panel B the inferior frontal gyrus (IFG) extending into the insula. No other regions survived cluster thresholding. Panel C shows beta-weights for the contrast of amygdala-DLPFC connectivity versus baseline for each condition for each group. Panel D shows beta-weights for the contrast of amygdala-IFG connectivity versus baseline for each condition for each group. Panel E demonstrates that amygdala-DLPFC functional connectivity is more strongly associated with increased planning accuracy in the easy condition in MDD relative to HCs.