Individual Submission Summary
Share...

Direct link:

Amygdala activation and connectivity is differentially associated with antisocial behaviors versus callous-unemotional traits

Fri, March 22, 8:00 to 9:30am, Baltimore Convention Center, Floor: Level 3, Room 321

Integrative Statement

Introduction
Prominent neuroetiologic theories of antisocial behavior (AB) emphasize amygdala reactivity to threat as key to the development of AB, but in divergent ways depending on the presence of co-occurring Callous-Unemotional (CU) traits (i.e., low empathy, guilt). These models highlight several small, clinical studies showing that AB is related to greater amygdala reactivity to threat, whereas AB with co-occurring CU traits is related to diminished amygdala reactivity to threat (Blair et al., 2014; Viding, Fontaine, & McCrory, 2012). However, these results have not replicated dimensionally in community samples (e.g., Dotterer, et al., 2017). Moreover, these theories hypothesize that amygdala deficits are specific to fearful and angry faces, but little work has examined whether these deficits emerge in response to other negative emotions or to positive emotions. Finally, though amygdala-vmPFC connectivity is central to theories of AB (Blair, 2007), little work has examined whether connectivity differences in AB may extend beyond the vmPFC. Thus, the goal of this study was to examine the relationship between AB, CU traits, and amygdala reactivity and connectivity during socioemotional processing in a representative community sample with enrichment for risk for AB.

Methods
165 adolescents (age range 15–17) from the Fragile Families and Child Wellbeing Study (FFCWS) participated in an fMRI session. The FFCWS is a representative sample of children born in the early 2000s in US Cities (population >200,000) with oversampling for non-marital births, resulting in a sample with substantial levels of poverty. The current sample was primarily African-American (73.3%) and, due to exposures related to urban poverty (annual income < $25,000 for 46.7% of families), participants were at increased risk for AB. Teens engaged in a socioemotional processing fMRI task in which they indicated the gender of an emotional face (angry, fearful, sad, happy; contrasted with neutral faces). Data were processed and analyzed using SPM12 and the gPPI toolbox (McClaren et al, 2012). AB and CU traits were measured using multi-informant latent factors that included child (SRD, ICU), parent (CBCL, ICU), and clinical report (K-SADS, CAPE).

Results
AB, but not CU traits, was related to greater amygdala reactivity to fearful (t(164)=3.28, k=2) and sad faces (t(164)=3.58, k=31). AB was related to reduced connectivity of the amygdala to regions of the default mode network (DMN) and salience network (SN) while viewing fearful and happy faces. CU traits were related to increased connectivity of the amygdala with the DMN and SN while viewing fearful and happy faces (all results p<.05 corrected for multiple comparisons using 3dClustSim).

Discussion
The current results from a well-sampled cohort including many participants under-represented in neuroimaging research (e.g., teens of color, low-income families) link AB to individual differences in amygdala reactivity to interpersonal distress. However, amygdala reactivity was not related to CU traits, nor were results for AB specific to fearful faces (i.e., they are also related to sad faces). We will discuss these findings as they challenge current theory in the field and discuss our novel connectivity findings as they may indicate a broad deficit in integrating information in AB.

Authors