Alterations in fear extinction neural circuitry and fear-related behavior linked to trauma exposure in children

Abstract

Exposure to childhood trauma is extremely common (>60 %) and is a leading risk factor for fear-based disorders, including anxiety and posttraumatic stress disorder. These disorders are characterized by deficits in fear extinction and dysfunction in underlying neural circuitry. Given the strong and pervasive link between childhood trauma and the development of psychopathology, fear extinction may be a key mechanism. The present study tests the impact of childhood trauma exposure on fear extinction and underlying neural circuitry. Children (N = 44, 45 % trauma-exposed; 6-11 yrs) completed a novel two-day virtual reality fear extinction experiment. On day one, participants underwent fear conditioning and extinction. Twenty-four hours later, participants completed a test of extinction recall during fMRI. Conditioned fear was measured throughout the experiment using skin conductance and fear-related behavior, and activation in fear-related brain regions was estimated during recall. There were no group differences in conditioned fear during fear conditioning or extinction learning. During extinction recall, however, trauma-exposed children kept more distance from both the previously extinguished and the safety cue, suggesting poor differentiation between threat and safety cues. Trauma-exposed youth also failed to approach the previously extinguished cue over the course of extinction recall. The effects on fear-related behavior during extinction recall were accompanied by higher activation to the previously extinguished cue in fear-relevant brain regions, including the dorsal anterior cingulate cortex and anterior insula, in trauma-exposed relative to control children. Alterations in fear-related brain regions and fear-related behavior may be a core mechanism through which childhood trauma confers heightened vulnerability to psychopathology.

Keywords: Childhood adversity; Childhood trauma; Extinction recall; Fear conditioning; fMRI.

Figure 1.. Experimental overview (a) and outcome measures

(b). (a) Days 1 and 2 were conducted approximately 24 hours apart. Day 1 consisted of fear conditioning, during which one conditioned stimulus (CS+) was paired with a white noise burst, which served as the unconditioned stimulus (US). A second conditioned stimulus (CS−) served as the safety cue and was not paired with the US. During the subsequent extinction learning phase, the CS+ was extinguished (i.e., presented in the absence of the US) and is thus referred to as the extinguished cue (CS+E). Day 2 was conducted during fMRI scanning to examine activation of fear-relevant brain regions during extinction recall. We focused on activation of fear-relevant brain regions based on our prior research showing poor extinction recall, as measured by expression of conditioned fear to the CS+E, coupled with activation of fear-relevant brain regions in pre-adolescent children. (b) Conditioned fear responses to the CS+ or CS+E and CS− were measured throughout each phase using skin conductance responses (SCR) and distance kept from the CS (virtual meters). Of note, for SCR, the third palm electrode was used as a ground during fMRI scanning only. Fear and US expectancy ratings were also collected during each phase for use in exploratory analyses (see Supplemental Material). We also measured activation in fear-relevant brain regions during extinction recall. In particular, we focused on the dorsal anterior cingulate cortex (dACC) and anterior insula (AI) given that these regions consistently showed activation during extinction recall in a meta-analysis of fMRI studies in adults [38] and that adults with PTSD show a consistent pattern of overactivation in these regions during extinction recall and other phases [13].

Figure 2.. Overall patterns of conditioned fear across initial learning phases (fear conditioning, extinction learning), measured by SCR (a) or distance from the CS (b), by group (control, trauma).

(a) The overall patterns (i.e., independent of trauma exposure) across phases have been previously described [31,32]. Abbreviations: SCR, skin conductance responses; CS, conditioned stimulus; CS+, conditioned stimulus paired with the US and subsequently underwent extinction; US, unconditioned stimulus; CS−, safety cue that was not paired with the US. *p < 0.05, paired sample t-test.

Figure 3.. Effects of childhood trauma on fear-related behavior during extinction recall.

(a) Children in the control group kept more distance from the previously extinguished cue (CS+E) relative to the safety cue (CS−) during a test of extinction recall whereas trauma-exposed youth kept distance from both cues. (b) Children in the control group approached the CS+E over the course of extinction recall (i.e., early to late recall), whereas trauma-exposed children failed to show this approach behavior to the CS+E. *p < 0.05

Figure 4.. Effects of childhood trauma exposure on activation of fear-relevant brain regions during extinction recall.

Children who have experienced trauma, shown in green, show greater activation in the left AI and the dACC to the previously extinguished cue (CS+E) relative to children who have not experienced trauma (i.e., controls), shown in grey. Results significant using small-volume family-wise error correction (pFWE < 0.05) using regions defined in a meta-analysis of fMRI studies on extinction learning and recall adults [38]. Abbreviations: CS+E, previously extinguished conditioned stimulus; dACC, dorsal anterior cingulate cortex; L AI, left anterior insula; BOLD, blood-oxygen-level-dependent.