Heterogeneity in threat extinction learning: substantive and methodological considerations for identifying individual...

Pavlovian threat (fear) conditioning (PTC) is an experimental paradigm that couples innate aversive stimuli with neutral cues to elicit learned defensive behavior in response to the neutral cue. PTC is commonly used as a translational model to study neurobiological and behavioral aspects of fear and anxiety disorders including Posttraumatic Stress Disorder (PTSD). Though PTSD is a complex multi-faceted construct that cannot be fully captured in animals PTC is a conceptually valid model for studying the development and maintenance of learned threat responses. Thus, it can inform the understanding of PTSD symptomatology. However, there are significant individual differences in posttraumatic stress that are not as of yet accounted for in studies of PTC. Individuals exposed to danger have been shown to follow distinct patterns: some adapt rapidly and completely (resilience) others adapt slowly (recovery) and others failure to adapt (chronic stress response). Identifying similar behavioral outcomes in PTC increases the translatability of this model. In this report we present a flexible methodology for identifying individual differences in PTC by modeling latent subpopulations or classes characterized by defensive behavior during training. We provide evidence from a reanalysis of previously examined PTC learning and extinction data in rats to demonstrate the effectiveness of this methodology in identifying outcomes analogous to those observed in humans exposed to threat. By utilizing Latent Class Growth Analysis (LCGA) to test for heterogeneity in freezing behavior during threat conditioning and extinction learning in adult male outbred rats (n = 58) three outcomes were identified: rapid extinction (57.3%), slow extinction (32.3%), and failure to extinguish (10.3%) indicating that heterogeneity analogous to that in naturalistic human studies is present in experimental animal studies strengthening their translatability in understanding stress responses in humans.