Explaining the effectiveness of fear extinction through latent-cause inference

Acquiring fear responses to predictors of aversive outcomes is crucial for survival. At the same time, it is important to be able to modify such associations when they are maladaptive, for instance in treating anxiety and trauma-related disorders. Standard extinction procedures can reduce fear temporarily, but with sufficient delay or with reminders of the aversive experience, fear often returns. The latent-cause inference framework explains the return of fear by presuming that animals learn a rich model of the environment, in which the standard extinction procedure triggers the inference of a new latent cause, preventing the extinguishing of the original aversive associations. This computational framework had previously inspired an alternative extinction paradigm -- gradual extinction -- which indeed was shown to be more effective in reducing fear. However, the original framework was not sufficient to explain the pattern of results seen in the experiments. Here, we propose a formal model to explain the effectiveness of gradual extinction, in contrast to the ineffectiveness of standard extinction and a gradual reverse control procedure. We demonstrate through quantitative simulation that our model can explain qualitative behavioral differences across different extinction procedures as seen in the empirical study. We verify the necessity of several key assumptions added to the latent-cause framework, which suggest potential general principles of animal learning and provide novel predictions for future experiments.

Destabilizing Different Strengths of Fear Memories Requires Different Degrees of Prediction Error During Retrieval

Reactivation of consolidated memories can induce a labile period, in which these reactivated memories might be susceptible to change and need reconsolidation. Prediction error (PE) has been recognized as a necessary boundary condition for memory destabilization. Moreover, memory strength is also widely accepted as an essential boundary condition to destabilize fear memory. This study investigated whether different strengths of conditioned fear memories require different degrees of PE during memory reactivation in order for the memories to become destabilized. Here, we assessed the fear-potentiated startle and skin conductance response, using the post-retrieval extinction procedure. A violation of expectancy (PE) was induced during retrieval to reactivate enhanced (unpredictable-shock) or ordinary (predictable-shock) fear memories that were established the day before. Results showed that a PE retrieval before extinction can prevent the return of predictable-shock fear memory but cannot prevent the return of unpredictable-shock fear memory, indicating that a single PE is insufficient to destabilize enhanced fear memory. Therefore, we further investigated whether increasing the degree of PE could destabilize enhanced fear memory using different retrieval strategies (multiple PE retrieval and unreinforced CS retrieval). We found that spontaneous recovery of enhanced fear memory was prevented in both retrieval strategies, but reinstatement was only prevented in the multiple PE retrieval group, suggesting that a larger amount of PE is needed to destabilize enhanced fear memory. The findings suggest that behavioral updating during destabilization requires PE, and the degree of PE needed to induce memory destabilization during memory retrieval depends on the strength of fear memory. The study indicates that memory reconsolidation inference can be used to destabilize stronger memories, and the findings shed lights on the treatment of posttraumatic stress disorders and anxiety disorders.

Proton receptors regulate synapse-specific reconsolidation in the amygdala

During retrieval, aversive memories become labile during a period known as the reconsolidation window. When an extinction procedure is performed within the reconsolidation window, the original aversive memory can be replaced by one that is less traumatic. Our recent studies revealed that acidosis via inhalation of carbon dioxide (CO2) during retrieval enhances memory lability. However, the effects of CO2 inhalation on the central nervous system can be extensive, and there is a lack of prior evidence suggesting that the effects of CO2 are selective to a reactivated memory. The specific effects of CO2 depend on acid-sensing ion channels (ASICs), proton receptors that are involved in synaptic transmission and plasticity in the amygdala. Our previous patch-clamping data suggests that CO2 inhalation during retrieval increases activities of neurons in the amygdala that involve in the memory trace. In addition, CO2 inhalation during retrieval increases exchanges from Ca2+-impermeable to Ca2+-permeable AMPA receptors. Thus, we hypothesize that CO2 selectively potentiates memory lability in mice when inhaled during retrieval of aversive memory. In addition, CO2 inhalation alters memory lability via synaptic plasticity at selectively targeted synapses. Alterations in spine morphology after CO2 and retrieval with a specific stimulus indicates that CO2 selectively enhances synaptic plasticity. Overall, our results suggest that inhaling CO2 during the retrieval event increases the lability of an aversive memory through a synapse-specific reconsolidation process.

A new model for recovery-from-extinction effects in Pavlovian conditioning and exposure therapy

Exposure therapy is an effective intervention for anxiety-related problems. A mechanism of this intervention has been the extinction procedure in Pavlovian conditioning, and their findings have provided many effective intervention strategies that can promote the effect of and prevent relapse following exposure sessions. However, traditional associative theories that have explained Pavlovian conditioning cannot comprehensively explain their findings. In particular, it was difficult to explain the recovery-from-extinction effects, which is the reappearance of conditioned response following extinction. In this study, we propose a new associative model that can deal with procedures that promote an effect of extinction and many recovery-from-extinction effects. The cores of this model are that the asymptotic strength of the inhibitory association depends on the degree of excitatory association retrieved in a context in which CS is presented and that the retrieval is determined by the similarity between contexts during reinforcement and non-reinforcement and the present context. Moreover, this model assumes that these similarities change under specific conditions. By adding these assumptions to the traditional framework, many difficulties in explaining these phenomena can be resolved. Our model can provide not only a new perspective in associative learning, but also many implications for exposure therapy.

Investigating the efficacy of the reminder-extinction procedure to disrupt contextual threat memories in humans using immersive Virtual Reality

Upon reactivation, consolidated memories can enter a temporary labile state and require restabilisation, known as reconsolidation. Interventions during this reconsolidation period can disrupt the reactivated memory. However, it is unclear whether different kinds of memory that depend on distinct brain regions all undergo reconsolidation. Evidence for reconsolidation originates from studies assessing amygdala-dependent memories using cue-conditioning paradigms in rodents, which were subsequently replicated in humans. Whilst studies providing evidence for reconsolidation of hippocampus-dependent memories in rodents have predominantly used context conditioning paradigms, studies in humans have used completely different paradigms such as tests for wordlists or stories. Here our objective was to bridge this paradigm gap between rodent and human studies probing reconsolidation of hippocampus-dependent memories. We modified a recently developed immersive Virtual Reality paradigm to test in humans whether contextual threat-conditioned memories can be disrupted by a reminder-extinction procedure that putatively targets reconsolidation. In contrast to our hypothesis, we found comparable recovery of contextual conditioned threat responses, and comparable retention of subjective measures of threat memory, episodic memory and exploration behaviour between the reminder-extinction and standard extinction groups. Our result provide no evidence that a reminder before extinction can prevent the return of context conditioned threat memories in humans.

Offline tDCS modulates prefrontal cortical–subcortical cerebellar fear pathways in delayed fear extinction

Transcranial direct current stimulation (tDCS) has been studied to enhance extinction-based treatments for anxiety disorders. However, the field shows conflicting results about the anxiolytic effect of tDCS and only a few studies have previously observed the extinction of consolidated memories. Off-line tDCS modulates subsequent fear response (fear recall and fear extinction) neural activity and connectivity, throughout changes in the fear pathway that is critically involved in the pathogenesis of anxiety disorders. Thirty four women participated in a two-day fear conditioning procedure. On day 1, women were randomly assigned to the control group (n=18) or the tDCS group (n=16) and went through a fear acquisition procedure. On day 2, the tDCS group received 20min tDCS at 1mA [cathode F4; anode contralateral deltoid] immediately before extinction and while inside the MRI scanner. The control group completed the extinction procedure only. fMRI whole brain contrast analysis showed stimulation dependent activity patterns with the tDCS group showing decreased neural activity during the processing of the CS+ and increased activity during the processing of the CS, in prefrontal, postcentral and paracentral regions, during late extinction. PPI analysis showed tDCS impact on the connectivity between the left dorsolateral prefrontal cortex and three clusters along the cortical amygdalo hippocampal cerebellar pathway, during the processing of the CS+ in late extinction (TFCE corrected at p < .05). The increased neuronal activity during the processing of safety cues and the stronger coupling during the processing of threat cues might well be the mechanisms by which tDCS contributes to stimuli discrimination.

Variance and Scale-Free Properties of Resting-State BOLD Signal Alter after Fear Memory Acquisition and Extinction

Previous studies showed differences in brain dynamics during rest and different tasks. We aimed to find changes of variance and scale-free properties of blood oxygenation level-dependent (BOLD) signal between resting-state sessions before and after fear learning and fear memory extinction in twenty-three healthy right-handed volunteers. During a 1-hour break between MRI-scanning, subjects passed through fear extinction procedure, followed by Pavlovian fear conditioning with weak electrical stimulation. After preprocessing, we extracted the average time course of BOLD signal from 245 regions of interest (ROI) taken from the resting-state functional atlas. The variance of the BOLD signal in and Hurst exponent (H), which reflects the scale-free dynamic, were compared in resting states before after fear learning. Six ROIs showed a significant difference in H after fear extinction, including areas from the fear and memory networks. In consistency with the previous results, H decreased during fear extinction but then increased higher than before, specifically in areas related to fear extinction network, whereas the other ROIs restored H to the initial level. The BOLD signal variance showed distinct behavior: the variance in subcortical regions increased permanently, while cortical areas demonstrated a decreasing variance during fear extinction and the reverse growth in resting state after fear extinction. A limited number of ROIs showed both changes in H and the variance. Our results suggest that the variability and scale-free properties of the BOLD signal are sensitive indicators of the residual brain activity related to the recent experience.