Characterizing the nature of emotional-associative learning deficits in panic disorder: An fMRI study on fear conditioning, extinction training and recall
Introduction
Emotional-associative learning serves as a translational model for the development, maintenance and treatment of anxiety disorders (Mineka and Oehlberg, 2008) as it represents a key paradigm for understanding the neurobiology of fear and the mechanisms underlying variations in fear memory strength (Johnson et al., 2012). Extinction training, the laboratory analogue to behavioral exposure (Bouton et al., 2001) is currently conceptualized as forming a new memory trace which confers the inhibition of the formerly learned fear reaction when the extinction memory is recalled (Milad and Quirk, 2002). Although extinction has been extensively studied in animals, less is known about alterations in extinction training and recall in anxiety disorders. Of note, most studies used one-session fear conditioning and extinction protocols neither allowing for memory consolidation nor for assessing extinction recall – a phenomenon with high translational value for better understanding the problem of clinical relapse.
Fear conditioning enables the organism to avoid future threats in that important information (conditioned stimuli, CS) signaling a potential threat (unconditioned stimulus, US) elicits defensive reactions (conditioned fear response, CR). During fear extinction training, a second fear-inhibitory learning process is initiated, resulting - after successful consolidation - in two memory traces existing in parallel: the conditioning memory (CS-US) and the extinction memory (CS-no-US). When encountering the CS after extinction training, both memory traces can be activated, with the magnitude of fear reactions being dependent on the extent to which the extinction memory is activated (Milad et al., 2006). The neurophysiology of fear acquisition is well identified (Maren and Quirk, 2004) and highlights the role of the amygdala as central structure for the acquisition and expression of learned fear in rodents (Duvarci and Pare, 2014, LeDoux, 2000) and humans (Phelps et al., 2004, Sehlmeyer et al., 2009). Animal and human neuroimaging studies have also shown overlapping neural systems involved in extinction learning, corroborating the role of the amygdala and the infralimbic cortex (IL) in rodents or, respectively, the ventromedial prefrontal cortex (vmPFC) as its human homologue (Delgado et al., 2008, Milad et al., 2014). In comparison to the acquisition, amygdala activity decreases as extinction proceeds while activity in the IL/vmPFC relatively increases (Milad et al., 2007) throughout extinction learning as the IL/vmPFC projects to inhibitory neurons within the amygdala (Milad and Quirk, 2012). The IL/vmPFC also constitutes a key region for extinction recall (Davis, 1992, Milad and Quirk, 2002) just as the hippocampus, indicating the importance of contextual information for retaining extinction memory (Kalisch et al., 2006, Milad et al., 2007).
Fear conditioning seems to play a pivotal role for the development and maintenance of PD, although the exact nature of the underlying fear learning and extinction deficits remains under debate (Duits et al., 2015, Lueken et al., 2014). Gorman et al., 2000, Gorman et al., 1989) presented a neuroanatomical hypothesis according to which the behavioral symptoms of PD are mediated via a neural network encompassing the amygdala, thalamus, hypothalamus, hippocampus, periaqueductal grey (PAG) and locus coeruleus. Accordingly, PD patients are thought to have an abnormally sensitive fear network with lowered activation thresholds resulting in excessive activation. Threat cues can then trigger defensive behavior by activating survival circuits in the brain, probably mediated by the PAG (Hamm et al., 2014). To date, several neuroimaging studies have corroborated the pivotal role of certain neural networks for PD pathophysiology (Dresler et al., 2013, Graeff and Del-Ben, 2008, Lueken et al., 2014, Sobanski and Wagner, 2017). Recent functional studies suggest aberrant activation in an extended network comprising the brainstem, insula, anterior and midcingulate cortices as well as medial and lateral parts of the PFC in PD (Sobanski and Wagner, 2017). Additional pathophysiological models of PD encompass among others interoceptive conditioning processes (e.g., Benke et al., 2018, Bouton et al., 2001, Khalsa and Lapidus, 2016, Pappens et al., 2015) or suffocation false alarm theory (Klein, 1993) resulting in CO2 hypersensitivity (Esquivel et al., 2010, Leibold et al., 2016), pointing towards the relevance of brain systems beyond fear conditioning circuits such as the periaqueductal grey and brainstem (Goossens et al., 2014, Wemmie, 2011). Regarding fear conditioning and extinction, altered neural processing of safety cues (Kircher et al., 2013, Tuescher et al., 2011), a proclivity towards fear overgeneralization (Lissek et al., 2009) or resistance to extinction indicated by a prolonged retention of the CR (Michael et al., 2007) have been considered to be accountable for deviant fear learning processes in PD.
Previous research on fear conditioning and extinction as a pathophysiological marker of PD is however limited by two major shortcomings: First, fear acquisition and extinction are usually conducted within one session (Milad et al., 2007) thus not allowing the fear and extinction memory to consolidate. Consequently, alterations pertaining to extinction training of one-session paradigms cannot be unequivocally interpreted as truly representing fear inhibitory learning, but rather a mixture resulting from the recall of the fear memory and the extinction process. Second, the recall of extinction memories is rarely tested at all (as compared to animal studies that typically include testing on a separate day). Therefore, knowledge about deficits in extinction recall as a correlate of PD is virtually not available, limiting the translation of animal findings to the patient level. From a clinical perspective, it is however essential to test whether patients recall their extinction memories, i.e. actively inhibit fear reactions, when encountering the CS again. The phenomenon of relapse after successful behavioral exposure frequently seen in anxiety disorders (Taylor et al., 2012) could thus be interpreted as a failure to consolidate or recall extinction memories.
Following this translational perspective, the present study aimed to closer match fear conditioning and extinction protocols based on animal research to the clinical level. Investigating the neural substrates of fear conditioning, extinction training and recall separated by distinct overnight consolidation phases, we applied a three-day fear conditioning and delayed extinction paradigm. Regarding the acquisition of newly conditioned fears during day 1, we expected patients to show heightened activity in defensive networks encompassing the amygdala and insula as an indicator of exaggerated conditionability (e.g., Mineka and Oehlberg, 2008, Phelps et al., 2004). Second, when recalling the fear conditioning memories at the beginning of day 2, we expected patients to show stronger activation of these networks compared to controls. After completion of the extinction training and following overnight consolidation, we hypothesized patients will show impaired recall of the extinction memory on day 3 as represented by stronger and prolonged activation of defensive networks compared to controls, where fear inhibition should take place faster.
Section snippets
Participants
As part of the multicenter national research network “Panic-Net” (2nd funding period) a total of n = 20 quality-controlled datasets with full data from all three days were included in this analysis, consisting of n = 10 patients with PD and n = 10 healthy controls (HC). Patients were recruited from the psychotherapy outpatient center at Technische Universität Dresden; HC responded to local advertisements. Patients and controls were matched for age, gender, smoking status, handedness (only
Sample characteristics and behavioral data
Sample characteristics are given in Table 1. Behavioral data for the fear conditioning task (ratings, SCR, contingency awareness) can be found in the supplement (supplemental data & figures S1, S2, S3). Briefly, we observed a main effect of conditioning in arousal ratings in the entire group, but no differences between patients and controls.
Differential conditioning and extinction effects in the combined sample
fMRI results for the main task effects are given in Table 2. During acquisition we observed, among others, activation in the right supplementary motor area
Discussion
The present study employed a differential fear conditioning and delayed extinction paradigm on three consecutive days in patients with PD for the purpose of disentangling neural networks involved in fear acquisition, extinction and recall of fear-related memories to gain more insight into altered patterns of brain activation as a function of PD. Focusing on extinction recall may improve our understanding of how fear-inhibitory learning induced by behavioral exposure may be consolidated and
Role of funding source
This study was funded by the Federal Ministry of Education and Research (psychotherapy network “Panic-Net”, 2nd funding period, 01GV0615) as part of the BMBF Psychotherapy Research Initiative. The BMBF had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. The study was approved by the ethics committee of the Technische Universität Dresden (EK 62,022,010).
Contributors
Author Schwarzmeier has contributed to the data analysis, literature searches, and manuscript preparation. Author Kleint has contributed to the data collection and approved the final manuscript. Authors Wittchen, Ströhle and Hamm have contributed to the study design and approved the final manuscript. Author Lueken has contributed to the study design, data analysis, literature searches, manuscript preparation and approved the final manuscript.
Conflict of interest
The following authors report no conflicts of interest concerning the content of this paper: H. Schwarzmeier, N. I. Kleint, A. Hamm, U. Lueken. H.-U. Wittchen has been member of advisory boards of several pharmaceutical companies. He received travel reimbursements and research grant support from Essex Pharma, Sanofi, Pfizer, Organon, Servier, Novartis, Lundbeck, Glaxo Smith Kline. A. Ströhle received research funding from the German Federal Ministry of Education and Research, the European
Acknowledgments
We would like to thank Jens Siegert for programming advice, the staff at the Neuroimaging Center at the TU Dresden for support during data acquisition and patients and controls for participating in this study. This study was funded by the Federal Ministry of Education and Research (psychotherapy network “Panic-Net”, 2nd funding period, 01GV0615) as part of the BMBF Psychotherapy Research Initiative. Hanna Schwarzmeier is supported by the German Research Foundation (Collaborative Research Center
References (67)
- et al.
Cue and context conditioning to respiratory threat: effects of suffocation fear and implications for the etiology of panic disorder
Int. J. Psychophysiol.
(2018) - et al.
Optimized EPI for fMRI studies of the orbitofrontal cortex
Neuroimage
(2003) - et al.
Neural circuitry underlying the regulation of conditioned fear and its relation to extinction
Neuron
(2008) - et al.
Amygdala microcircuits controlling learned fear
Neuron
(2014) - et al.
Neurobiology of panic disorder: from animal models to brain neuroimaging
Neurosci. Biobehav. Rev.
(2008) - et al.
Pavlovian fear memory circuits and phenotype models of PTSD
Neuropharmacology
(2012) - et al.
Effect of cognitive-behavioral therapy on neural correlates of fear conditioning in panic disorder
Biol. Psychiatry
(2013) - et al.
(Don't) panic in the scanner! How panic patients with agoraphobia experience a functional magnetic resonance imaging session
Eur. Neuropsychopharmacol.
(2011) - et al.
The role of precuneus and left inferior frontal cortex during source memory episodic retrieval
Neuroimage
(2005) - et al.
Fear extinction in rats: implications for human brain imaging and anxiety disorders
Biol. Psychol.
(2006)
Neuroscience of fear extinction: implications for assessment and treatment of fear-based and anxiety related disorders
Behav. Res. Ther.
Recall of fear extinction in humans activates the ventromedial prefrontal cortex and hippocampus in concert
Biol. Psychiatry
The relevance of recent developments in classical conditioning to understanding the etiology and maintenance of anxiety disorders
Acta Psychol. Amst.
Transient neural activity in the medial superior frontal gyrus and precuneus time locked with attention shift between object features
Neuroimage
Extinction learning in humans: role of the amygdala and vmPFC
Neuron
Anxiety sensitivity, anxiety frequency and the prediction of fearfulness
Behav. Res. Ther.
Non-adherence and non-response in the treatment of anxiety disorders
J. Anxiety Disord.
Differential activity of subgenual cingulate and brainstem in panic disorder and PTSD
J. Anxiety Disord.
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
Neuroimage
From Pavlov to PTSD: the extinction of conditioned fear in rodents, humans, and anxiety disorders
Neurobiol. Learn. Mem.
The anatomical and functional specialization of the fusiform gyrus
Neuropsychologia
Panic and Agoraphobia Scale (PAS)
An inventory for measuring depression
Arch. Gen. Psychiatry
Decomposition of skin conductance data by means of nonnegative deconvolution
Psychophysiology
Functions of the left superior frontal gyrus in humans: a lesion study
Brain
A modern learning theory perspective on the etiology of panic disorder
Psychol. Rev.
The role of the amygdala in fear and anxiety
Annu. Rev. Neurosci.
Revise the revised? New dimensions of the neuroanatomical hypothesis of panic disorder
J. Neural Transm.
Updated meta-analysis of classical fear conditioning in the anxiety disorders
Depress. Anxiety
Facial Expressions of emotion: new findings, New Questions
Acids in the brain: a factor in panic?
J. Psychopharmacol.
Inhibitory control is slowed in patients with right superior medial frontal damage
J. Cogn. Neurosci.
Cited by (17)
Extinction of negative conditioned stimulus valence in human fear conditioning
2024, Behaviour Research and TherapyCausally mapping human threat extinction relevant circuits with depolarizing brain stimulation methods
2023, Neuroscience and Biobehavioral ReviewsCitation Excerpt :Correlative fMRI studies suggest anxiety related abnormalities in the threat extinction neurocircuitry cited above. For example, vmPFC hypoactivation during extinction learning and recall has been identified in large transdiagnostic samples and meta-analyses (Cano et al., 2021; Marin et al., 2020; but see: Schwarzmeier et al., 2019; Suarez-Jimenez et al., 2020). Moreover, consistent with the speculated anterior-posterior vmPFC functional parcellation introduced above, posterior vmPFC hyperactivation to threat stimuli has also been detected in PTSD (Suarez-Jimenez et al., 2020).
Fear conditioning and extinction in obsessive-compulsive disorder: A systematic review
2021, Neuroscience and Biobehavioral ReviewsCitation Excerpt :A recent meta-analysis of seven studies found that PTSD was associated with a robust amygdala response across acquisition and extinction, with aberrant insula and ACC activity related to PTSD during acquisition and vmPFC during extinction (Suarez-Jimenez et al., 2019). Similar results have been found in studies of panic disorder (e.g., Kircher et al., 2013; Schwarzmeier et al., 2019) and generalized anxiety disorder (e.g., Greenberg et al., 2013), although not all anxiety-related disorders are associated with this pattern of neural results, including specific phobia (e.g., Lange et al., 2019) and social anxiety disorder (e.g., Savage et al., 2020). Relatively consistent vmPFC and ACC results from the two neuroimaging studies of conditioning in OCD suggest that, at the neural level, OCD might have more in common with PTSD and the other anxiety-related disorders characterized by vmPFC and ACC dysfunction.
Neural adaptation of cingulate and insular activity during delayed fear extinction: A replicable pattern across assessment sites and repeated measurements
2021, NeuroImageCitation Excerpt :However, the neural trajectory may contain important additional information about learning types that could be beneficial for subtyping patients for treatment strategies. Schwarzmeier et al. (2019) found evidence for the importance of neural trajectories in accelerated acquisition of conditioned responses and attenuated extinction recall during delayed extinction training in patients with panic disorder, compared to healthy subjects. Since extinction processes depend on learning competing information challenging well established memories, changes over time are rather subtle and extinction learning is a fragile phenomenon (Bouton, 2002; Giustino and Maren, 2015).
Dysfunctional frontal activation of mismatch negativity in panic disorder: A magnetoencephalographic study
2021, Journal of Affective DisordersCitation Excerpt :Neuroimaging studies in panic disorder (PD) have indicated the importance of brain structures, such as the striatum, thalamus, amygdala, cingulate cortex, and prefrontal cortex (Demenescu et al. 2013; Engel et al. 2016; Feldker et al. 2016) in the processing of panic-provoking or emotional stimuli. There is also ample imaging evidence showing that patients with PD exhibited disturbances in the executive (top-down) function involving frontal and parietal regions (Dresler et al. 2012; Schwarzmeier et al. 2019). Recently, there is an increasing number of studies investigated the perceptual (bottom-up) processing of the neutral stimuli in patients with PD (Ludewig et al. 2005; Wise et al. 2009; Pfleiderer et al. 2010) since a clearer understanding of the cortical processing of sensory information in the brain would help understand the mechanisms of cognitive dysfunction in this clinical population.
Improvement on perception of welding noise by fountain sound masking: Insights from fMRI
2021, Applied AcousticsCitation Excerpt :By fMRI technology, some research carried out the auditory perception and analyzed the correlations between brain response and some fundamental acoustic indexes, such as frequency [21], sound pressure level (SPL) [22], and timbre [23]. Other studies pointed on the pathology of auditory related disease, such as schizophrenia [24], tinnitus [25], and panic disorder [26]. Moreover, the neural mechanisms of sound property identification [27], the individual differences in sound sensitivity [28], and sound source localization [29] were also studied by fMRI.