Dental phobia - one form of specific phobia - is characterized by an excessive fear in response to and a strong avoidance of phobic situations  as for instance dental treatment. Regarding the development of dental phobia, individuals with high levels of dental anxiety report distressing dental experiences up to 18 times more often than low-anxious individuals [2, 3]. Classical conditioning is considered as one important mechanism underlying the development of phobic fear in response to distressing events [4, 5]. Thereby, a former neutral stimulus (e.g. dentist) acquires a negative affective value by being paired with an unconditioned stimulus (e.g. pain during dental treatment) that elicits an unconditioned response (e.g. fear response). Subsequently, the former neutral stimulus (conditioned stimulus) elicits a conditioned response (e.g. fear) that is similar to the unconditioned response, without being paired with the unconditioned stimulus again. Numerous studies have been conducted in recent years trying to elucidate the neural mechanisms underlying the acquisition of fear in healthy individuals as well as the neural basis underlying phobic fear responses (for an overview see ). Supporting a learning theory-based perspective on specific phobia, fear conditioning in healthy subjects and symptom provocation in specific phobia seem to be characterized by similar neural correlates including enhanced amygdala, insula, and dorsal anterior cingulate cortex (dACC) activation (for an overview see ). In dental phobia, activation has also been found in basal ganglia and prefrontal cortex regions [7, 8].
Based on learning theory, the extinction of conditioned responses is supposed to be the central process underlying exposure based treatments in anxiety disorders as for instance specific phobia . Extinction is defined as the repeated confrontation with a conditioned stimulus without subsequent presentation of the unconditioned stimulus, leading to a reduction of conditioned responding over time. Especially the ventromedial prefrontal cortex (vmPFC) has been emphasized as a key structure underlying extinction learning and retrieval (for an overview see ). Correspondingly, reduced vmPFC activation has been shown in patients with specific phobias during symptom provocation [11–13]. Studies investigating the effects of cognitive-behavioral therapy (CBT) on neural correlates of symptom provocation in spider phobia indicate reduced activation of the insula and the dACC  as well as increased activation of the vmPFC  as a result of successful CBT. All in all, these findings demonstrate that fear acquisition and symptom provocation as well as fear extinction and symptom modification via CBT are characterized by similar neural circuits.
Despite the vast literature on the neural correlates of symptom provocation in specific phobia and the few studies on the effects of CBT on these neural correlates, little is known about the mechanisms of action (e.g. habituation, extinction) underlying such processes of change in phobic patients. There is one H2 15O-positron emission tomography study showing habituation of several brain regions including the amygdala and the insula during prolonged exposure to phobic compared with non-phobic stimuli in spider phobic subjects . These results give first evidence for the neural basis underlying habituation or extinction processes in specific phobia. Generally, it has been claimed that fear extinction in humans comprises more than mere passive learning, namely cognitive processes , which might additionally influence conditioned responding during extinction. In line with this, cognitive interventions are important components of CBT . Moreover, it has been shown that adding guided threat reappraisal to exposure treatment in specific phobia led to enhanced between-trial habituation  as well as decreased return of fear .
Cognitive reappraisal is one prominent form of cognitive emotion regulation defined as reinterpreting a stimulus or situation in a way that reduces its emotional impact . Several functional magnetic resonance imaging (fMRI) studies investigating the neural basis of emotion regulation via cognitive reappraisal found enhanced activation of lateral and medial prefrontal cortical control regions accompanied by reduced activation of emotional arousal-related brain structures like the amygdala and the insula during (successful) emotion regulation (for an overview see ). One of our own studies investigated cognitive reappraisal of phobic compared with general aversive stimuli in specific phobia . Results of this study show that cognitive down-regulation of emotional responses to phobic pictures led to reduced activation of insula and dACC compared with just looking at the pictures. In addition, down-regulation of phobic compared with general aversive emotional responses was associated with an increased regulation effort and diminished activation of the right rostral ACC (rACC) and the dorsomedial PFC (dmPFC), regions crucially involved in the cognitive regulation of emotions . Altogether, these results indicate that patients with specific phobia exhibit a phobia-specific regulation deficit reflected in a dysfunctional recruitment of rACC and dmPFC.
Beyond that, previous research has shown that individuals vary in their habitual use of cognitive reappraisal (dispositional cognitive reappraisal) as a strategy to regulate emotions. This individual tendency has been shown to be stable in time . A more frequent use of cognitive reappraisal has been related to better interpersonal functioning, enhanced psychological well-being, and reduced depressive symptoms . On the neurobiological level, dispositional cognitive reappraisal has been found to be correlated with reduced insula activation during the anticipation of negative affective stimuli , with reduced amygdala and stronger dACC responses while viewing negative emotional faces , and with stronger dACC and dorsolateral prefrontal cortex (dlPFC) responses during response inhibition towards negative emotional material (sad vs. happy faces; ). Moreover, studies investigating brain structural correlates found an association of dispositional cognitive reappraisal with dACC  and vmPFC  gray matter volumes. In sum, these numerous findings regarding individual differences in cognitive reappraisal usage indicate a pivotal role of this factor in the modulation of emotional responses.
Despite generally large effect sizes of exposure-based treatments in specific phobia , one challenging future task is to figure out possible factors contributing to individual differences in treatment response and return of fear at follow-up. Considering the abovementioned correlates of dispositional cognitive reappraisal, this factor might be a promising candidate for modulating treatment responses in specific phobia. Individuals might differ in the extent they use cognitive reappraisal to regulate their emotions during an exposure session. This might moreover result in different outcomes of exposure therapy at the end of treatment or at follow-up. Previous results on mechanisms of change in CBT have shown that the ‘performance’ during exposure (e.g. rate of habituation, initial fear responding) is not a good predictor for final therapy success . Furthermore, Craske and colleagues assume that the toleration of fear during exposure is far more important to exposure therapy than the reduction of fear. This general ability to manage ones own emotions is also represented in the strategy of cognitive reappraisal. However, until now there are no studies investigating the influence of individual differences in cognitive reappraisal usage on the neural correlates of extinction-related processes in anxiety disorders, as for instance specific phobia. Because it is difficult to conduct typical exposure therapy during scanning, one first approximation is to investigate the time course of brain activation during symptom provocation. The repeated presentation of phobic stimuli without negative consequences (e.g. pain) might be a plausible equivalent to emotional relearning/extinction learning, habituation as well as to a first exposure session in the treatment of specific phobias.
Hence, we investigated a sample of 27 dental phobic patients and 21 healthy control subjects who underwent fMRI during symptom provocation with phobia-specific, generally disgust and fear inducing as well as affectively neutral pictures. The aim of the study was to identify brain structures showing differential activation over time for phobic compared with neutral stimuli depending on the extent of habitual cognitive reappraisal usage in phobic patients. The emotion regulation questionnaire (ERQ, . German version ) was used to measure individual differences in habitual cognitive reappraisal usage. A stronger dispositional use of cognitive reappraisal is supposed to lead to enhanced habituation or emotional relearning (e.g. extinction) because these individuals might be better at spontaneously using reappraisal during symptom provocation. This is expected to result in a stronger reduction of activation in brain regions associated with phobia specific processing (amygdala, insula, dACC, basal ganglia). In addition, extinction-related structures as for instance the vmPFC should exhibit less activation decrease over time because individuals with more pronounced habitual reappraisal use are supposed to show stronger emotional relearning respectively extinction learning. Furthermore, cortical control-related areas (rACC, dmPFC, vlPFC, dlPFC) involved in cognitive reappraisal processes are assumed to exhibit a different time course as a function of habitual cognitive reappraisal usage. It is conceivable that subjects high in cognitive reappraisal are characterized by enhanced activation in these areas over time due to sustained cognitive reappraisal of phobic stimuli during the course of symptom provocation. On the other hand, these individuals might be more effective in reducing emotional responses and therefore might show a stronger reduction of activation in these PFC structures over time because regulation might take place faster automatically rather than by explicit and effortful cognitive reappraisal processes. Additionally, we were interested in the specificity of the effects and therefore compared phobic with general affective responses (fear and disgust) as well as phobic individuals with healthy control subjects.