Reappraisal has been defined as an adaptive and antecedent-focused regulatory strategy and describes the process of changing the interpretation of an event that triggers an emotional response (Gross, 1998). In healthy participants, reappraisal engages a network of regions associated with cognitive control, (prefrontal cortex; PFC), conflict monitoring (anterior cingulate cortex; ACC), and semantic processing or perspective taking (middle temporal gyrus; Kanske et al., 2011; Buhle et al., 2014; see Figure 1).

Few studies, even though not explicitly stating that they study reappraisal, instructed participants to “down-regulate” negative emotions (New et al., 2009; Xiong et al., 2013; Schweizer et al., 2016). As the instructions resemble reappraisal, we review these studies together with direct reappraisal instructions. Summarizing the findings in trauma-exposed individuals with PTSD, most studies reported reduced activation of prefrontal regions during reappraisal, suggesting impaired top-down regulatory control during effortful emotion regulation. Specifically, the results showed reduced reappraisal-related activation in key prefrontal areas such as the dorsolateral PFC (New et al., 2009; Rabinak et al., 2014; Bryant et al., 2021), dorsomedial PFC and inferior frontal gyrus (IFG; Bryant et al., 2021; Keller et al., 2022). However, a closer examination reveals that only three studies showed consistent reductions in prefrontal activity in both whole-brain and region-of-interest (ROI) analyses when comparing trauma-exposed individuals with PTSD to healthy controls (New et al., 2009; Bryant et al., 2021) or to trauma-exposed controls without PTSD (New et al., 2009; Rabinak et al., 2014). Other studies, also reported reduced prefrontal activity, but did not observe effects using whole-brain analysis (Keller et al., 2022), or were of lower methodological quality and reported no between-condition contrast (Xiong et al., 2013) or no between-group results (Fitzgerald et al., 2017; Lee S. W. et al., 2021). As such, the results are not specifically attributable to reappraisal or group differences.

Interestingly, one study comparing trauma-exposed controls with and without PTSD distinguished between task instruction and strategy application while measuring brain activity (Butler et al., 2019). In contrast to the expected reduced activity in cognitive control and conflict monitoring regions, they reported heightened dorsal ACC activity in PTSD during strategy application. This finding diverges from other studies suggesting that some PTSD patients may exert greater effort during emotion regulation but with potentially reduced efficiency. Nevertheless, they found lower ventromedial PFC activation during the instruction phase in PTSD, aligning with theories of reduced regulatory control and highlighting the importance of differentiating between stages of emotion processing.

Notably, reduced brain activity does not necessarily indicate emotion dysregulation, as success is also reflected by reduced negative affect or arousal. For reappraisal, studies showed that people with PTSD reported higher negative ratings than controls (New et al., 2009; Butler et al., 2019). Within-group analyses revealed mixed findings: some reported reduced negative responses following reappraisal (vs. maintain/feel) in PTSD (Rabinak et al., 2014), while others found no differences (Butler et al., 2019). These results complicate interpreting reduced prefrontal engagement as a marker of emotion dysregulation but overall hint a PTSD-specific deficiency in reappraisal. However, when comparing trauma-exposed individuals without PTSD to healthy controls, some studies suggest that reduced frontal activation in combination with reduced negative affect (successful downregulation) may be more indicative of efficiency. More specifically, Schweizer et al. (2016) reported that young adults with (vs. without) experiences of early adversity exhibited more successful downregulation in regions such as the amygdala, middle frontal, and temporal areas. Based on this pattern of reduced activity along with effective downregulation of negative emotions, the authors suggested that the early adversity group may have developed a more efficient neural network for emotion regulation, as they were used to manage emotional distress during childhood. In support of this hypothesis, New et al. (2009) showed that trauma-exposed individuals without PTSD exhibited reduced reappraisal-related activity in the left superior and middle frontal gyri compared to healthy controls, while showing no group differences in self-reported affect after reappraisal trials. Similarly, another study reported reduced activity in orbitofrontal regions, but did not report between-group results on reappraisal success (Mao et al., 2023). At a behavioral level, within-group analysis yielded reduced negative affect following reappraisal (vs. maintain), which is also indicative of successful regulation (Rabinak et al., 2014; Butler et al., 2019).

Overall, comparing findings on people with and without PTSD indicate that the reduced prefrontal activity during reappraisal could be a specific effect related to PTSD but not to trauma exposure in general. Correspondingly, some studies also expected changes in the amygdala activation due to the failed prefrontal down-regulation after trauma exposure. However, there were no differences in amygdala activation in trauma-exposed individuals with compared to those without PTSD (Rabinak et al., 2014) or to healthy participants (Bryant et al., 2021) when using whole-brain or ROI analysis.

In summary, studies assessing neural underpinnings of reappraisal related to trauma exposure hint that specifically PTSD appears to be associated with reduced prefrontal engagement, in the dorsolateral PFC. Although there is a growing body of research assessing reappraisal, small sample sizes, lack of reported whole-brain results or between-group contrasts still make it difficult to draw final conlusions considering other brain regions.

In contrast to reappraisal, suppression targets the response directly by attempting to inhibit or prevent the full expression of the emotion but seems less effective (Gross, 1998; Gyurak et al., 2011). Similar to reappraisal, suppressing emotional expressions and memories engages prefrontal (e.g., dorsolateral, ventrolateral) and parietal regions (Guo et al., 2018; Sikka et al., 2022; see Figure 1). While expressive suppression has been specifically linked to reduced amygdala and insula activity, suggesting top-down control (Sikka et al., 2022), memory suppression involves striatal activation, indicating inhibitory pathways (Guo et al., 2018). Only few neuroimaging studies instructed participants to suppress negative emotions (Butler et al., 2019; Lee K. H. et al., 2021) or negative memories (Sullivan et al., 2019; Steward et al., 2020). Moreover, one study used instructions to suppress emotions but did not report any related results (Mao et al., 2023). Given the small set of studies, results are much more inconclusive compared to reappraisal trials.

Lee K. H. et al. (2021) reported no differences in prefrontal regions using ROI or corrected whole-brain analysis. However, with small volume correction, refugees (with and without PTSD) compared to healthy controls showed stronger activation in the lateral PFC related to suppression. Hence, refugees may exert more effort to regulate negative emotions, although suppression (vs. the control condition) did not show success on reducing the intensity of negative emotions. Similarly, Butler et al. (2019) reported no differences between combat-exposed individuals with and without PTSD in brain activity at the whole-brain and behavioral level.

Two studies used the Think-/No-Think paradigm, which assesses suppression of aversive memories rather than suppressing emotional responses. Using ROI analysis, Sullivan et al. (2019) found reduced activity in the middle frontal gyrus related to general and successful memory suppression for trauma-exposed people with and without PTSD compared to controls, suggesting a general effect of trauma, not specific to PTSD. In contrast, Steward et al. (2020) did not report similar findings. However, they reported that PTSD patients showed decreased parahippocampal activation during No-Think > Baseline at the whole-brain level compared to healthy controls. Because this contrast does not show brain activity unique to suppressing (vs. thinking about) a memory, it remains unclear whether the effect is suppression-specific or merely due to general attention effects.

In summary, none of these studies reported robust differences between people with and without PTSD and control groups related to emotion or memory suppression. The use of different comparisons, samples (e.g., mixed group with and without PTSD vs. each group separated), and correction methods makes it difficult to aggregate these findings, calling for more research on neural correlates of suppression in trauma-exposed people with and without PTSD compared to healthy controls. Thus, it remains unclear whether potential underlying neural mechanisms of suppression, such as reduced prefrontal activation, are due to the experience of trauma in general or specific to PTSD.

Other explicit emotion regulation strategies have been far less studied, although on a behavioral level various maladaptive regulation strategies have been linked to PTSD, including, rumination, worry, or self-blame (Seligowski et al., 2015; Kaczkurkin et al., 2017). We identified one previous neuroimaging study using a rumination induction task, which showed no differences between individuals with and without adverse childhood experiences, despite differences in functional connectivity were reported (Sokołowski et al., 2022).

Interestingly, one set of adaptive emotion regulation strategies has been neglected altogether in the neuroscientific research of PTSD, that is acceptance and compassion. While acceptance may be described as acknowledgement of the current states without being attached, or judgmental (Messina et al., 2021), compassion is defined as a caring feeling directed towards the suffering of others or to oneself (self-compassion; Neff, 2003; Goetz et al., 2010). When compassion is consciously evoked (e.g., through meditation) to reduce personal distress, it may be conceptualized as explicit emotion regulation (Engen and Singer, 2015). Compassion for others can be a special form of adaptive interpersonal emotion regulation, as it may be used not only to reduce personal distress in social situations, but also to maintain a connection with others (Engen and Singer, 2015). Since the induction of acceptance and compassion is usually associated with mindfulness-based interventions, studies already intersect with intervention research.

We identified one study, directly assessing compassion in people with PTSD though not as an explicit regulation strategy, but as direct emotional response towards the suffering of others, reflecting the propensity of compassion. Pino et al. (2016), reported reduced activation in the left anterior insula and left IFG in participants with PTSD compared to trauma-exposed controls during the question of how much empathic concern (compassion) they were feeling in response to pictures of people. This finding supports the idea that training of compassion might be promising target of future research.

Fonzo et al. (2017a, 2017b) investigated effects of prolonged exposure therapy on emotion regulation. Using ROI analysis they found a time-by-treatment effect indicating neural plasticity of reappraisal-related activation in the left middle frontal gyrus after prolonged exposure vs. waitlist (Fonzo et al., 2017b). In the same project, they did not find that reappraisal-related brain activity at baseline moderated the effect of treatment on symptom change (Fonzo et al., 2017a). These findings highlight that while exposure is associated with neural plasticity underpinning reappraisal, initial reappraisal-related brain activity seems not to be a marker of who will benefit most from treatment.

Another project assessed reappraisal ability before and after trauma-focused exposure including one session of cognitive reframing (Bryant et al., 2021; Korgaonkar et al., 2023). Contrary to the results of Fonzo et al. (2017b), here reduced dorsolateral PFC activation during reappraisal from pre- to post-treatment was associated with reduced PTSD symptoms after treatment (Korgaonkar et al., 2023). Contrasting the hypothesis of increased prefrontal activation, this finding could be explained by increased efficiency in down-regulating aversive emotions. However, they did not find a reappraisal-related time-by-group interaction, indicating that activity changes were not uniquely driven by the treatment.

In summary, these studies do show neural plasticity related to trauma-focused (exposure) therapy, but the exact mechanism remains unclear, as they observed both increased and decreased prefrontal activation. In contrast, one study combined exposure therapy with placebo or sertraline or applied medical treatment alone, but did not find significant differences between pre- and post-treatment (Joshi et al., 2020).

Mindfulness-based interventions have gained great attention for PTSD treatment (Boyd et al., 2018). Yet, we could not identify studies specifically investigating neural plasticity of compassion or acceptance in trauma-exposed people applying task-based fMRI at pre- and post-treatment. We did identify one study reporting increased resting-state connectivity of the posterior cingulate cortex the with dorsolateral PFC and dorsal ACC following mindfulness-based exposure therapy (including self-compassion exercises) in combat veterans with PTSD (King et al., 2016). Although resting-state connectivity is not the focus of our review, these findings provide initial evidence for emotion regulation-related neural plasticity in the context of mindfulness-based interventions in trauma-exposed individuals.

Within the last decade, real-time fMRI neurofeedback has shown potential in treating PTSD by promoting direct neuroplasticity. Using neurofeedback, participants are asked to regulate brain activity of a region, for instance, related to emotional experience. Changes in brain activity are visually presented to participants during training runs, followed by transfer runs without visual neurofeedback to assess learning. This form of regulation is—like explicit emotion regulation—a volitional control of the response to an emotional stimulus. When targeting the amygdala, Nicholson et al. (2017, 2018) reported that PTSD patients were able to downregulate amygdala activity in response to trauma-related words. This effect was sustained during transfer run, but did not increase between runs, indicating no learning. However, increased dorsolateral PFC activity between training runs suggested neuroplasticity, though this was not evident when comparing the first training and transfer run (Nicholson et al., 2018). The same research group showed that participants with PTSD and healthy controls were similarly able to decrease brain activity in the posterior cingulate cortex during downregulation vs. viewing of emotional words, without group differences (Nicholson et al., 2022; Lieberman et al., 2023).