The best magazine
Discussion
Based on Brothers (1990) model of social cognition, the aim of this study was to investigate affective ToM, as well as more basic stages of social cognition, i.e. face processing, and emotion recognition in BPD. It was assumed that patients show hyperactivation in the amygdala independent of the specific social-cognitive process, and with increasing complexity of the social-cognitive task have aberrations in brain activation in areas of the MNS, too.
Patients with BPD had no alterations on the behavioral level in any of the tested stages of social cognition. However, as hypothesized, they showed alterations in brain activation. We found evidence for general aberrations in the processing of social stimuli, as well as aberrations that were linked to the complexity of the social-cognitive task: patients with BPD showed hypoactivation in areas of the MNS and hyperactivation in the amygdala. While a hypoactivation of BA44 in BPD was found across all tasks, this difference between groups increased over the three stages of social cognition, with the most marked difference occurring during the attribution of intentions. In addition, only during affective ToM a hypoactivation of right STS was found (Supplementary Results). These data reflect that in healthy subjects, the recruitment of cerebral areas, which are assumed to be part of the MNS, was boosted with increasing demands on social-cognitive processes, as reported before (Mier et al., 2010a). In contrast, a comparable activation pattern could not be found in BPD patients. The hypoactivation of the STS is in agreement with a recent study by Dziobek et al. (2011) who found reduced STS-activation in BPD during cognitive empathy. Furthermore, and again in agreement with the literature (Herpertz et al., 2001; Minzenberg et al., 2007; Koenigsberg et al., 2009b), our data clearly point to a hyperactivation of the amygdala in BPD that was not modulated by task. Activation neither in areas of the MNS nor in the amygdala varied with the task-demands in the BPD-group. However, since we applied only social stimuli (faces) in this paradigm, it is not clear whether this activation pattern can be linked to the processing of social information or whether it is caused by a more basal perceptual aberration that would be evident with non-social material, too.
In addition, to these activation differences between groups in the pre-defined ROIs (amygdala, BA44 and STS), hyper- as well as hypo-activations could be shown in several others brain regions. Additionally, to the hyperactivation in the amygdala, we found a task-independent hyperactivation in the somatosensory cortex in the BPD group. Both the amygdala as well as the somatosensory cortex seem to be involved in emotional simulation processes (Adolphs and Spezio, 2006; Decety et al., 2008; Hooker et al., 2008). On the other hand, in addition to the described task modulated effect in BA44 and STS, we found a general hypoactivation in the thalamus and inferior prefrontal gyrus. These areas are associated with the conscious representation and simulation of actions and intentions (Coricelli, 2005; Iacoboni et al., 2005). It is supposed that the amygdala conveys information about emotional aspects of facial expressions via the insula to the inferior prefrontal cortex where action representation and the recognition of intentions take place (Carr et al., 2003; Pfeifer et al., 2008). Our data suggest that BPD patients and HCs achieve a comparable performance in their social-cognitive functions by a differential usage of a network of brain areas that were recently found to be involved in emotion simulation (Carr et al., 2003; Pfeifer et al., 2008) and affective ToM: BA44, STS, amygdala and somatosensory cortex (Mier et al., 2010a). While BPD patients show enhanced demands on areas associated with an emotional simulation process (Adolphs and Spezio, 2006; Decety et al., 2008; Hooker et al., 2008), HCs have enhanced demands on areas associated with the conscious representation of intentions with increasing complexity of the social-cognitive task (Coricelli, 2005). Thus, the results give evidence not only for general alterations in the processing of social stimuli (amygdala-hyperactivation) but also for alterations that are associated with the specific social-cognitive task (hypoactivation in areas of the MNS that was pronounced for affective ToM). This pattern of activation could represent a more rigid affect dominated processing of social stimuli in BPD that is not adaptive to the specific requirements of the particular social-cognitive task. This interpretation is in agreement with the clinical picture of BPD patients that is marked by high emotionality and a prominent emotion regulation deficit (Putnam and Silk, 2005). A recent study showed, indeed, that hyperactivation in the amygdala while viewing negative pictures was associated with deficits in emotion regulation in BPD (Niedtfeld et al., 2010).
Such an affect dominated processing of social stimuli in BPD could lead to mis-identification of intentions if there are indeed no affective aspects to be considered. Based on heightened amygdala activation, patients with BPD might have an increased sensitivity for negative emotions and intentions (Lynch et al., 2006) and are thus prone to search for negative intentions resulting in a negative bias (Wagner and Linehan, 1999; Dyck et al., 2009). On the other hand, deficits in intention recognition could also occur if intentions must be identified in situations that are highly emotional and self-relevant (Fonagy and Bateman, 2006; Domes et al., 2009). One might speculate that the increased usage of this more 'emotional network' for the identification of intentions, in combination with a reduction in prefrontal control mechanisms (Ruchsow et al., 2008; Koenigsberg et al., 2009a) interferes with the processing of the actual emotional content and thus result in false attribution of intentions in BPD patients. Domes et al. (2009) conclude that BPD patients' emotional arousal interferes with the cognitive processing of emotions which is mirrored in a limbic hyperactivation and a hypoactivation in frontal areas. Fonagy and Bateman (2008) assume that mentalizing deficits in BPD lead to a reduced ability to differentiate between the feelings of one and of others. Hence, one could imagine that a hyperactive amygdala, along with reduced prefrontal activation in BPD-patients during social interactions results in a feeling of threat that is directly transferred into the perception of the other person and results in the negative expectations of others.
However, since we could not observe any performance alterations or deficits in the BPD group, this emotional processing of social stimuli, i.e. face stimuli, seems to allow appropriate functioning at least for affective ToM tasks that are not presented in an emotional context. This assumption is in line with the findings from Harari et al. (2010), showing even enhanced abilities in affective ToM, but reduced abilities in cognitive ToM and with the results of the study by Fertuck et al. (2009) and our own data (Franzen et al., 2011) showing enhanced ability to recognize affect-related mental states in BPD. It might be that the amygdala hyperactivation in BPD cannot only be seen as a disadvantage that leads to emotion regulation problems and deficits in social interactions but under certain circumstances also as an advantage; i.e. that this amygdala hyperactivation helps overcoming deficits caused by an hypoactive MNS, and thereby supporting emotional interpersonal abilities. Although, it cannot be ruled out that deficits in ToM occur also in a standardized laboratory setting when more subtle methods are used to assess ToM-abilities (Preissler et al., 2011).
From a developmental point of view, one could speculate whether the hypoactivation in areas of the MNS and the hyperactivation in the amygdala seen in BPD might result from a reduced emotional validation of the BPD patients during childhood (Dziobek et al., 2011), what is a core assumption about the etiology of BPD (Linehan, 1993). The MNS is a network of brain areas that are used early in ontogeny and are inherent of plasticity. Even infants, as young as 6–12 month, show activation in areas of the MNS when predicting actions (Falck-Ytter et al., 2006; Southgate et al., 2009). Beyond this early development of the cerebral system, its activation adapts according to environmental demands, i.e. it can by re-organized and has plasticity (Pierno et al., 2009). Fonagy and Bateman (2008) assume that in BPD, the ability to mentalize develops only partially based on reduced mirroring of the emotional state of the child by the caretaker. Ghiassi et al. (2010) could indeed show that the quality of parental care is correlated with mentalizing abilities in patients with BPD. While healthy persons seem to learn understanding their own feelings and predicting actions of their caregiver by a mirroring process, this is refused to children who grow up in an invalidating environment and might lead to a more intuitive emotional assessment. This deviating processing strategy might be reflected in an activation pattern characterized by an amygdala hyperactivation and a MNS hypoactivation as observed in this study.
However, there are several shortcomings that have to be mentioned. The sample size was rather small, what results in a reduced power to detect group differences. It could be shown that the extent of clinical relevant depressive symptoms did not affect brain activation alterations in amygdala, BA 44 and STS. However, patients had various comorbidities and medication that might have influenced the neural processing of the social stimuli. Although all patients had a clinical diagnosis of BPD, the diagnosis and the diagnoses of comorbidities were not further confirmed by a diagnostic interview or clinical scales. Hence, we investigated a relatively heterogeneous sample of BPD patients, varying in gender, medication and occurrence of comorbidities. While this can be regarded as a representative sample of BPD patients (Zanarini et al., 1998; Zimmerman and Mattia, 1999), it holds the disadvantage that the contribution of specific symptom domains upon the investigated social cognition stages remains uncertain. Further studies with larger sample sizes are needed to investigate how different symptom expressions, the various comorbidities, and the medication might contribute to the observed alterations in brain activation during social cognition. Furthermore, it would be interesting for future studies to explore behavioral and neural differences between groups, depending on the valence of the emotional intentions.
In conclusion, to our knowledge, the study reported here is the first study in which the neuronal correlates of affective ToM were investigated in BPD. Our data suggest that BPD patients achieve comparable performance as HCs, not only in affective ToM but also in more basic social cognitions, such as basal face processing and emotion recognition with a differential recruitment of task-relevant brain areas. Patients with BPD had alterations in brain activation that were independent of the specific social-cognitive task (hyperactivation in the amygdala) that point to a deviant processing of social stimuli, in general, as well as to changes in brain activation that got augmented with increasing complexity of the social-cognitive task and point to alterations in more complex social cognitions, such as affective ToM (hypoactivation in areas of the MNS). This pattern of increased amygdala activation and decreased activation in areas of the MNS may reflect a more affect dominated processing of social stimuli that can be appropriate for task solving in affective ToM tasks but results in deficits in social situations that afford a less emotional and more cognitive interaction style.
Source: ...