The Relationship of Self-Relatedness and Emotional Processing

Transcription

1 Alexander Heinzel 1 and Georg Northoff 2 The Relationship of Self-Relatedness and Emotional Processing Abstract: Self and emotion have been discussed separately in philosophy, psychology, and neuroscience whereas their relationship has been rather neglected. We here suggest that they are intrinsically linked by demonstrating that self-relatedness accounts for the subjectivity of emotions, and emotional valence reflects affectivity of the self. Based on functional imaging studies, we discuss different possible spatio-temporal patterns of neural activity in anterior cortical midline structures (acms) that allow us to link and, at the same time, to distinguish self-relatedness and emotional valence. We conclude that the acms may be crucial in affectively colouring our self and associating our emotional feelings with the self. Keywords: self; self-related processing; emotions; emotional experience; anterior cortical midline structures (acms). Introduction The understanding of the self and emotions are among the most widely debated subjects in philosophy, psychology, and recently in Correspondence: Alexander Heinzel, MD, PhD, Dept. of Nuclear Medicine, RWTH Aachen University, Pauwelsstr. 30, Aachen, Germany. Phone: Fax: aheinzel@ukaachen.de [1] Dept. of Nuclear Medicine, RWTH Aachen University, Aachen, Germany; and Institute of Neuroscience and Medicine, Forschungszentrum Jülich, Jülich, Germany. [2] Mind, Brain Imaging and Neuroethics Research Unit, Institute of Mental Health Research, University of Ottawa, Ottawa, ON, Canada. Journal of Consciousness Studies, 21, No. 9 10, 2014, pp.????

2 2 A. HEINZEL & G. NORTHOFF neuroscience. Both the self and emotions can be considered umbrella concepts that include various distinct components. They are often considered as separate conceptions whereas their intricate relationship has been largely neglected. A possible reason might be that the self and emotions are often considered with regard to their functions within different kinds of cognition-based models of the brain. Although such models are of undeniable value for our understanding of the brain, they contain the potential problem of disregarding non-functional properties of the brain such as phenomenal experience and subjectivity. Consequently, these properties may often be neglected when considering the relationship between the self and emotions. In this paper, we analyse their relationship in detail by considering conceptual arguments and empirical findings by explicitly focusing on phenomenal experience and subjectivity. Firstly, we will demonstrate that at the core of the self there is an experiential component that is constitutional for any kind of self-related processing. Secondly, we will show that emotional processing is closely, if not necessarily, linked to self-relatedness. Thirdly, we will explore the different components of this relationship, focusing on the hedonic tone fundamentally involved in emotional processing and the relationship of mineness as a core constituent of self-related processing. Fourthly, we will consider empirical evidence from functional imaging studies suggesting that the anterior cortical midline regions may be of crucial importance for the interaction between self-related processing and emotional processing. Finally, we will discuss possible consequences for further research. The Affective Component of Self-Related Processing The question of how to understand one s own self has been one of the most widely discussed problems throughout many centuries of philosophical debate. In these philosophical debates, the conception of the self is usually related to questions of subjectivity and emotions. Some contemporary authors speak of a bodily, or proto-self (Panksepp, 1998; 2005; Damasio, 1999; 2010) while others focus more on cognitive and social components by emphasizing self-recognition, selfawareness, autobiographical memory, and ecological and narrative aspects (Gallagher, 2000; Turk et al., 2002; 2003; Keenan et al., 2003). Recently, the self has become a subject of neuroscientific investigation especially using various techniques of functional imaging. In

3 SELF-RELATEDNESS & EMOTIONS 3 order to address the self by functional imaging the conception of the self is often replaced by self-related processing (Northoff and Bermpohl, 2004; Northoff et al., 2006). Self-related processing represents an operationalization of the self that permits empirical investigation in experimental paradigms. It is sometimes intended to represent a purely scientific conception based on behaviour-related and neuroscientific data. However, such an operationalization is merely based on functional terms. It runs the risk of disregarding a core component that constitutes the self and also self-related processing: the affective component related to the subjective experience of the self and selfrelated processing. Let us first consider the self-related processing of computers. Computers are able to relate to their own processing, for example by evaluating how much memory certain types of processing need, etc. Moreover, they have at their disposal information on their hardware, their software, and whether they are connected to other devices. Thus, we may assume that they have information on their own processing and on their own physical constitution. In addition, we may assume that computers are in principle able to recognize this information as related to themselves and to distinguish such information from other information. Still all these kinds of processing do not seem comparable to self-related processing in human beings. For any human being the distinction between self-relatedness and non-self-relatedness is of crucial importance. Stimuli that concern our self are linked to many types of cognitive processing such as focusing of attention, but above all they are linked to a certain phenomenal experience. It somehow feels like being our self and we also somehow feel personally affected by certain stimuli as opposed to being a neutral observer. This difference is so fundamental for human beings that self-relatedness without this phenomenal experience, i.e. without a special qualitative feeling, does not seem complete. Afundamental constituent is missing without this qualitative component so that self-related processing in human beings appears impossible without it. With regard to computers, it is not plausible that a deeper investigation of their processing will reveal anything similar to the phenomenal quality that humans experience in self-related processing because it is not plausible that they have such a phenomenal quality at all. However, if this holds true then we cannot investigate human beings from a purely computational perspective, i.e. we cannot investigate their processing as if it was similar to the processing of a computer. The fundamental difference between pure information-based processing and self-related processing which includes a phenomenal

4 4 A. HEINZEL & G. NORTHOFF quality becomes evident in psychopathological phenomena where the pure information contradicts the phenomenal quality. Such phenomena comprise the objective loss of a body part combined with the persistent feeling of its presence and the objective presence of a body part combined with the feeling of strangeness as if the body part no longer belongs to one s own body. The first types of phenomena refer to phantom sensations. A striking property of phantom sensations is the feeling of their presence instead of their obvious absence. For example, some patients experience this presence even many years after the loss of a limb. The presence may be perceived as so real that they are still afraid to injure their eyes with extended phantom fingers. They are still and irresistibly linked to the experience of their own body although the body part is no longer present (Northoff, 2004; Heinzel, 2010). Similarly, some patients appear to lose the feeling of relatedness to their own body with regard to certain parts of the body. For example, some patients experience a feeling of strangeness or unrelatedness with respect to a limb after suffering a stroke. Again the objective information of the limb being attached to one s own body clearly contradicts the subjective feeling of it not being part of one s self. These examples demonstrate that self-relatedness and self-related processing are not purely neutral information and cannot be investigated as such. Such information contains a phenomenological quality that fundamentally shapes this kind of processing in human beings and if it is neglected self-related processing may not be adequately investigated in human beings. The Self-Relatedness of Emotional Processing Our analysis is based on the concept of emotions put forward by Panksepp (2005) as well as by Lambie and Marcel (2002). It shares many aspects of the component model of emotion presented by Scherer (1982; 2005). Like Scherer we hold that emotions include various different components including affective, cognitive, behavioural, expressive, and physiological changes as well as the fact that emotions can be considered to be episodes of interrelated changes of these different components that he calls the five organismic subsystems (Scherer, 2005). However, our characterization of feeling differs from his account. Whereas Scherer emphasizes the function of feeling (i.e. integration of appraisal-driven response organization in emotion), we focus on the phenomenal character of feeling characterized predominately by qualitativeness, referring to the experiential

5 SELF-RELATEDNESS & EMOTIONS 5 quality of the subjective experience, i.e. what it is like to experience an emotion (Nagel, 1974; Northoff, 2004; Northoff and Heinzel, 2006). It has to be noted that we do not identify emotion with affectivity. Affectivity is considered to be a component of emotion. This affective component of emotion is considered an intrinsic property in the sense that it cannot be explained in terms of cognitive functions (Northoff, 2012a,b). It refers to a fundamentally subjective experience and it presupposes a specific perspective. This perspective has two aspects; the first refers to the subjectivity of the experience and the second refers to the individuality of the experience. Emotional experience is subjective in the sense that it can only be experienced from the perspective of the first person. For example, it has been argued that extraterrestrial researchers coming from another planet may not understand what it is like for human beings to have emotions, if their mental structure is fundamentally different from ours. Thus, if they do not have a human first-person perspective they will be unable to understand first-person emotional experience in human beings (Nagel, 1974). Emotional experiences in this sense are not exclusively accessible by one individual person because every human being is able to take the point of view of the first person and may thus, in principle, understand emotional experience in other humans. However, emotional experience also contains an individual component that is related to the experiencing person. My experience of fear of a specific dog in a specific situation is only conceivable in the individual context of my person, i.e. my former experiences with dogs, my perception of the character of the actual dog, my general emotional state at that time point, etc. Another person in the same situation may not experience any fear at all. Therefore, my emotional experience of fear has to be considered as related to my individual self. From that point of view, it appears implausible that an emotional experience may not be self-related. Thus, we have established, on the one hand, that self-related processing has an affective component and, on the other hand, that emotional experience is closely linked to self-relatedness. In the next paragraph possible realizations of this linkage are discussed. Possible Linkages Between Self-Relatedness and Affectivity Our characterization of self-related processing is closely linked to what N Diaye, Sander and Vuilleumier (2009) refer to as self-relevance. However, their analysis focuses on the behavioural aspect of

6 6 A. HEINZEL & G. NORTHOFF self-relevance supporting a role for the appraisal of self-relevance in the decoding of facial expression. In our analysis, we assume similar behavioural effects, but we emphasize the phenomenal character of self-related processing in line with the conception of Lambie and Marcel (2002). According to them, a stimulus is linked to an individual organism by self-related processing. By this means, self-related processing allows the stimulus to be viewed through the eyes of the organism and thus to become mine. The concept of mineness describes the person s private and specific relationship with a particular stimulus, which is available only to that particular individual. The person s private and specific relationship is reflected in their phenomenal experience of the hedonic or affective tone of the stimulus. Subsequently mineness, as based upon self-relatedness, is essentially affective-phenomenal. Conversely, the affective component describes the phenomenal experience of emotions, i.e. emotional feelings. Emotional feelings can be characterized by emotional valence, a psychological concept that refers to their hedonic tone (i.e. pleasure versus displeasure) (Russell and Barrett, 1999; Colombetti, 2005). The hedonic tone is specific and private for a particular person, who consequently experiences its emotional feelings as mine and thus as subjective. However, the mechanism of how emotional valence or hedonic tone is linked to mineness remains unclear. Some authors (LeDoux, 2002; Rolls, 2005) associate the subjective nature of emotional feelings with metacognitive functions while others seem rather to link it to bodily and vegetative functions (Panksepp, 1998; 2005; Damasio, 1999). Since they associate mineness with either cognitive-reflective (LeDoux, Rolls) or vegetative-physiological (Panksepp, Damasio) functions, both solutions may miss the affective-phenomenal character of emotional feelings. How can mineness be linked to affectivity and how can emotional valence be associated with mineness? Mineness as a fundamental feature of self-relatedness constitutes any affective relationship towards stimuli. Stimuli that are not self-related at all are perceived as neutral because they are of no relevance in any sense for the perceiving subject. Therefore, there are no positive or negative affections towards non-self-related stimuli. In that sense, non-self-related stimuli lack any hedonic tone. On the other hand, if a stimulus is self-related this self-relatedness is strongly associated with positive or negative valence. Self-relatedness forces the perceiving subject to evaluate the stimuli with regard to possible positive or negative effects and subsequent behavioural adaptation in terms of approach to or withdrawal

7 SELF-RELATEDNESS & EMOTIONS 7 from the stimuli. Moreover, a higher degree of self-relatedness is assumed to be associated with stronger emotional experience than a lower degree of self-relatedness and vice versa. Thus, the hedonic tone of emotional experience is not only closely related to self-relatedness as such, but also to the degree of self-relatedness. The relationship of hedonic tone and self may be related to the construct of motivated self-processing (Beer, 2012; Flagan and Beer, 2013). Based on the premise that human beings tend to approach what is perceived as positive and avoid what is perceived as negative, the association of positive feelings with the self appears a logical consequence. Self-enhancement in order to see ourselves in a more positive light and also neglecting negative feedback while preferentially remembering positive feedback are psychological strategies to realize a more positive self-image. While our analysis focuses predominately on the phenomenal aspect of emotional and self-related processing, these factors must not be neglected. It may be speculated that in addition to the linkage on a phenomenal level based on the mineness of self-related processing and hedonic tone of emotional processing there might be a cognitively based linkage realized by motivated selfprocessing. So far it remains, however, unclear how these different linkages might interact. Does the analysis imply that self-relatedness is the more basal process and emotional processing is a secondary phenomenon? From a conceptual point of view such a hierarchical conception is not justified. One may argue that the hedonic tone of stimuli determines selfrelatedness because completely neutral stimuli are of no relevance and therefore not self-related. So far, it seems impossible to decide whether self-related processing or emotional processing is the more fundamental type of processing. Our analysis has demonstrated so far that they are closely linked to each other and that this linkage is based on the experiential component that lies at the core of self-related processing and emotional processing. This close linkage is evident for the personal relationships we have with our husbands, wives, children, close friends, etc. as well as personal objects such as our homes, our cars, etc. It may, however, be argued that a group of objects that belong to us are experienced as entirely neutral with regard to their hedonic tone. Many objects that we use in our daily life such as computers or mobile phones are selfrelevant but appear emotionally neutral. We argue that these objects as such may have no hedonic tone for us, but we need them to achieve goals that have a strong hedonic tone. For example, the cell phone is relevant because it may permit us to call close friends. Thus, the cell

8 8 A. HEINZEL & G. NORTHOFF phone has no direct hedonic tone itself, but it has an indirect hedonic tone as an instrument that permits us to stay in contact with our friends who have a strong hedonic tone for us. Without these friends or any other goal containing a hedonic tone those objects would lose their relevance for us. If I have nobody to call who interests me in any possible way, I wouldn t mind giving my cell phone away. Thus, we argue that the high degree of self-relevance of seemingly neutral objects derives from their relationship with objects or goals that have a strong emotional tone for us. It has to be noted that we do not argue that the hedonic tone is the only factor predicting the extent to which we relate certain stimuli to ourselves. But we argue that it is an essential factor and that self-relatedness without a direct or indirect emotional relationship is difficult to imagine. If we are completely neutral toward any given object it appears difficult to see how this object could become self-relevant for us. However, we do not hold that all stimuli which are considered positive are consequently highly self-related. Emotional tone is a necessary rather than a sufficient condition in this context. Moreover, stimuli with a negative emotional tone can also be of high self-relatedness. For example, the fear of dangerous situations can be highly self-relevant. In contrast, neutral stimuli are of low self-relatedness. Thus, any stimuli which a person does not consider to be self-related are likely to have a neutral emotional tone. Based on this analysis, we suggest that the hedonic tone or emotional valence of the stimulus determines its degree of self-relatedness and thus its strength of mineness in the person s phenomenal experience. This allows us to realize and express the degree of self-relatedness of the stimulus and thus the subjectivity in its hedonic tone or emotional valence and, at the same time, to associate the latter with self-relatedness and mineness in emotional feelings (see Figure 1 later). If this is true, different degrees of self-relatedness should be accompanied by different emotional valences, which is supported by both psychological and psychiatric evidence. Greenwald and Farnham (2000), for example, found that participants were much faster to respond when self-related items (e.g. their birth month) were paired with pleasant words than when self-related items were paired with unpleasant words. Another study (Moran et al., 2006) demonstrated that positively valenced highly self-relevant words were associated with faster reaction times than negatively valenced highly self-relevant words and,

9 SELF-RELATEDNESS & EMOTIONS 9 conversely, positively valenced low self-relevance words were associated with slower reaction times than negatively valenced low selfrelevance words. Moreover, they found that subjects were more likely to endorse positive information as self-descriptive and negative information as low in self-relevance. Depression, an affective disorder, is characterized by abnormal negative emotional feelings and negative self-focus (Rimes and Watkins, 2005) while patients with borderline personality disorder show feelings of inner emptiness of their self and emotions. These examples lend support to our assumption that the degree of self-relatedness of the stimulus may be closely linked to its hedonic tone and emotional valence, which, however, needs to be demonstrated directly in future studies. Neural Linkage Between Subjectivity and Affectivity Both self and emotions have been shown to involve the acms (Phan et al., 2002; Northoff and Bermpohl, 2004; Northoff et al., 2006). This raises not only the question of how subjectivity and affectivity can be linked, but also how they can be distinguished on neural grounds. We will now discuss different possible spatio-temporal patterns of neural activity in the acms that may allow for both linkage and distinction between self-relatedness and emotional valence. Linkage and distinction between self-relatedness and emotional valence may occur on neuroanatomical grounds. Imaging studies have demonstrated the correlation between emotional valence and neural activity in the medial orbitofrontal cortex (MOFC) (Anderson et al., 2003; Kringelbach et al., 2003; Small et al., 2003; Heinzel et al., 2005; Kringelbach, 2005; Grimm et al., 2006). Moreover, the MOFC seems to be involved in processing socio-emotional information permitting us to understand the emotions expressed by others, and to adjust our behaviour accordingly (Goodkind et al., 2012). In contrast, emotional intensity, reflecting arousal, is related rather to subcortical regions such as the tectum, the midbrain periaqueductal grey (PAG), etc. How is this related to the self? Phan et al. (2004) observed that activity in the amygdala was specifically associated with emotional picture intensity ratings while neural activity in the ventromedial prefrontal cortex (VMPFC) and the dorsomedial prefrontal cortex (DMPFC) (and the insula) correlated with their degree of self-relatedness. Neural activity in the ventral striatum/n.accumbens correlated with the picture ratings of both emotional intensity and self-relatedness. Moran et al. (2006) demonstrated differential

10 10 A. HEINZEL & G. NORTHOFF activity in the ventral anterior sub/precingulate cortex (vacc) depending on whether highly self-related words were valenced positively (high activity) or negatively (low activity). In contrast, neural activity in the VMPFC/DMPFC was determined solely by the degree of self-relatedness (high or low) of the words while remaining independent of positive and negative emotional valence. This is in line with findings by Fossati et al. (2003), who tested for the effects of the self-relatedness of emotional pictures and found the DMPFC to be active during self-judgment independent of whether it concerned positive or negative pictures (see also Ochsner et al., 2004). Based on these findings, we assume a ventral-dorsal dissociation in the acms. The ventral part including the MOFC and the vacc may be involved in processing emotional valence (MOFC) and its interaction with self-relatedness (vacc). Emotional valence and hedonic tone may converge with self-relatedness in the ventral part of the acms so that the former can signify the latter. While the exact neural mechanisms of such signifying remain to be explored, it should allow the realization and expression of different degrees of self-relatedness in terms of hedonic tone and emotional valence and, at the same time, the association of the latter with the former in emotional feeling. If this is true, one would expect changes in MOFC and vacc during emotion processing in depression and borderline personality disorders. Such changes have indeed been observed (Phillips et al., 2003; McCloskey Phan and Coccaro, 2005; Groenewold et al., 2012). Similar changes were also reported with regard to self-related processing. There is initial evidence that abnormally increased negative self-attribution as a hallmark of the increased self-focus in MDD might be mediated by altered neural activity in the midline structures (Grimm et al., 2009). Moreover, studies on patients with other mental health disorders indicated altered functionality of the orbitofrontal cortex and the medial prefrontal cortex (Heatherton, 2011). In contrast to the ventral part, the dorsal part of the acms, including the DMPFC, may be involved in self-related processing, particularly in judging and evaluating stimuli as self-related (Gusnard et al., 2001; Ochsner et al., 2005). This may allow us to distinguish selfrelatedness from emotional valence on a cognitive-reflective level. To that end, self-related judgment should be directly compared with emotional judgment, which needs to be investigated. Our assumption is very compatible with recent findings in subjects with a high degree of alexithymia. Such subjects can be characterized by difficulties in identifying subjective feelings and relating them to themselves. Contrasting subjects with high degrees of alexithymia to subjects with low

11 SELF-RELATEDNESS & EMOTIONS 11 degrees of alexithymia, an increased activation of the dorsal ACC was observed using various different kinds of emotional stimuli. Moreover, there was a positive correlation of the activation in the dorsal ACC with the individual degree of alexithymia (Heinzel et al., 2010). This might be interpreted as an increased effort in self-related processing reflected by an increased recruitment of associated neural resources when confronted with emotional stimuli. The impaired understanding of the emotional tone of stimuli would thus lead to an alteration of the self-related processing of the stimuli. Linkage and distinction between self-relatedness/mineness and emotional valence/hedonic tone could also be accounted for by the temporal course of neural activity over ventral and dorsal acms. EEG/MEG studies in emotion processing show an early ( ms) and a late ( ms) component (Aftanas et al., 2006; Herbert et al., 2006). In addition, several studies investigated the EEG correlates of selfrelated processing. Some studies focusing on event-related potential (ERP) found, in accordance with fmri studies, that medial cortical structures (especially MPFC and ACC) are crucial for self-related processing and that the time frame of this processing most frequently coincides with the P300 ERP component (Knyazev, 2013). For example, the processing of self-related possessive pronouns resulted in a significantly larger P300 amplitude than non-self-related possessive pronouns with sources of this activity in MPFC, anterior cingulate, and postcentral cortex (Zhou et al., 2010; Shi et al., 2011). These results were confirmed by Esslen et al. (2008), who compared selfreference versus other-reference using trait adjectives in reference to the self or to a close friend. They found that the MPFC was more active during self-related processing. In addition, Mu and Han (2010) demonstrated that self-referential traits versus other referential traits induced event-related synchronization of theta-band activity over the frontal area ( ms) and of alpha-band activity over the central area ( ms). Moreover, two studies found evidence for the interaction of emotional and self-related processing. Herbert et al. (2011a) looked at the effect of emotional valence on ERPs elicited by self-relevant and non-self-relevant pronoun-noun expressions. They found that the processing of self-related unpleasant words elicited greater frontal activity (from 350 ms onward), whereas self-related processing of pleasant words elicited greater parietal activity (from 450 ms onward). In addition, it was demonstrated that implicit self-reference modulates the processing of emotionally positive and negative nouns during a silent

12 12 A. HEINZEL & G. NORTHOFF reading task. It was shown that the early posterior negativity (EPN), a brain potential reflecting early conceptual processing of salient information, was greater for reading emotional nouns compared to neutral nouns (Herbert et al., 2011b). These results indicate that self-reference is involved in the very early processing of emotional stimuli. The ventral acms show high neural activity in the resting state (Raichle et al., 2001), which has been psychologically associated with self-related processing (Northoff and Bermpohl, 2004; McKiernan et al., 2006; Northoff et al., 2006). If self-related processing is indeed sustained, the subsequent resting state neural activity should be modulated by the degree of self-relatedness of the preceding stimulus rather than by its emotional valence. The degree of self-relatedness may then be coded by both the level of neural activity of the resting state and its duration. Unfortunately, no imaging studies investigating the impact of self-relatedness on the subsequent neural activity of the resting state have yet been reported. Future combined fmri-eeg/meg studies may also investigate whether early emotional valence and selfrelated processing occurs in the ventral acms whereas late sustained processing involves the dorsal acms. Conceptually, sustained neural processing could contribute to distinguishing self-relatedness from emotional valence. This may also be related to our phenomenal experience of a temporally extended self that contrasts with our more short-lived emotional feelings. In contrast, early neural processing may reflect temporal convergence by means of which self-relatedness can be associated with emotional valence. Linkage and distinction between self-relatedness/mineness and emotional valence/hedonic tone could also occur on procedural grounds. Some studies (Heinzel et al., 2005; Grimm et al., 2006) observed a bipolar continuum of neural activity in the VMPFC in relation to emotional valence, with higher activity coding positive valences and lower activity being associated with negative valences. This was contradicted by Lewis et al. (2007), who show neural coding of positive and negative emotional valence along independent axes or as absolute distance from neutral (hedonic value) located in different subregions in the ventral acms. Another option would be an exponential or U-shaped relationship between neural activity and positive and negative emotional valences. Such a relationship has been observed in the MOFC in response to financial reward (Elliott et al., 2003). In contrast to emotional valence, neural coding of self-relatedness remains to be investigated. Self-relatedness and emotional valence might be coded in the same way as a bipolar continuum in

13 SELF-RELATEDNESS & EMOTIONS 13 some regions, as for instance in the ventral acms, while they might be coded differentially or independently in other regions as, for instance, the dorsal acms. One might subsequently speak of affect-related and affect-unrelated processing of self-relatedness. Affect-related processing may link self-relatedness to emotional valence whereas affect-unrelated processing may distinguish self-relatedness from emotional valence. Different ways of neural processing, the way neural activity is modulated, may thus allow for both linkage and distinction between self-relatedness and emotional valence at the same time. Figure 1. Schematic presentation of the hypothesized overlap between self and emotions. The figure above points out the distinct components in both self and emotions and the interaction between the phenomenal-subjective component of the self and the affective-experiential component of emotions. Hedonic tone may contribute to selecting those stimuli which are self-related as distinguished from those that are non-selfrelated. This makes self-related stimuli accessible to (affective) experience as such. In contrast, mineness or self-relatedness is, at the same time, linked to emotional valence that makes the experience of emotional feelings essentially subjective. The mutual relationship is indicated by the arrows and permits the linkage between subjectivity and experience.

14 14 A. HEINZEL & G. NORTHOFF Conclusions and Open Questions to be Addressed in Future Studies We have demonstrated in this paper that self-related processing and emotions are closely linked to each other. Moreover, we have suggested that self-relatedness as a hallmark of the subjectivity of our self is intrinsically linked to emotional valence reflecting the affectivity of our emotions; the degree of self-relatedness of the stimulus is assumed to be signified by its hedonic tone or emotional valence. The intrinsic linkage between self-relatedness and emotional valence may be mediated by neural activity in the acms. Although the specific spatio-temporal patterns of neural activity allowing both linkage and distinction between self-relatedness and emotional valence remain unclear, the acms seem to be crucially involved in affectively colouring our self and associating our emotional feelings with the self. Based on our analysis, we will briefly address some points mainly related to the empirical studies discussed above that might be considered for future research. These points are related to studies on healthy subjects, studies on patients with altered emotional or self-related processing, and general methodological considerations. Firstly, although there are many functional imaging studies on healthy volunteers most of them only provide indirect evidence with regard to the relationship of self-related processing and emotional processing. It has not been clearly demonstrated whether there is a neural overlap between self-relatedness and emotional feelings. Moreover, the association of specific hedonic tones with self-relatedness should be explored in more detail, for example by questioning the extent to which self-relatedness is predominantly associated with positive emotional feelings compared to negative feelings. Furthermore, other aspects of emotions might be considered such as emotional intensity. In this paper, we focus on emotional valence processing, but the relation to emotional intensity remains unclear. Moreover, it would be of interest to investigate the temporal characterization of self-related processing in relation to emotional feelings especially focusing on the acms. Secondly, it has to be noted that we focus on the acms. This does not exclude the fact that other regions could be additionally implicated in the interaction of emotional and self-related processing. One possible candidate might be the insula. Studies indicate that the insula is involved in all kinds of subjective feelings since it is organized in network nodes that integrate all salient neural activity. Therefore, the insula has been termed the sentient self. It has been suggested that

15 SELF-RELATEDNESS & EMOTIONS 15 the anterior insula may create a cinemascopic model of human awareness and subjectivity (Craig, 2009; 2010; 2011). However, this concept does not address the processing of emotional valence. It seems that it is rather related to emotional arousal and may therefore not be involved in the specific interaction of the processing of self-relatedness and emotional valence processing (Northoff, 2012a). Thirdly, the investigation of patients with altered emotional or self-related processing is of interest. However, there are few patients who suffer from a purely emotionally based impairment. Mostly, such as in depressive patients or patients with personality disorders, various kinds of emotional, self-related, and cognitive processing are altered leading to a complex psychopathology. Therefore, it might be more interesting to look at otherwise healthy subjects with more subtle impairments of a specific kind. One interesting example is the case of the subjects discussed above with high degrees of alexithymia. These subjects may be selected in such a way that other kinds of mental disorders are excluded so that a purely emotionally based disorder could be investigated. With these subjects, it would be very interesting to directly compare emotional processing to self-related processing and to compare both types of processing to subjects with low degrees of alexithymia. Fourthly, in a recent review Flagan and Beer (2013) characterized the role of the vacc and the MOFC by motivational influence on social cognition. In our model, the same regions are regarded as being involved in processing emotional valence (MOFC) and its interaction with self-relatedness. Since the construct of the motivational self is closely linked to phenomenal aspects of the self it may be difficult to disentangle these two factors by empirical studies. Moreover, the role of the DMPFC was described as a region that indexes certainty about evaluation and the VMPFC is thought to reflect the type of information brought to bear on evaluations of people we know well (ibid.). In our model, the dorsal part of the acms is thought to be involved in self-related processing, particularly in judging and evaluating stimuli as self-related on a cognitive-reflective level. It may be of great interest to develop an experiment that allows a direct comparison of the two models. The experimental design has to be chosen carefully since both models are closely linked and most stimuli will not only evoke genuine target factors but also accompanying factors that are unavoidably involved in self-processing tasks. These concern, for example, the induction method as well as cognitive functions such as attention, working memory, and evaluation.

16 16 A. HEINZEL & G. NORTHOFF Fifthly, we argue that the relationship between self-related processing and emotional processing is closely linked to subjectivity and phenomenal experience. This raises specific methodological problems for empirical studies. As discussed above, subjectivity might be linked to the perspective of the first person. Therefore, a specific methodology has to be considered. A possible approach is first-person neuroscience (Varela, 1996; Varela and Shear, 1999; Panksepp, 1998; Northoff and Heinzel, 2006). Another methodological problem, difficult if not impossible to account for, is the exclusion of cognitive confounds in the study of self-relatedness and emotions because all of these processings seem to be accompanied by implicit cognitive processes such as implicit judgment as well as episodic and working memory, etc. We are thus confronted with the problem of excluding cognitive confounds in imaging paradigms focusing on emotional feeling and self-relatedness. In addition to improvements in our empirical designs, future methodological and philosophical discussions are needed to discover whether these methodological limitations are of a fundamental nature thus setting apriorilimits to empirical investigation of the self and emotions, or whether they can be overcome in the future by developing more appropriate methods, techniques, and designs. References Aftanas, L.I., Reva, N.V., Savotina, L.N. & Makhnev, V.P. (2006) Neurophysiological correlates of induced discrete emotions in humans: An individually oriented analysis, Neuroscience and Behavioral Physiology, 36, pp Anderson, A.K., Christoff, K., Stappen, I., Panitz, D., Ghahremani, D.G., Glover, G., Gabrieli, J.D. & Sobel, N. (2003) Dissociated neural representations of intensity and valence in human olfaction, Nature Neuroscience, 6, pp Beer, J.S. (2012) This time with motivation: The implications of social neuroscience for research on motivated self- and other-perception, Motivation and Emotion, 36, pp Colombetti, G. (2005) Appraising valence, Journal of Consciousness Studies, 12 (8 10), pp Craig, A.D. (2009) How do you feel now? The anterior insula and human awareness, Nature Reviews Neuroscience, 10, pp Craig, A.D. (2010) The sentient self, Brain Structure and Function, 214, pp Craig, A.D. (2011) Significance of the insula for the evolution of human awareness of feelings from the body, Annals of the New York Academy of Sciences, 1225, pp Damasio, A.R. (1999) The Feeling of What Happens: Body and Emotion in the Making of Consciousness, New York: Harcourt Brace. Damasio, A.R. (2010) Self Comes to Mind: Constructing the Conscious Brain, New York: Harcourt Brace. Elliott, R., Newman, J.L., Longe, O.A. & Deakin, J.F. (2003) Differential response patterns in the striatum and orbitofrontal cortex to financial reward in humans:

17 SELF-RELATEDNESS & EMOTIONS 17 A parametric functional magnetic resonance imaging study, Journal of Neuroscience, 23, pp Esslen, M., Metzler, S., Pascual-Marqui, R. & Jancke, L. (2008) Pre-reflective and reflective self-reference: A spatiotemporal EEG analysis, Neuroimage, 42, pp Flagan, T. & Beer, J.S. (2013) Three ways in which midline regions contribute to self-evaluation, Frontiers in Human Neuroscience, 7, p Fossati,P.,Hevenor,S.J.,Graham,S.J.,Grady,C.,Keightley,M.L.,Craik,F.& Mayberg, H. (2003) In search of the emotional self: An fmri study using positive and negative emotional words, American Journal of Psychiatry, 160, pp Gallagher, I.I. (2000) Philosophical conceptions of the self: Implications for cognitive science, Trends in Cognitive Sciences, 4, pp Goodkind, M.S., Sollberger, M., Gyurak, A., Rosen, H.J., Rankin, K.P., Miller, B. & Levenson, R. (2012) Tracking emotional valence: The role of the orbitofrontal cortex, Human Brain Mapping, 33, pp Greenwald, A.G. & Farnham, S.D. (2000) Using the implicit association test to measure self-esteem and self-concept, Journal of Personality and Social Psychology, 79, pp Grimm, S., Schmidt, C.F., Bermpohl, F., Heinzel, A., Dahlem, Y., Wyss, M., Hell, D., Boesiger, P., Boeker, H. & Northoff, G. (2006) Segregated neural representation of distinct emotion dimensions in the prefrontal cortex an fmri study, Neuroimage, 30, pp Grimm, S., Ernst, J., Boesiger, P., Schuepbach, D., Hell, D., Boeker, H. & Northoff, G. (2009) Increased self-focus in major depressive disorder is related to neural abnormalities in subcortical-cortical midline structures, Human Brain Mapping, 30, pp Groenewold, N.A., Opmeer, E.M., De Jonge, P., Aleman, A. & Costafreda, S.G. (2012) Emotional valence modulates brain functional abnormalities in depression: Evidence from a meta-analysis of fmri studies, Neuroscience and Biobehavioral Reviews, 37, pp Gusnard, D.A., Akbudak, E., Shulman, G.L. & Raichle, M.E. (2001) Medial prefrontal cortex and self-referential mental activity: Relation to a default mode of brain function, Proceedings of the National Academy of Sciences USA, 98, pp Heatherton, T.F. (2011) Neuroscience of self and self-regulation, Annual Review of Psychology, 62, pp Heinzel, A., Bermpohl, F., Niese, R., Pfennig, A., Pascual-Leone, A., Schlaug, G. & Northoff, G. (2005) How do we modulate our emotions? Parametric fmri reveals cortical midline structures as regions specifically involved in the processing of emotional valences, Cognitive Brain Research, 25, pp Heinzel, A. & Heinzel, T. (2010) The phenomonology of virtual reality and phantom sensations, Studia UBB. Philosophia, LV, pp Heinzel, A., Schafer, R., Muller, H.W., Schieffer, A., Ingenhag, A., Eickhoff, S.B., Northoff, G., Franz, M. & Hautzel, H. (2010) Increased activation of the supragenual anterior cingulate cortex during visual emotional processing in male subjects with high degrees of alexithymia: An event-related fmri study, Psychotherapy and Psychosomatics, 79, pp Herbert, C., Kissler, J., Junghofer, M., Peyk, P. & Rockstroh, B. (2006) Processing of emotional adjectives: Evidence from startle EMG and ERPs, Psychophysiology, 43, pp

18 18 A. HEINZEL & G. NORTHOFF Herbert, C., Herbert, B.M., Ethofer, T. & Pauli, P. (2011a) His or mine? The time course of self-other discrimination in emotion processing, Society for Neuroscience, 6, pp Herbert, C., Pauli, P. & Herbert, B.M. (2011b) Self-reference modulates the processing of emotional stimuli in the absence of explicit self-referential appraisal instructions, Social Cognitive and Affective Neuroscience, 6, pp Keenan, J.P., Wheeler, M., Platek, S.M., Lardi, G. & Lassonde, M. (2003) Selfface processing in a callosotomy patient, European Journal of Neuroscience, 18, pp Knyazev, G.G. (2013) EEG correlates of self-referential processing, Frontiers in Human Neuroscience, 7, p Kringelbach, M.L. (2005) The human orbitofrontal cortex: Linking reward to hedonic experience, Nature Reviews Neuroscience, 6, pp Kringelbach, M.L., O Doherty, J., Rolls, E.T. & Andrews, C. (2003) Activation of the human orbitofrontal cortex to a liquid food stimulus is correlated with its subjective pleasantness, Cerebral Cortex, 13, pp Lambie, J.A. & Marcel, A.J. (2002) Consciousness and the varieties of emotion experience: A theoretical framework, Psychological Review, 109, pp Ledoux, J.E. (2002) Synaptic Self: How Our Brains Become Who We Are, New York: Viking. Lewis, P.A., Critchley, H.D., Rotshtein, P. & Dolan, R.J. (2007) Neural correlates of processing valence and arousal in affective words, Cerebral Cortex, 17, pp McCloskey, M.S., Phan, K.L. & Coccaro, E.F. (2005) Neuroimaging and personality disorders, Current Psychiatry Reports, 7, pp McKiernan, K.A., D Angelo, B.R., Kaufman, J.N. & Binder, J.R. (2006) Interrupting the stream of consciousness : An fmri investigation, Neuroimage, 29, pp Moran, J.M., Macrae, C.N., Heatherton, T.F., Wyland, C.L. & Kelley, W.M. (2006) Neuroanatomical evidence for distinct cognitive and affective components of self, Journal of Cognitive Neuroscience, 18, pp Mu, Y. & Han, S. (2010) Neural oscillations involved in self-referential processing, Neuroimage, 53, pp N diaye, K., Sander, D. & Vuilleumier, P. (2009) Self-relevance processing in the human amygdala: Gaze direction, facial expression, and emotion intensity, Emotion, 9, pp Nagel, T. (1974) What is it like to be a bat?, Philosophical Review, 83, pp Northoff, G. (2004) Philosophy of the Brain: The Brain Problem, Amsterdam: John Benjamins. Northoff, G. (2012a) From emotions to consciousness a neuro-phenomenal and neuro-relational approach, Frontiers in Psychology, 3, p Northoff, G. (2012b) Unlocking the Brain. Volume II: Consciousness, Oxford: Oxford University Press. Northoff, G. & Bermpohl, F. (2004) Cortical midline structures and the self, Trends in Cognitive Sciences, 8, pp Northoff, G. & Heinzel, A. (2006) First-person neuroscience: A new methodological approach for linking mental and neuronal states, Philosophy, Ethics & Humanities in Medicine, 1, E3. Northoff, G., Heinzel, A., De Greck, M., Bermpohl, F., Dobrowolny, H. & Panksepp, J. (2006) Self-referential processing in our brain a meta-analysis of imaging studies on the self, Neuroimage, 31, pp

19 SELF-RELATEDNESS & EMOTIONS 19 Ochsner, K.N., Knierim, K., Ludlow, D.H., Hanelin, J., Ramachandran, T., Glover, G. & Mackey, S.C. (2004) Reflecting upon feelings: An fmri study of neural systems supporting the attribution of emotion to self and other, Journal of Cognitive Neuroscience, 16, pp Ochsner, K.N., Beer, J.S., Robertson, E.R., Cooper, J.C., Gabrieli, J.D., Kihsltrom, J.F. & D Esposito, M. (2005) The neural correlates of direct and reflected self-knowledge, Neuroimage, 28, pp Panksepp, J. (1998) Affective Neuroscience: The Foundations of Human and Animals Emotions, New York: Oxford University Press. Panksepp, J. (2005) Affective consciousness: Core emotional feelings in animals and humans, Consciousness and Cognition, 14, pp Phan, K.L., Wager, T., Taylor, S.F. & Liberzon, I. (2002) Functional neuroanatomy of emotion: A meta-analysis of emotion activation studies in PET and fmri, Neuroimage, 16, pp Phan, K.L., Taylor, S.F., Welsh, R.C., Ho, S.H., Britton, J.C. & Liberzon, I. (2004) Neural correlates of individual ratings of emotional salience: A trial-related fmri study, Neuroimage, 21, pp Phillips, M.L., Drevets, W.C., Rauch, S.L. & Lane, R. (2003) Neurobiology of emotion perception II: Implications for major psychiatric disorders, Biological Psychiatry, 54, pp Raichle, M.E., Macleod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A. & Shulman, G.L. (2001) A default mode of brain function, Proceedings of the National Academy of Sciences USA, 98, pp Rimes, K.A. & Watkins, E. (2005) The effects of self-focused rumination on global negative self-judgements in depression, Behaviour Research and Therapy, 43, pp Rolls, E.T. (2005) Emotion Explained, New York: Oxford University Press. Russell, J.A. & Barrett, L.F. (1999) Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant, Journal of Personality and Social Psychology, 76, pp Scherer, K.R. (1982) Emotion as a process: Function, origin, and regulation, Social Science Information, 21, pp Scherer, K.R. (2005) What are emotions? And how can they be measured?, Social Science Information, 44, pp Shi, Z., Zhou, A., Liu, P., Zhang, P. & Han, W. (2011) An EEG study on the effect of self-relevant possessive pronoun: Self-referential content and first-person perspective, Neuroscience Letters, 494, pp Small, D.M., Gregory, M.D., Mak, Y.E., Gitelman, D., Mesulam, M.M. & Parrish, T. (2003) Dissociation of neural representation of intensity and affective valuation in human gustation, Neuron, 39, pp Turk, D.J., Heatherton, T.F., Kelley, W.M., Funnell, M.G., Gazzaniga, M.S. & Macrae, C.N. (2002) Mike or me? Self-recognition in a split-brain patient, Nature Neuroscience, 5, pp Turk, D.J., Heatherton, T.F., Macrae, C.N., Kelley, W.M. & Gazzaniga, M.S. (2003) Out of contact, out of mind: The distributed nature of the self, Annals of the New York Academy of Sciences, 1001, pp Varela, F. (1996) Neurophenomenology: A methodological remedy for the hard problem, Journal of Consciousness Studies, 3 (4), pp Varela, F. & Shear, J. (1999) First-person methodologies: What, why, how?, Journal of Consciousness Studies, 6 (2 3), pp Zhou, A., Shi, Z., Zhang, P., Liu, P., Han, W., Wu, H., Li, Q., Zuo, Q. & Xia, R. (2010) An ERP study on the effect of self-relevant possessive pronoun, Neuroscience Letters, 480, pp