Interacting and dissociable effects of alexithymia and depression on empathy

Accepted Manuscript Interacting and dissociable effects of alexithymia and depression on empathy Christian Banzhaf , Ferdinand Hoffmann , Philipp Kanske , Yan Fan , Henrik Walter , Stephanie Spengler , Stefanie Schreiter , Tania Singer , Felix Bermpohl PII: S0165-1781(18)30235-X DOI: https://doi.org/10.1016/j.psychres.2018.10.045 Reference: PSY 11821 To appear in: Psychiatry Research Received date: 10 February 2018 Revised date: 18 October 2018 Accepted date: 19 October 2018 Please cite this article as: Christian Banzhaf , Ferdinand Hoffmann , Philipp Kanske , Yan Fan , Henrik Walter , Stephanie Spengler , Stefanie Schreiter , Tania Singer , Felix Bermpohl , Interact- ing and dissociable effects of alexithymia and depression on empathy, Psychiatry Research (2018), doi: https://doi.org/10.1016/j.psychres.2018.10.045 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 1 Highlights  Empathy deficits in participants with high alexithymia regarding cognitive empathy.  Particularly high personal distress in depression when affected by alexithymia.  Alexithymia seems to affect empathy at nearly all levels. T  Empathy deficits in depression can mostly be attributed to concurrent alexithymia. IP CR US AN M ED PT CE AC  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 2 Interacting and dissociable effects of alexithymia and depression on empathy Christian Banzhaf,a,*,1 Ferdinand Hoffmann, b,1 Philipp Kanske,b, d Yan Fan,c Henrik Walter,a Stephanie Spengler,a Stefanie Schreiter,a Tania Singer,b Felix Bermpohla T a IP Department of Psychiatry and Psychotherapy, Charité Campus Mitte, Universitätsmedizin Berlin, Germany CR b Department of Social Neuroscience, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany c US Department of Psychiatry and Psychotherapy, Charité Campus Benjamin Franklin, Universitätsmedizin Berlin, Germany AN d Institute of Clinical Psychology and Psychotherapy, Department of Psychology, Technische Universität Dresden, Dresden, Germany 1 These authors contributed equally to this work. M ED PT * Corresponding author. Christian Banzhaf CE Psychiatric University Hospital Charité at St. Hedwig Hospital AC Große Hamburger Str. 5-11 10 115 Berlin T: 0049/179/2985062 E-mail: christian.banzhaf@charite.de  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 3 Abstract Major-depressive-disorder (MDD) and alexithymia have both been associated with empathy T deficits. We examined whether depression and alexithymia show dissociable or interacting IP effects on cognitive and emotional trait and state empathy. Healthy controls with high and low CR alexithymia and MDD-patients with high and low alexithymia were assessed. We used the Interpersonal–Reactivity-Index-questionnaire (IRI) for trait cognitive and emotional empathy US and the Multifaceted-Empathy-Test (MET) for state cognitive and emotional empathy. Firstly, we found a main effect of alexithymia, irrespective of depression, on trait and state cognitive AN empathy: High alexithymia subjects showed lower scores in perspective taking (IRI) and in the cognitive-empathy-component of the MET. Secondly, we found main effects of alexithymia and M depression on trait emotional empathy (IRI-subscale personal distress). Moreover, we found a ED significant depression-by-alexithymia-interaction on trait emotional empathy: MDD-patients showed particularly high personal distress when affected by alexithymia (IRI). Thirdly, PT alexithymia and depression had no impact on state emotional empathy (MET). However, analyzing positive and negative trials separately, we found more emotional empathy in MDD- CE patients concerning negatively valenced stimuli. Our data suggest dissociable and interacting AC effects of MDD and alexithymia on empathy. Importantly, except for heightened personal distress, empathy deficits in MDD-patients were entirely due to concurrent alexithymia. Keywords: empathy; depression; alexithymia  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 4 1. Introduction 1.1. Effects of depression on empathy Depression is often associated with deficits in social functioning (Hirschfeld et al., 2000). One important reason for impaired social functioning in Major Depressive Disorder (MDD) might be T deficits in understanding other minds. Fundamental for understanding other minds is the ability IP to understand and share the feelings of others. This ability has been called empathy CR (DeVignemont and Singer, 2006), a multifaceted construct that according to some researchers involves at least three aspects (e.g., Decety and Jackson, 2004; Gonzalez-Liencres et al., 2013): US Firstly, a cognitive component concerning the understanding of emotional states of others. Secondly, an emotional component enabling sharing other people‟s emotional states. Thirdly, a AN mechanism featuring the distinction between self and other. Self-other-distinction refers to the ability to be aware of the other‟s context and to distinguish between the other person's and one's M own situation. Without this ability, an empathic response is considered to be a primitive form of ED empathy, referred to as emotional contagion (DeVignemont and Singer, 2006; Lamm et al. 2007). This mere reactive response to another‟s condition can lead to personal distress, which is PT characterized by the degree a person feels uncomfortable when confronted with the suffering of others (Davis, 1980; 1983; Batson, 2009; Singer and Klimecki, 2014). CE Investigations regarding empathy in depression are sparse and findings are inconsistent. There is AC greatest consensus concerning the effect of depression on trait emotional empathy, more specifically concerning the findings of increased personal distress (O‟Connor et al., 2002; Wilbertz et al., 2010; Thoma et al., 2011; Cusi et al., 2011; for a review see Schreiter et al., 2013). Cusi et al. (2011) found empathy deficits in depression in both the cognitive (perspective taking) and emotional component (empathic concern) of empathy using the IRI, whereas Thoma  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 5 et al. (2011) using the same instrument did not. In addition to examining trait empathy by applying the IRI, Thoma et al. (2011) explored state empathy using the Multifaceted Empathy Test (MET; Dziobek et al., 2008). This photo-based test measures cognitive and emotional components of empathy. However, Thoma and collegues found no alterations between patients and controls. Moreover, no alternations concerning cognitive empathy were found by T Wolkenstein et al. (2011) using the Reading the Mind in the Eye Test (Baron-Cohen et al., 2001). IP However, a separate analysis of stimulus valence in this perspective-taking task revealed that CR depressed patients showed a higher accuracy rate in decoding negative mental states compared to healthy controls (Wolkenstein et al., 2011), indicating that MDD-patients show a hypervigilance US to negative emotional stimuli (Joormann and Gotlib, 2010). Such negative processing bias has AN been observed in a variety of cognitive and emotional processes (beyond social cognition), e.g., a specific attentional and memory bias towards negative cues in MDD patients (Leppänen, 2006; M Sterzer et al., 2011; Kluczniok et al., 2016; Hoffmann et al., 2016a) and could be expected to occur also in emotional empathy. ED When studying empathy in depression, it may be critical to consider that MDD patients may PT exhibit deficits in executive functioning and that these deficits may impact on empathic abilities (Thoma et al. 2011). Empathy requires inhibition of one‟s own emotional state (Hoffmann et al., CE 2016b) and flexible shifting between one‟s own and other people‟s perspectives. Potentially deficient inhibitory control in depression might exert a modulating effect on empathy (Singer AC and Lamm, 2009) irrespective of alexithymia and should therefore be taken into account.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 6 1.2. Effects of alexithymia on empathy One reason for the lack of clear evidence for empathy deficits in depression might be that comorbid alexithymia has so far not been accounted for. Alexithymia is a multifaceted T personality construct, which describes the impaired ability to identify and describe one‟s own IP emotional state (Sifneos, 1973). It is a common concurrent personality trait in depression CR (Honkalampi et al., 2000, Taylor and Bagby, 2004) and associated with deficits in emotional and cognitive empathy: Participants with high alexithymia scored significantly lower on empathic US concern and perspective taking (Moriguchi et al., 2007, Silani et al., 2008). Moreover, they reported more personal distress (Moriguchi et al., 2007). Individuals with Autism Spectrum AN Disorder (ASD) are known to exhibit high levels of alexithymia (Frith, 2004; Hill et al. 2004; Berthoz and Hill, 2005). Accumulating evidence seems to suggest that empathy deficits in ASD M can be explained by comorbid alexithymia rather than the diagnosis of ASD itself. (Bird et al., ED 2010; Bird and Cook, 2013) . Likewise, empathy deficits observed in depression could thus be related to concurrent alexithymia. Recent evidence suggests that in the case of emotional PT empathy, MDD patients might indeed be unimpaired, when no alexithymia is present and emotional states of self and other are not incongruent (Hoffmann et al., 2016b). CE In addition, depression and alexithymia effects on empathy may also interact. Thus, in addition AC to test for differential deficits in the affective and cognitive components of empathy, the second aim of the present study was to identify interacting and dissociable effects of depression and alexithymia on both components of trait and state empathy.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 7 1.3. The present study To address the above mentioned issues we used two well-established instruments for trait (Interpersonal Reactivity Index (IRI); Davis, 1980; 1983) and state empathy (Multifaceted T Empathy Test (MET); Dziobek et al., 2008) in four study groups: Patients suffering from an IP acute major depressive episode with high (n=18) and low (n=11) alexithymia and healthy CR controls with high (n=28) and low (n=13) alexithymia. As alexithymia has been linked to emotional and cognitive empathy deficits, we firstly hypothesized that alexithymia irrespective US of depression should decrease emotional and cognitive state and trait empathy. Secondly, when alexithymia is accounted for we expected MDD patients to instead show heightened emotional AN and cognitive empathy to negative emotions, consistent with previously reported mood- congruent processing biases (Leppänen, 2006, Sterzer et al., 2011, Kluczniok et al., 2016). M ED 2. Methods PT 2.1. Participants Twenty-nine patients suffering from a moderate to severe major depressive episode with high CE (n=18) and low (n=11) alexithymia were recruited through the inpatient clinic of the Charité - AC Universitätsmedizin Berlin, or were referred to us by specialized clinicians. Twenty-eight matched (with regard to education, age, and gender) healthy control (HC) participants with no history of psychiatric or neurological disorders were recruited through public notices and from project databases of the Charité - Universitätsmedizin Berlin. Participants completed the Toronto Alexithymia Scale (Bagby et al., 1994). High alexithymia was defined based on a large (N =  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 8 1859) representative German population sample (Franz et al., 2008), using the 66th percentile as a cut-off (Parker et al., 1993). The 66th percentile equaled the TAS-20-sum score of 53 for men and 52 for women and was subsequently used as a cut-off to represent high alexithymia (equal to and above 53/52). For the other participants (below 53/52), we subsequently used the term low alexithymia, indicating that their TAS-20-sum scores were relatively lower in the current sample. T To assure equal proportions of individuals with low and high alexithymia for both samples we IP further recruited healthy controls with high alexithymia from project databases of the Free CR University Berlin. This way, another 13 healthy controls, matched for education and gender, were included in the final sample (Table 1). By including these participants with high US alexithymia, a group difference in age emerged. Subsequently, age was used as a covariate in all AN analyses. Diagnosis of acute depressive episode (DSM-IV; American Psychiatric Association, 2000) was M confirmed with no other primary diagnoses (SCID-I; Wittchen et al., 1997). In addition, we ED applied the German version of the SCID-II screening questionnaire (Fydrich et al., 1997): No participant fulfilled the criteria for personality disorder. All participants were assessed with the PT Hamilton Depression Rating Sale (HAMD-17, Hamilton, 1960). Only patients scoring 15 or more points at the HAMD-17 were included. CE Every patient except one was medicated. Seven patients were treated with a selective serotonin AC reuptake inhibitor (SSRI), four with Mirtazapine, two with Venlafaxine, one with Buproprion, one with Doxepine, one with Quetiapine, three with a SSRI augmented with Lithium, one with Tianeptine augmented with Lithium, one with Mirtazapine augmented with Lithium, one with Sertraline augmented with Aripiprazole and one with Venlafaxine augmented with Olanzapine, one with a combination of Escitalopram and Trimipramine, one with Sertraline and Agomelatine,  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 9 one with Quetiapine and Lithium, one with Escitalopram and Mirtazapine and one with Venlafaxine and Amitriptyline. All patients were free of benzodiazepines for at least 48 hours at the time of the study. None of the patients had a history of electroconvulsive therapy (ECT). The study was approved by the local research ethics committee (Charité - Universitätsmedizin Berlin) and written informed consent was obtained from all participants. T IP CR Table 1. Demographic and clinical characteristics of the participants Healthy Controls US MDD-patients Significant AN effects (p<0.05) Low High Low High M alexithymia alexithymia alexithymia alexithymia Sample size 28 13 11 18 Gender 9 males 8 males 2 males 8 males Age 42.8 (12.7) 60.5 (19.0) 47.6 (12.7) 39.4 (11.7) d ED Education (yrs.) 15.8 (3.0) 16.7 (2.7) 16.4 (3.9) 14.3 (3.0) HAMD-17 0.7 (1.5) 0.8 (1.4) 19.0 (2.2) 21.3 (3.8) d, a TAS-20 38.9 (7.6) 59.3 (5.8) 44.0 (4.6) 61.0 (6.6) a d=significant depression group main effect; a=significant alexithymia group main effect PT CE 2.2 Measures AC 2.2.1. State empathy: MET The Multifaceted Empathy Test (MET) (for a detailed description of a related version see Dziobek et al., 2008) is a well-established measure of state empathy. In this task, a series of photographs (20 with a positive and 20 with a negative emotional valence) show people in emotionally charged situations, e.g., a crying child or a member of a rock band ecstatically  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 10 playing drums. Each photograph is presented in two conditions. One condition is referred to as cognitive empathy (i.e.: emotion recognition), the other condition as emotional empathy (i.e.: empathic concern). Cognitive empathy is assessed by asking participants to choose out of four adjectives the one that describes best the emotional state of the person presented in each photograph. Each correct answer scores one point. Emotional empathy is tested by asking the T participants to rate how strongly they feel for the person presented in the picture on a bipolar IP nine-point rating scale ranging from “not at all” to “very strongly”. Blocks of 10 subsequently CR presented pictures always involve the same question type (“How does the person feel?”, “How strongly do you feel for the person?”), but the order of the pictures within the conditions is US pseudorandomized. The completion of the MET takes about 20 minutes on average. For viewing AN and responding to the stimuli there were no time constraints. The number of correct responses computed separately for negative and positive valence was used as the dependant variable for M cognitive empathy. The summed rating scores also computed separately for positive and negative emotional valences, served as the dependent variable for emotional empathy. ED 2.2.2. Trait empathy: IRI PT Dispositional empathy was assessed using the Interpersonal Reactivity Index (IRI; Davis, 1980; 1983), differentiating 4 components of empathy (empathic concern, personal distress, CE perspective taking, fantasy), each referred to by 7 items. The empathic concern subscale AC measures the tendency to feel compassion and concern to unfortunate others. The personal distress scale reflects the degree to which the participant feels uncomfortable when he is confronted with the suffering of others, that is the negative side of experiencing too much empathy, referring to feeling with and not for someone (Batson, 2009). Fantasy refers to the ability to imaginatively transpose oneself into fictitious situations, e.g., when watching a movie  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 11 or reading a book. Whereas empathic concern and personal distress refer to emotional responses towards the feelings of others, the subscale perspective taking refers to a more cognitive aspect of empathy. The ability to adopt another person‟s point of view in everyday life is assessed by this scale. Participants rate their agreement on a five-point scale. The summed scores per scale represent the dependent variables for this task. T IP CR 2.3. Inhibitory control Inhibitory control was assessed using a Go/NoGo task, to control for differences in executive US functions. In this task, a blue square represented a Go stimulus and a red square a NoGo stimulus (intertrial intervals randomly altered between 950 ms and 1500 ms). Ninety Go trials and 60 AN NoGo trials were randomly presented. Participants had to respond as quickly as possible with a button-press to the presentation of the Go stimuli, while withholding a response to the M presentation of the NoGo stimuli. Response inhibition was measured as the ability to inhibit the ED response to NoGo stimuli. PT 2.4. Behavioral analysis CE To study state empathy (MET), we performed two 222 ANCOVAs on the variables emotional empathy and cognitive empathy using depression (depression vs. healthy controls) and AC alexithymia (high vs. low alexithymia) as between-subjects factors and valence (positive vs. negative valence) as within-subject factor. To investigate trait empathy (IRI), we performed four 22 ANCOVAs on the emotional empathy subscales (empathic concern, personal distress, fantasy) and the cognitive empathy subscale (perspective taking) using depression (depression vs.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 12 healthy controls) and alexithymia with the factors depression (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors. As age differed significantly between the groups, we included age as a covariate in all ANCOVAs. Pearson correlations were computed between the empathy measures (IRI subscales and MET T scores for emotional and cognitive empathy) and HAMD-17 symptom scores. IP To analyze the inhibitory control task, a d-prime score was calculated as a measure of response CR sensitivity (d' = Z(hit rate) - Z(incorrect presses during NoGo)). An ANCOVA was performed on the d-prime scores with depression and alexithymia as between-subjects factors as well as age as a covariate. US AN Data analysis was performed using the IBM SPSS statistics software, version 22.0. The significance level was set to p<0.05. All participants completed all assessments and there were M no missing values. ED 3. Results PT 3.1. State empathy (MET) CE 3.1.1 State emotional empathy To study state emotional empathy (MET) we performed a 222 ANCOVA, using depression AC (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors and valence (positive vs. negative) as within-subject factor. We found no significant effect of alexithymia (F(1,65) = 2.649; p = 0.108) and depression (F(1,65) = 1.700; p = 0.197) on the emotional empathy score of the MET. Moreover, we found no effect of valence (F(1,65) = 1.422; p  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 13 = 0.237). There was also no significant interaction between alexithymia and depression (F(1,65) = 0.135; p = 0.715), no valence by alexithymia interaction (F(1,65) = 0.274; p = 0.603) and no valence by alexithymia by depression interaction (F(1,65) = 2.608; p = 0.111). However, we found a significant valence by depression interaction (F(1,65) = 12.076; p = 0.001; 2 = 0.16, 90% CI: 0.04; 0.29). T IP To explore this valence by depression interaction, we performed separate 22 ANCOVAs using CR depression (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors in separate analyses of the positive and negative picture conditions. US This analysis revealed a significant effect of depression in the negative (F(1,65) = 7.0; p = 0.010; 2 = 0.1, 90% CI: 0.01; 0.21), but not in the positive (F (1,65)= 0.307; p = 0.581) picture condition AN (Fig. 1, A): Patients with MDD - independently of alexithymia - scored higher on emotional empathy (compared to healthy controls) when confronted with pictures depicting negative M emotions. We found no specific main effect of alexithymia in the positively (F(1,65) = 3.226; p = ED 0.077) nor in the negatively valenced condition (F (1,65)= 1.234; p = 0.271). PT 3.1.2 State cognitive empathy CE To investigate state cognitive empathy (MET) we performed a 222 ANCOVA, using AC depression (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors and valence (positive vs. negative) as within-subject factor. We found a main effect of alexithymia (F(1,65) = 4.105; p = 0.047; 2 = 0.06, CI 90%: 0.000; 0.17) on cognitive empathy (i.e., emotion recognition) with lower emotion recognition scores in participants with high alexithymia (Fig. 1, B). However, we found no main effect of depression  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 14 (F(1,65) = 0.000; p = 0.990): MDD-patients did not perform worse than participants without depression. There was also no significant effect of valence (F(1,65) = 0.001; p = 0.976) nor any interaction effect (all Fs < 0.607, ps > 0.439). T IP Please insert Figure 1 around here CR US Effects of depression and alexithymia on cognitive and emotional empathy. AN A) State emotional empathy: Displayed is emotional empathy (mean +SD) in the Multifaceted Empathy Test (MET; negative valence) for healtyh controls (HC) and patients with a major depressive episode (MDE) with high and low alexithymia, respectively. B) State cognitive empathy: Displayed is emotion recognition (mean+SD) in the MET. C) M Trait emotional empathy: Displayed is personal distress (mean + SD) in the Interpersonal Reactivity Index (IRI). D) Trait cognitive empathy: Shown is perspective taking in the IRI (mean+SD). ED PT 3.2. Trait empathy (IRI) CE 3.2.1 Trait emotional empathy AC To study trait emotional empathy we performed three 22 ANCOVAs on the emotional empathy subscales of the IRI (fantasy, empathic concern, personal distress), using depression (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors. We found no significant effect of alexithymia on fantasy (F(1,63) = 1.042; p = 0.311) and empathic  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 15 concern (F(1,63) = 3.334; p = 0.073). Moreover, there were no significant effects of depression on fantasy (F(1,63) = 0.267; p = 0.607) and empathic concern (F(1,63) = 0.816; p = 0.370). However, we found a main effect of alexithymia on personal distress (F(1,63)= 5.015; p = 0.029; 2 = 0.07, 90% CI: 0.004; 0.19), with higher personal distress in the high alexithymia group. Furthermore, we found a main effect of depression on personal distress (F(1,63)= 15.93; p = 0.001; 2 = 0.2, T IP 90% CI: 0.07; 0.33), with higher personal distress in the depression group. There was also a significant interaction between alexithymia and depression (F(1,63)= 4.149; p = 0.046; 2 = 0.06, CR CI: 0.00; 0.17), with a stronger effect of alexithymia in the depression group (T (27)= 2.6; p = US 0.015) compared to the healthy control group (T(37)= -0.6, p = 0.518) (Fig. 1, C). We found no further depression by alexithymia interactions on fantasy (F(1,63)= 0.079; p = 0.780) and empathic AN concern (F(1,63)= 0.589; p = 0.446). M 3.2.2 Trait cognitive empathy ED To study trait cognitive empathy, we performed a 22 ANCOVA on the cognitive empathy PT subscale of the IRI (perspective taking) using depression (depression vs. healthy controls) and alexithymia (high vs. low alexithymia) as between-subjects factors. We found a main effect of CE alexithymia on perspective taking (F(1,63) = 12.00; p = 0.001; 2 = 0.16, 90% CI: 0.04; 0.29), with lower scores in participants with high alexithymia (Fig. 1, D). However, we found no main effect AC of depression (F(1,63) = 1.769; p = 0.185) and no interaction between alexithymia and depression (F(1,63) = 1.405; p = 0.240) on perspective taking.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 16 3.3 Correlations to symptom severity To correlate empathy measures to clinical symptom severity, separate Pearson correlations were carried out for both healthy controls and MDD patients. Specifically, correlation analyses were T computed between empathy measures (IRI subscales and MET scores for emotional and IP cognitive empathy) and symptom severity as measured using the Hamilton Depression Rating CR Scale (HAMD-17). Within the depression group, we found a positive correlation between personal distress and HAMD-17 symptom scores (r(27) = 0.415, p = 0.025). No other significant correlations were found. US AN 3.4. Inhibitory control M We found no significant main effects of depression (F1,201 = 0.784, p = 0.379) and alexithymia ED (F1,201 = 1.064, p = 0.306) on inhibitory control, and no significant depression by alexithymia interaction (F1,201 = 1,034, p = 0.313). PT CE AC 4. Discussion To investigate whether inconclusive findings of empathy in depression may be related to interactions between alexithymia and different components of empathy, we studied two well- established instruments for testing trait (IRI; Davis, 1980; 1983) and state empathy (MET;  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 17 Dziobek et al., 2008) within four groups: Patients with Major Depressive Episode and healthy controls with high and low alexithymia, respectively. We hypothesized that alexithymia irrespective of depression should decrease emotional and cognitive state and trait empathy. Secondly, we expected MDD patients to show heightened T emotional and cognitive empathy to negative emotions, irrespective of alexithymia. IP Firstly, we found a main effect of alexithymia, irrespective of depression, on trait and state CR aspects of cognitive empathy. High alexithymia subjects showed lower scores in perspective taking (IRI) and in the cognitive-empathy-component of the MET. Secondly, we found main US effects of depression and alexithymia on trait emotional empathy (IRI-subscale personal distress). Moreover, we found a significant interaction effect of alexithymia and depression on AN trait emotional empathy. MDD-patients showed particularly high personal distress when affected by alexithymia (IRI). Thirdly, irrespective of alexithymia, we found a main effect of depression M on state emotional empathy (MET) in the negative emotion condition. MDD-patients showed ED higher scores on emotional empathy concerning negatively valenced stimuli. These findings suggest that concurrent alexithymia, irrespective of depression, affects trait and state cognitive as PT well as trait emotional empathy and has to be taken into account when assessing empathy deficits in depression. The only sub-component of empathy deficiently affected by depression, CE irrespective of alexithymia, is trait emotional empathy (IRI-subscale personal distress). AC We will discuss these findings in more detail below, starting with the effects of alexithymia and depression on cognitive empathy (MET and IRI), followed by the discussion of the effects on emotional empathy (MET and IRI).  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 18 4.1. Effects of alexithymia and depression on cognitive empathy As hypothesized, we found decreased levels of trait and state cognitive empathy in participants T with high alexithymia: Participants with high alexithymia showed difficulties in perspective IP taking on a trait empathy subscale (IRI) and emotion recognition in a state-based empathy task CR (MET). These findings are in line with earlier studies in high alexithymic participants (Moriguchi et al., 2007) reporting lower scores in perspective taking as measured with the IRI US and at the same time - like alexithymic participants in our study - higher levels of personal distress. This finding points to the inverse relationship between personal distress and perspective AN taking: Both subscales have been shown to be negatively associated (Davis, 1983; Moriguchi et al. 2007). Our study adds to the literature in that we were able to demonstrate the impairments in M alexithymia also at the state level, using the MET, while earlier studies have relied on the IRI ED capturing trait empathy. Moreover, we extended the findings by Davis (1983) and Moriguchi et al. (2007) concerning the inverse relationship between personal distress and perspective taking PT to depressed patients with high alexithymia. CE However, other than hypothesized we did not find a significant effect of depression on cognitive empathy, neither at the state nor at the trait level. In this context, it is noteworthy to mention, that AC cognitive empathy, as conceptualized in the present study, includes processes such as emotion recognition and affect sharing. Recent studies have shown a dissociation in healthy adults as well as psychopathological populations between two routes of social cognition, on the one hand an affective-motivational route including social emotions such as emotion contagion, empathy and  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 19 empathic distress, and on the other hand a purely cognitive route referring to our ability to infer beliefs and cognitive thoughts of others, referred to as Theory of Mind or mentalizing (Singer, 2006; 2012; Walter, 2012; Bird and Viding, 2014; Kanske et al., 2015a, 2015b, 2016). Future studies may clarify whether observed findings of a lack of cognitive empathy in depression accompanied by a modulation of this empathy component by alexithymia may generalize to pure T Theory of Mind tasks necessitating cognitive inferences about false-believe inferences. IP According to such models, alexithymia should only affect the affective-motivational route (e.g., CR empathy) subserved by limbic and para-limbic structures such as insula and anterior cingulated cortex (ACC) (Bird et al., 2010; Silani et al., 2013), but not the Theory of Mind network US involving parietal-temporal and prefrontal structures (Singer, 2012). AN M 4.2. Effects of alexithymia and depression on emotional empathy ED As hypothesized we found increased personal distress in high alexithymic participants. PT Moreover, we found higher rates of personal distress in depressed patients: Compared to healthy controls both groups reported more discomfort when relating to other peoples suffering. CE Furthermore, we found a relationship between the level of personal distress and symptom severity in patients. This is in accordance with previous findings, showing alexithymia AC (Moriguchi et al., 2007; Grynberg et al., 2010) as well as depression (O‟Connor et al., 2002; Thoma et al., 2011; Schreiter et al., 2013; see however Cusi et al., 2011) related to heightened personal distress. In addition to previous studies, which have not been able to disentangle the different effects of depression and alexithymia on personal distress, the novelty of our findings  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 20 is that, irrespective of alexithymia, a specific effect of depression on personal distress remains. From a clinical perspective, different psychological processes may be underlying each of the observed effects: One could speculate that the effect of depression might be due to a self- oriented perspective characteristic for depressed patients (Flory et al., 2000; Mor et al., 2010). Confronted with other people‟s suffering this self-orientation may induce ruminations on what it T would be like to be affected oneself, resulting in the feeling of negative affect and discomfort. IP Another explanation for heightened personal distress might be the problem of depressed patients CR with self-other distinction. As it has been shown lately, depressed patients showed an increased altrocentric bias when judging their own emotional state (Hoffmann et al., 2016b). This influence US of the emotional state of the other on the judgment of one's own emotional experience was AN related to heightened personal distress suggesting problems of depressed patients in keeping self and other separate (Hoffmann et. al., 2016b). Moreover, known problems of depressed patients M considering emotion regulation (for a recent review see Joormann and Stanton, 2016) might contribute to personal distress as well. Likewise, subjects with alexithymia may experience ED personal distress in the presence of other people‟s suffering, because they cannot identify, differentiate, and, as a consequence, regulate the feelings evoked (Grynberg et al., 2010). PT In addition we also found a significant interaction between depression and alexithymia: the effect CE of alexithymia on personal distress was stronger in the depression compared to the control group. Thus, both depression and alexithymia independently contribute to increased personal AC distress and, in addition, these factors interact in their effect on personal distress. This interaction is of clinical relevance because of the high prevalence of alexithymia in MDD (Leweke et al., 2012). Both depression and alexithymia might show heightened personal distress because of known problems in emotion regulation (Johnstone and Walter, 2014; Joormann and  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 21 Stanton, 2016). Additional high alexithymia seems to lead to even higher personal distress in MDD patients, as diminished emotional awareness seems to further increase the regulatory problems. Other than hypothesized, we did not find decreased state or trait empathic concern within the T alexithymia group. Neither did we find a group difference concerning state or trait empathic IP concern between MDD patients and healthy controls: Patients did not report lower feelings of care and concern in response to someone else‟s emotional experience than healthy controls. This CR finding is in line with prior studies (Thoma et al., 2011; Schreiter et al., 2013; see however Cusi US et al., 2011). It is also in line with previous differentiation between empathic concern and empathic distress (Batson, 2009; Eisenberg, 2000) and recent differentiations between empathy AN and compassion on the level of behavior, brain and its plasticity (Singer and Lamm, 2009; Singer and Klimecki, 2014; Klimecki et al., 2014; Klimecki, 2015). According to these differentiations, M there are two possible quite different responses to the suffering of others. One has been termed ED empathy, defined as feeling with another person, which can turn into empathic distress. The other is a response of empathic concern or compassion, which is defined as feeling for someone. PT Whereas the former is associated with negative affect and underlying brain activation related to threat, the latter is rather associated with positive feelings of warmth and love and a strong CE motivation for the welfare of the other. While empathy training increases brain activation in networks associated to negative affect, succeeding compassion training in the same people AC activates a non-overlapping brain network known to underlie reward and affiliate feelings (Klimecki et al., 2014). In line with such a more nuanced view, our findings suggest that while compassion/empathic concern is spared in depression, the risk of having an empathic distress response when exposed with the suffering of others is not.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 22 While the observation that state emotional empathy (MET) including both positive and negative trials of MDD patients showed no group difference for empathic concern, our additional analyses treating negative and positive trials separately revealed increased state emotional empathy in the negative condition within the depression group: Depressed patients reported more empathic concern when confronted with pictures of humans in negatively charged situations. In contrast, T increased empathic concern was not found in positive trials. Our data in the depression group IP suggests that the mood-congruent processing bias, which has been demonstrated in a number of CR cognitive domains, e.g., memory and attention (Leppänen, 2006; Sterzer et al., 2011; Kluczniok et al., 2016), may also exist for emotional empathy. They further may suggest that due to the US predominantly negative processing in MDD, the empathic concern ratings may have been AN interpreted by the patients as empathy ratings in the case of negative stimuli. Further studies asking for negative and positive affect ratings after each picture trial may help disentangle the M underlying reasons for such an effect. ED Finally, we found no differences in inhibitory control between groups, suggesting that differences between depressed patients and healthy controls observed in the present study are not PT simply related to differences in executive functioning (c.f. Thoma et al., 2011). It is acknowledged that inhibitory control only relates to one aspect of executive functioning. Future CE studies of empathy in depression may consider further components of executive functioning (e.g. cognitive flexibility, working memory). AC 4.3. Clinical relevance These findings are of considerable clinical relevance, as MDD patients with comorbid alexithymia will have different treatment outcomes (Taylor and Bagby, 2004; Ogrodniczuk et al., 2005). Empathy deficits in MDD can be mostly attributed to concurrent alexithymia, whereas  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 23 alexithymia seems to affect empathy on all levels. Diminished emotional awareness appears to lead to deficient sharing of emotional states, but also to impaired perspective taking and attribution of emotional state. Thus therapeutic work in this subgroup of MDD patients might have to focus firstly on increasing rudimentary emotional awareness and emotion regulation. Clinically, the concept of alexithymia becomes increasingly important to consider in therapy, T which raises the question of the etiology of alexithymia in depression. Generally, alexithymia is IP considered a trait rather than a state characteristic of an individual. However, it is not clear CR whether alexithymia observed in acutely depressed patients does improve after remission of symptoms (Luminet et al., 2001; Taylor and Bagby, 2004). The cross-sectional nature of the US present study limits our ability to determine if alexithymia observed in our MDD patient group is AN transient or permanent. In the former case, alexithymia could be considered a symptom of acute depression; in the latter case, alexithymia could be considered a vulnerability factor for MDD M (Luminet et al, 2001; Leweke et al., 2012). Longitudinal studies are needed to explore if and how empathic responding and alexithymia change over the course of depressive illness. To further ED explore the relationship between alexithymia and depression is a promising field with clinical relevance. PT From a clinical perspective, one might consider introducing a screening for high alexithymic CE traits in patients with depression which could allow the identification of patients who would particularly benefit from metacognitive or compassion training: Bornemann & Singer (2017) AC have shown that nine months of contemplative training can improve emotional awareness (as measured in the TAS) especially in participants with initially high alexithymia. However, the long duration of nine months might make this kind of intervention more interesting for patients after remission and might particularly aim at relapse prevention. Nonetheless, it has been shown  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 24 that even short-term compassion training (one-day training consisting of 6 hours) increases positive mood and decreases depressive symptoms (Leiberg et al., 2011). This intervention might even be feasible for inpatients during a depressive episode. T 4.4. Limitations IP A point of note is the age bias within the healthy controls with high alexithymia. We do CR acknowledge that by additionally recruiting a second control group, i.e., participants with high alexithymia, from specific databases an age bias emerged. To consider the age differences, we US used age as a covariate in all analyses. Moreover, we do acknowledge that due to the exploratory nature of our study we did not correct for multiple testing, which might have resulted in Type I AN error accumulation. Furthermore, while the MET is highly suitable to dissociate different M components of empathy we are aware that real life social interactions are much more complex. There is a need for novel, more ecologically valid measures that, at the same time, better ED dissociate affective and cognitive components of empathy (Kanske et al., 2015b; Schilbach, 2016; Schilbach et al., 2013). Finally, in this study we focused on alexithymia to investigate PT empathy in depression. However, other personality traits might play a critical role for empathy CE deficits in individuals suffering from depression. It could be promising for future studies to assess the relationship between the Big-Five dimensions of personality (McCrae and Costa, AC 1989) and alexithymia and depression, respectively, in their impact on empathy measures. 4.5. Conclusion  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 25 In conclusion, this study aimed to disentangle the different effects of depression and alexithymia on different components of empathy. We explored empathy accounting for alexithymia and looking at cognitive and emotional empathy in state and trait. Alexithymia seems to affect empathy at nearly all levels. Irrespective of alexithymia, MDD patients show only deficient empathy regarding personal distress, which seems to be a core empathy deficits in MDD and T associated with symptom severity. Other empathy deficits observed in individuals with MDD IP (decreased levels of state and trait cognitive empathy) seem to be linked to concurrent CR alexithymia. Thus, impairments in alexithymia become increasingly important to consider in therapy. US AN Acknowledgements M The authors wish to thank all the participants of this study. The research was supported by a grant from the German Federal Ministry of Education and Research (BMBF-01KR1207C to ED Felix Bermpohl) and by the Max Plank Society. PT Declaration of interest CE None. AC  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 26 References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, fourth ed. (text-rev.) American Psychiatric Association, Washington DC. Bagby, R.M., Parker, J.D., Taylor, G.J., 1994. The twenty-item Toronto Alexithymia Scale—I. T Item selection and cross-validation of the factor structure, J. Psychosom. Res. 38, 23-32. IP Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., Plumb, I., 2001, The "Reading the Mind in CR the Eyes" test revised version: a study with normal adults, and adults with Asperger syndrome or high-functioning autism. J. Child Psychol. Psychiatry 42, 241-251. US Batson, C.D., 2009. These things called empathy: eight related but distinct phenomena, in: Decety, J., Ickes, W. (Eds.), The Social Neuroscience of Empathy. MIT Press, Cambridge, MA, AN pp. 3-15. Berthoz, S., Hill, E.L., 2005. The validity of using self-reports to assess emotion regulation M abilities in adults with autism spectrum disorder. Eur. Psychiat. 20, 291-298. ED Bird., G., Viding, E., 2014. The self to other model of empathy: Providing a new framework for understanding empathy impairments in psychopathy, autism, and alexithymia. Neurosci. PT Biobehav. Rev. 47, 520-532. Bird, G., Cook, R., 2013. Mixed emotions: the contribution of alexithymia to the emotional CE symptoms of autism. Transl. psychiatry 3, e285. Bird, G., Silani, G., Brindley, R., White, S., Frith, U., Singer, T., 2010. Empathic brain responses AC in insula are modulated by levels of alexithymia but not autism. Brain 133, 1515-1525. Bornemann, B., Singer, T., 2017. Taking time to feel our body: steady increases in heartbeat perception and decreases in alexithymia over 9 months of contemplative mental training. Psychophysiology 54, 469-482.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 27 Cusi, A.M., MacQueen, G.M., Spreng, R.N, McKinnon, M.C., 2011. Altered empathic responding in major depressive disorder: Relation to symptom severity, illness burden, and psychosocial outcome. Psychiatry Res. 188, 231-236. Davis, M.H., 1980. A multidimensional approach to individual differences in empathy. JSAS, Cat. Sel. Doc. Psychol. 10, 85-104. T Davis, M.H., 1983. Measuring individual differences in empathy - evidence for a IP multidimensional approach. J. Pers. Soc. Psychol. 44, 113-126. CR Davis, M.H., 1996. Empathy: a social psychological approach, Westview Press, Boulder, CO. De Vignemont, F., Singer, T., 2006. The empathic brain: how, when and why? Trends Cogn. Sci. 10, 435-441. US AN Dziobek, I., Rogers, K., Fleck, S., Bahnemann, M., Heekeren, H.R., Wolf, O.T. et al., 2008. Dissociation of cognitive and emotional empathy in adults with Asperger syndrome using the M Multifaceted Empathy Test (MET). J. Autism Dev. Disord. 38, 464-473. Eisenberg, N., 2000. Emotion, regulation, and moral development. Annu. Rev. Psychol. 51, 665- ED 697. Flory, J.D., Raikkonen, K., Matthews, K.A., Owens, J.F., 2000. Self-focused attention and mood PT during everyday social interactions. Pers. Soc. Psychol. Bull. 26, 875-883. CE Franz, M., Popp, K., Schaefer, R., Sitte, W., Schneider, C., Hardt, J. et al., 2008. Alexithymia in the German general population. Soc. Psychiatry Psychiatric Epidemiol. 43, 54-62. AC Frith, U., 2004. Emanuel Miller lecture: confusions and controversies about Asberger syndrome. J. Child Psychol. Psychiatry 45, 672-686. Fydrich, T., Renneberg, B., Schmitz, B., Wittchen, H.-U., 1997. SKID II. Strukturiertes Klinisches Interview für DSM-IV, Achse II: Persönlichkeitsstörungen. Interviewheft. Eine  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 28 deutschspeachige, erw. Bearb. d. amerikanischen Originalversion d. SKID-II von: MB First, RL Spitzer, M. Gibbon, JBW Williams, L. Benjamin, (Version 3/96). Hogrefe, Göttingen. Gonzalez-Liencres, C., Shamay-Tsoory, S.G., Brüne, M., 2013. Towards neuroscience of empathy: Ontogeny, phylogeny, brain mechanisms, context and psychopathology. Neurosci. Biobehav. Rev. 37, 1537-1548. T Grynberg, D., Luminet, O., Corneille, O., Grèzes, J., Berthoz, S., 2010. Alexithymia in the IP interpersonal domain: A general deficit of empathy? Pers. Inidvid. Dif. 49, 845-850. CR Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56-62. Hill, E., Berthoz, S., Frith, U., 2004. Brief report: cognitive processing of own emotions in US individuals with autistic spectrum disorder and in their relatives. J. Autism Dev. Disord. 34, 229- AN 235. Hirschfeld, R., Montgomery, S.A., Keller, M.B., Kasper, S., Schatzberg, A.F., Möller, H.-J. et M al., 2000. Social functioning in depression: a review. J. Clin. Psychiatry 61, 268-275. Hoffmann, F., Banzhaf, C., Kanske, P., Bermpohl, F., Singer, T., 2016a. Where the depressed ED mind wanders: Self-generated thougt patterns as assessed through experience sampling as a state marker of depression. J. Affect. Disord. 198, 127-134. PT Hoffmann, F., Banzhaf, C., Kanske, P., Gärtner, M., Bermpohl, F., Singer, T. (2016b). Empathy CE in depression: Egocentric and altercentric biases and the role of alexithymia. J. Affect. Disord. 199, 23-29. AC Honkalampi, K., Hintikka, J., Tanskanen, A., Lehtonen, J., Viinamäki, H., 2000. Depression is strongly associated with alexithymia in the general population. J. Psychosom. Res., 48, 99-104. Johnstone, T., Walter, H., 2014. The neural basis of emotion dysregulation, in: Gross, J. (Ed.), Handbook of Emotion Regulation, second ed. The Guilford Press, New York, pp. 58-75.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 29 Joormann, J., Stanton, C.H., 2016. Examining emotion regulation in depression: A review and future directions. Behav. Res. Ther. 86, 35-49. Joormann, J., Gotlib, I.H., 2010. Emotion regulation in depression: relation to cognitive inhibition. Cogn. Emot. 24, 281-298. Kanske, P., Böckler, A., Trautwein, F.M., Parianen Lesemann, F.H., Singer,T., 2016. Are strong T empathizers better mentalizers? Evidence for independence and interaction between the routes of IP social cognition. Soc. Cogn. Affect. Neurosci. 9, 1383-1392. CR Kanske, P., Böckler, A., Singer, T., 2015a. Models, mechanisms and moderators dissociating empathy and Theory of Mind, in: Wöhr, M., Krach, S. (Eds.), Current Topics in Behavioral US Neurosciences - Social Behavior from Rodents to Humans: Neural Foundations and Clinical AN Implications. Springer, Luxemburg, pp. 1-14. Kanske, P., Böckler, A., Trautwein, F.M., Singer, T., 2015b. Dissecting the social brain: M introducing the EmpaToM to reveal distinct neural networks and brain-behavior relations for empathy and theory of mind. Neuroimage 122, 6-19. ED Klimecki, O.M., Leiberg, S., Ricard, M., Singer, T., 2014. Differential pattern of functional brain plasticity after compassion and empathy training. Soc. Cogn. Affect. Neurosci. 9, 873-879. PT Klimecki, O.M., 2015. The plasticity of social emotions. Soc. Neurosci. 10:5, 466-473. CE Kluczniok, D., Hindi Attar, C., Fydrich, T., Fuehrer, D., Jaite, C., Domes, G., et al., 2016. Transgenerational effects of maternal depression on affect recognition in children. J. Affect. AC Disord. 189, 233-239. Lamm, C., Batson, C.D., Decety, J., 2007. The neural substrate of human empathy: effects of perspective-taking and cognitive appraisal. J. Cogn. Neurosci. 19, 42-58.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 30 Leiberg, S., Klimecki, O., Singer, T., 2011. Short-term compassion training increases prosocial behaviour in a newly developed prosocial game. PLoS ONE, 6(3), e17798. doi:10.1371/journal.pone.0017798. Leppänen, J.M., 2006. Emotional information processing in mood disorders: a review of T behavioral and neuroimaging findings. Curr. Opin. Psychiatry 19, 34-39. IP Leweke, F., Leichsenring, F., Kruse, J., Hermes, S., 2012. Is alexithymia associated with specific CR mental disorders? Psychopathology 45, 22-28. Luminet, O., Bagby, R.M., Taylor, G.J., 2001. An evaluation of the absolute and relative stability US of alexithymia in patients with major depression. Psychother. Psychosom. 70, 254-260. McCrae, R.R., Costa, P.T., 1989. More reasons to adopt the five-factor model. Am. Psychol. 44, AN 451-452. Mor, N., Doane, L.D., Adam, E.K., Mineka, S., Zinbarg, R.E., Griffith, J.W. et al., 2010. Within- M person variations in self-focused attention and negative affect in depression and anxiety: a diary ED study. Cogn. Emot. 24, 48-62. Moriguchi, Y., Decety, J., Ohnishi, T., Maeda, M., Mori, T., Nemoto et al., 2007. Empathy and PT judging other's pain: an fMRI study of alexithymia. Cereb. Cortex 17, 2223-2234. O'Connor, L.E., Berry, J.W., Weiss, J., Gilbert, P., 2002. Guilt, fear, submission, and empathy in CE depression. J. Affect. Disord. 71, 19-27. Ogrodniczuk, J.S., Piper, W.E., Joyce, A.S., 2005. The negative effect of alexithymia on the AC outcome of group therapy for complicated grief: what role might the therapist play? Compr. Psychiatry 46, 206-213. Parker, J.D., Taylor, G.J., Bagby, R.M., 1993. Alexithymia and the recognition of facial expressions of emotion. Psychother. Psychosom. 59, 197-202.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 31 Schreiter, S., Pijnenborg, G.H.M., aan het Rot, M., 2013. Empathy in adults with clinical or subclinical depressive symptoms. J. Affect. Disord. 150, 1-16. Schilbach, L., 2016. Towards a second-person neuropsychiatry. Phil. Trans. R. Soc. Lond. B 371: 20150081. http://dx.doi.org/10.1098/rstb.2015.0081. Schilbach, L., Timmermans, B., Vasudevi, R. Costall, A., Bente, G., Schlicht, T., et al., 2013. T Toward a second-person neuroscience. Behav. Brain Sci. 36(4), 393-414. doi: IP 10.1017/S0140525X12000660. CR Sifneos, P.E., 1973. The prevalence of „alexithymic‟characteristics in psychosomatic patients. Psychother. and Psychosom. 22, 255-262. US Silani, G., Bird, G., Brindley, R., Singer, T., Frith, C., Frith, U., 2008. Levels of emotional AN awareness and autism: an fMRI study. Soc. Neurosci. 3, 97-112. Silani, G., Lamm, C., Ruff, C. C., Singer, T., 2013. Right Supramarginal Gyrus Is Crucial to M Overcome Emotional Egocentricity Bias in Social Judgments. J. Neurosci. 33, 15466-15476. Singer, T., Klimecki, O.M., 2014. Empathy and Compassion. Curr. Biol. 18, R875-R878. ED Singer, T., 2012. The past, present and future of social neuroscience: A European perspective. Neuroimage 61, 437-449. PT Singer, T., Lamm, C., 2009. The social neuroscience of empathy. Ann. N. Y. Acad. Sci. 1156, CE 81-96. Singer, T., 2006. The neural basis and ontogeny of empathy and mind reading: Review of AC literature and implications for future research. Neurosci. Biobehav. Rev. 30, 855-863. Sterzer, P., Hilgenfeldt, T., Freudenberg, P., Bermpohl, F., Adli, M., 2011. Access of emotional information to visual awareness in patients with major depressive disorder. Psychol. Med. 41, 1615-1624.  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 32 Taylor, G.J., Bagby, R.M., 2004. New trends in alexithymia research. Psychother. Psychosom. 73, 68-77. Thoma, P., Zalewski, I., von Reventlow, H. G., Norra, C., Juckel, G., Daum, I. (2011). Cognitive and affective empathy in depression linked to executive control. Psychiatry Res. 189, 373-378. Walter, H., 2012. Social cognitive neuroscience of empathy. Concept, circuits and genes. Emot. T Rev. 4, 9-17. IP Wilbertz, G., Brakemeier, E.-L., Zobel, I., Härter, M., Schramm, E., 2010. Exploring CR preoperational features in chronic depression. J. Affect. Disord. 124, 262-269. Wittchen, H.-U., Wunderlich, U., Gruschwitz, S., Zaudig, M., 1997. SKID I. Strukturiertes US Klinisches Interview für DSM-IV. Achse I: Psychische Störungen. Interviewheft und AN Beurteilungsheft. Eine deutschsprachige, erweiterte Bearb. d. amerikanischen Originalversion des SKID I, Hogrefe, Göttingen. M Wolkenstein, L., Schönenberg, M., Schirm, E., Hautzinger, M., 2011. I can see what you feel, but I can't deal with it: Impaired theory of mind in depression. J. Affect. Disord. 132, 104-111. ED PT CE AC  ACCEPTED MANUSCRIPT C. Banzhaf et al. Interacting and dissociable effects of alexithymia and depression on empathy 33 Figure 1 T IP CR US AN M ED PT CE AC