Clinical Neurophysiology 126 (2015) 1331–1337

Contents lists available at ScienceDirect

Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph

Neural correlates of causal attribution in negative events of depressed patients: Evidence from an fMRI study Lei Hao a,b,1, Junyi Yang a,b,1, Yanqiu Wang a,b, Songyan Zhang a,b, Peng Xie c,d,e,⇑, Qinghua Luo c,d,e, Gaoping Ren c,d,e, Jiang Qiu a,b,⇑ a

Key Laboratory of Cognition and Personality (SWU), Ministry of Education, Chongqing, PR China School of Psychology, Southwest University, Chongqing 400715, PR China Department of Neurology, The First Affiliated Hospital of Chongqing Medical University, Chongqing 400016, PR China d Chongqing Key Laboratory of Neurobiology, Chongqing 400016, PR China e Institute of Neuroscience, Chongqing Medical University, Chongqing 400016, PR China b c

a r t i c l e

i n f o

Article history: Accepted 22 October 2014 Available online 29 October 2014 Keywords: Causal attribution Negative events Functional magnetic resonance imaging (fMRI) Inferior parietal lobe (IPL)

h i g h l i g h t s  Depressed patients showed balanced attribution style versus controls.  The inferior parietal lobe (IPL) showed a greater activation in patients versus controls for the non-self-

serving in negative events.  The duration of illness of patients was correlated with a BOLD signal change of the IPL.

a b s t r a c t Objective: The causal attribution of depressive patients in negative events was significantly important. However, few previous studies have explored its neural underpinnings. Methods: The current study examines the neural basis of causal attribution in negative events of depressed patients (12) compared with healthy controls (12) by using the Attribution Style Task (AST). Briefly, participants were presented with 80 sentences describing 40 positive and 40 negative social events. Participants were asked to imagine the event happening to them and select the most likely cause with a button press: 1 self (internal), 2 another person, and 3 situation (external). Results: Behaviorally, controls showed a self-serving bias, whereas patients demonstrated a balanced attributional pattern with the attribution scores. Our fMRI results found a significant group difference in the inferior frontal gyrus and middle temporal gyrus in depressed participants compared to normal controls. Moreover, there was a significantly increased activation in the IPL during non-self-serving attributions in negative events of patients compared to controls. Most interestingly, we also found the BOLD signal change of the region of IPL was positively related to the duration of the illness of the patients. Conclusion: Based on our findings, we may infer that a stronger activation of the IPL in depression may demonstrate that depressed patients always pay more attention to self-reference in negative events. Significance: These interesting findings might provide a biomarker of subtle differences in brain signal alterations associated with depressive cognitive characteristics. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction

⇑ Corresponding authors at: School of Psychology, Southwest University, Beibei, Chongqing 400715, PR China. Tel.: +86 23 6836 7942 (J. Qiu). Institute of Neuroscience, Chongqing Medical University, Yuzhong, Chongqing 400016, PR China. Tel.: +86 23 6848 5490 (P. Xie). E-mail addresses: [email protected] (P. Xie), [email protected] (J. Qiu). 1 These authors equally contributed to this paper.

Causal attribution is very important to explain and construct the word in our daily life. However, depression patients often exhibit an abnormal attribution style compared to healthy people. Indeed, depression has been conceptualized as a cognitive disorder (Beck, 1976, 1967). Depressed people (DP) make less internal attributions for positive events and external attribution for negative

http://dx.doi.org/10.1016/j.clinph.2014.10.146 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

1332

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337

events (Abramson et al., 1978), that is known as nonself-serving attributions (Diez-Alegría et al., 2006), which is one of the attributional-style abnormalities. Generally, healthy people usually display a so-called ‘‘self-serving’’ bias: the propensity to attribute success to the self and failure to external factors (Miller and Ross, 1975). Nevertheless, a number of theorists have further suggested that depressive symptoms might be profitably understood by taking into account the causal attributions for the bad events in their lives (Abramson and Sackheim, 1977; Abramson et al., 1978; Golin et al., 1981; Gong-Guy and Hammen, 1980; Harvey, 1981; Harvey et al., 1978). Extensive previous studies examining the neural circuitry of attribution style have shown an involvement of the frontal cortex and temporal lobe. Specifically, the activation of these regions have been known to be involved in self-control (Amodio and Frith, 2006; Elliott et al., 2000; Ochsner et al., 2004); this suggests that attributing events in a nonself-serving manner is due to a suppression process of the dominant self-enhancement tendencies. In addition, the ability to ‘‘mentalize’’ (understand others’ behavior in terms of their mental states) may depend upon the medial prefrontal cortex (MPFC) and the superior temporal sulcus, which represents states of the self and the actions of others (Frith and Frith, 1999; Gallagher and Frith, 2003). Moreover, Frith et al. (1999) have examined that the dorsal striatum, previously implicated in motivated behavior, mediates the self-serving bias by using fMRI during the performance of attributional decision tasks. Then, Seidel et al. (2010) conducted an fMRI study in normal people and found that the self-serving bias was associated with the activation in the dorsal anterior cingulate and the dorsal striatum, which might be related to the rewarding value of these attributions. Recently, Seidel et al. (2012) also made a neuroimaging research investigating the differences of DPs and controls, which revealed a significant group difference in a frontotemporal network. Interestingly, the higher activation of this network was associated with nonself-serving attributions in controls but self-serving attributions in patients, which might reflect that DPs experience a greater conflict with the emotional context of their own negative selfconcept when a self-serving judgment is made during causal attributions. Previous research has demonstrated that major depression is primarily a disorder of emotion and its regulation (Beevers et al., 2010; Gotlib and Joormann, 2010; Hamilton et al., 2011; Wang et al., 2012). Various neuroimaging studies have postulated prefrontal areas (primarily MPFC), anterior insular and rostral anterior cingulate cortex to regulate attention and cognitive features of depression (Diener et al., 2012; Fitzgerald et al., 2006). Furthermore, the default mode network (DMN) is frequently cited as a region of abnormal synchrony in major depression (Anand et al., 2005; Broyd et al., 2009). The DMN is one of the commonly recognized resting state networks in the brain, containing structures such as the MPFC, the precuneus, the posterior cingulate cortex (PCC), and the lateral, medial, and inferior parietal cortices. The DMN is believed to mediate attention-orienting and self-referential thought (Broyd et al., 2009). Thus, abnormal activation of the above regions in this network in DPs suggests that abnormalities in selfreferential thought and attentional shifts contribute to major depression (Wang et al., 2012). As Seidel et al. (2012) pointed out, a limitation of their study was that individual valence-attribution conditions were combined into self-serving and non-self-serving categories and an individual analysis of valence and attribution effects, particularly their interactions, was not possible. However, as we discussed above, a depressive attributional style constitutes a cognitive diathesis that interacts with negative life events to produce depression disorder. That is, it is important for researchers to examine the neural correlates in causal understanding of DPs to attributional style in negative life

events. Meanwhile, we used an fMRI approach to investigate whether the frontotemporal network was associated with causal attribution in DPs compared with healthy controls (HCs). 2. Materials and methods 2.1. Participants Twenty DPs were recruited from the local hospital and a group of 22 controls were recruited from the local community. Participants were between 18 and 60 years of age and group-matched based on age. However, our study focused on exploring the difference of external attribution or internal attribution in the negative events between two groups; thus, 12 qualified patients and 12 HCs were selected in the analysis. All participants were right handed and had normal or corrected-to-normal vision. All assessments were conducted by well-trained psychiatrists or postgraduate research assistants. The psychiatrists diagnosed participants as an unipolar depressed group according to the Structured Clinical Interview for the DSM-IV (SCID; First et al., 1995) criteria. Patients with a history of neurologic illness, alcohol or substance abuse, electroconvulsive therapy, or transcranial magnetic stimulation within the last 2 years were excluded from the current study. The DPs and HC subjects were both assessed by using the Hamilton Rating Scale for Depression (HAMD-21). However, HCs scored in the normal range. Meanwhile, exclusion criteria for DPs and HCs included severe head trauma, color blindness, learning disabilities, current panic disorder, psychotic symptoms, and alcohol or substance abuse within the past 6 months. This study was approved by the local ethics committee of Southwest China University. All participants provided written informed consent prior to the study, which was approved by the Institutional Human Participants Review Board of Southwest University Imaging Center for Brain Research. 2.2. Measures Demographic information, including self-reported age, gender, and education level, was collected via a questionnaire. Illness severity in DPs was assessed using the HAMD (Hamilton, 1960), a 21-item questionnaire. Meanwhile, we also assessed the attribution style scores of all participants by using the Attribution Style Questionnaire (ASQ) (Peterson et al., 1982). 2.3. Experiment task The Attribution Style Task (AST) was designed as our experiment task. First, we used the statements of the Internal, Personal, and Situational Attributions Questionnaire (IPSAQ) and extend the 32 statements from the IPSAQ to a total of 80 short sentences in the task paradigm, similar to reference of Seidel et al. (2012), including 40 positive and 40 negative social events. Then, participants were asked to complete the AST in the scanner. They were presented with 40 positive (e.g., ‘‘A friend sent you a gift.’’) and 40 negative (e.g., ‘‘A friend ignored you.’’) social events. Subjects were asked to read the social events silently and then to vividly imagine the events happening to them and decide the most likely cause with a button press: 1 self (internal), 2 another person, and 3 situations (external). The subjects’ correct use of the button box to indicate their attributional decisions was checked before each experiment. The fixation cross was randomly presented 2s/4s/6s, followed by these positive or negative events, together with the response categories (internal, another person, and situation) were maximally presented 6s until a response was given. In addition, the IPSAQ was similar to the ASQ (Peterson et al., 1982); the ASQ has been criticized for its low internal reliability.

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337

2.4. Behavioral data analysis Behavioral statistical analyses were performed using SPSS 16.0 (Statistical Packages for the Social Sciences, Version 16.0, SPSS Inc., USA) with a level of significance of p < 0.05. The task contained four conditions: positive valence, external attribution (positive external (PE)), positive valence, internal attribution (positive internal (PI)), negative valence, external attribution (negative external (NE)), and negative valence, internal attribution (negative internal (NI)). We performed 2 (group)  2 (attribution bias) repeated measures ANOVAs on the attributional response. The attribution bias specifically represented select positive events as internal and negative events as external (PI and NE values), select positive events as external, and negative events as internal (NI and PE values). In addition, we examined the means of PI + NE and NI + PE between two groups with SPSS. The attributional response represented the mean number of attribution decisions for each of the four conditions during the attribution-style fMRI task, the ASQ (positive and negative events) scores and on the reaction time (RT). Planned post hoc group comparisons regarding attribution biases, ASQ scores, and RT were performed by between-group independent ttests comparing both groups as well as paired-samples t-tests comparing percentages of self-serving and nonself-serving attributions and comparing the attribution scores in positive events and negative events. Within-group correlation analyses between attributional data and HAMD and ASQ were performed using the Pearson product moment correlation coefficient. 2.5. Image acquisition and analysis Images were acquired with a Siemens 3T scanner (Siemens Magnetom Trio TIM, Erlangen, Germany). An echo-planar imaging (EPI) sequence was used for data collection, and 432 T2⁄-weighted images were recorded per run (TR = 2000 ms; TE = 30 ms; flip angle = 90°; field of view (FOV) = 220  220 mm2; matrix size = 64  64; 32 interleaved 3-mm thick slices; in-plane resolution = 3.4  3.4 mm2; interslice skip = 0.99 mm). T1-weighted images were recorded with a total of 176 slices at a thickness of 1 mm and in-plane resolution of 0.98  0.98 mm2 (TR = 1900 ms; TE = 2.52 ms; flip angle = 9°; FOV = 250  250 mm2). We used SPM8 (Wellcome Department of Cognitive Neurology, London, UK, http://www.fil.ion.ucl.ac.uk/spm/software/spm8/) to preprocess the functional images (Friston et al., 1994). Slice timing correction was used to correct slice order, the data were realigned to estimate and modify the six parameters of head movement, and the first six images were discarded to achieve magnet-steady images. These images were then normalized to Montreal Neurological Institute (MNI) space in 3  3  3 mm3 voxel sizes. The normalized data were spatially smoothed with a Gaussian kernel; the full width at half maximum (FWHM) was specified as 8  8  8 mm3. After preprocessing, the four regressors from each run (i.e., NI, NE, PI, and PE) were modeled to create the design matrix, and for each subject all the four runs were modeled in one general linear model (GLM). We have chosen the onset of the stimulus (sentence and response categories) as the onset time point and the RT as the duration of the event. They were convolved with the canonical hemodynamic response function, and the six realignment parameters for each subject were also included as confounding factors. 2.6. Statistical image analysis The image threshold for fMRI data significance was first set to p < 0.005, one tailed in the individual voxel level, uncorrected. Then we performed AFNI’s AlphaSim program (http://afni.nimh. nih.gov/pub/dist/doc/manual/AlphaSim.pdf) for multiple comparisons. We ran 1000 Monte Carlo simulations with the correct value

1333

p < 0.05, two tailed, a Gaussian filter width of 8 mm, and Cluster connection radius of 5 mm, and previous studies have demonstrated that 1000 simulations were adequate in the fMRI study (Brass et al., 2007; Chen et al., 2013). The correction at p < 0.05 for multiple comparisons revealed a cluster size of 50 contiguous voxels. We used this corrected AlphaSim threshold to report our fMRI data below. The coordinates of activation images were transformed to the MNI space (1.5 mm isotropic voxel). Moreover, as in the previous research, the subject-level contrast map was entered into group-level and two-sample t-tests (Seidel et al., 2012). The main contrast of interest compared nonself-serving versus self-serving attributions. This contrast was composed of: (NI + PE) versus (NE + PI), and is equivalent to a within-subject valence-attribution interaction effect. We evaluated this bidirectional contrast on a within-group and between-group basis. For completeness, valence (positive vs. negative) contrast was compared. Based on previous studies (Abramson et al., 1977, 1978; Harvey et al., 1978; Gong-Guy et al., 1980), depression attribution in negative events plays a key role in DPs. To examine the different attribution bias in negative life events, we separately examine the effects of each condition for two groups. According to previous studies (Beck, 1976, 1967), DPs make less external attributions for negative events. We would explore the different attribution in the negative events between two groups. For superposition of functional brain activity on the whole brain, we chose the superposition times of the trials P15 in each condition (Buckner et al., 1996; Wager et al., 2004). The number of the qualified patients was 12. Accordingly, we also randomly chose 12 HCs from the previous sample whose superposition times of the trials P15 in each condition. Meanwhile, we have found that there were no significant differences between the 12 controls and other 10 control participants in age, years of education, and the HAMD scores. That is, the two control subgroups exhibited homogeneity. Thus, our contrasts were composed of: patient NI versus control NI, patient NE versus control NE. 3. Results 3.1. Participant characteristics The characteristics of demographic variables of the study sample are shown in Table 1. There were no significant differences between the DPs and control participants in age and years of education. As expected, DPs had significantly higher scores on the HAMD (Hamilton, 1960) than did control participants (t = 9.85, p < 0.05). At the time of assessment, their depression illness had lasted an average of 31.6 months. The characteristics of demographic variables of the subgroup are shown in Table 1. 3.2. Behavioral data The repeated measures ANOVA of attribution scores (ASQ) owed significantly group  attribution bias interaction (F = 23.12, p < 0.05). In addition, we performed planned post hoc comparisons to examine group differences regarding attributional biases. It suggested that there were significantly (p = 0.003) more self-serving attributions in HC (mean = 3.77, SD = 0.65) compared to the patients (mean = 3.23, SD = 0.50). A direct comparison within each group in the positive scores versus negative scores attribution style revealed that controls showed a more self-serving style (t = 6.95, p < 0.05), whereas patients showed a balanced attribution style (t = 0.92, p < 0.05). These effects are illustrated in Fig. 1. The repeated measures ANOVA on the RT showed no significant main effect of group indicating both groups responded equally fast. Moreover, there remained no significant differences in all interactions with the factor group (group by attribution).

1334

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337

Table 1 Clinical and demographic characteristics of patients with depression and matched controls. Variable

DP (SD) (n = 20)

HC (SD) (n = 22)

Subsample of DP (SD) (n = 12)

Subsample of HC (SD) (n = 12)

Age (years) Female Education (years) HAMD score Duration of illness

32.3 (8.4) 6 12.4 (3.2) 20.3 (7.8) 32.6 months

29.7 (5.7) 9 14.2 (2.9) 2.2 (2.5) –

32.8 (8.7) 4 13.0 (3.3) 19.3 (8.0) 25.1 months

29.7 (4.4) 6 14.4 (2.8) 2.4 (3.3) –

SD = standard deviation; HAMD = Hamilton Rating Scale for Depression.

The repeated measures ANOVA of attributional response showed no significant group  attribution bias interaction (p > 0.05). 3.3. Imaging data 3.3.1. Group attribution analysis When we compared the group effects of the self-serving and nonself-serving attributions, the contrast revealed a significant activation in the whole-brain analysis. Both groups showed similar patterns of greater activation with the self-serving attribution in temporal lobe and frontal cortex respect to nonself-serving attribution. Patients showed additional activations in posterior midline regions (middle cingulate gyrus and postcentral gyrus) in the non-self-serving attribution. Imaging analysis of group differences for the non-self-serving attribution (NI + PE) contrast was found in the right temporal lobe. The region of the right middle temporal gyrus showed a greater activation in the DPs for the nonself-serving versus controls (p < 0.05, corrected: peak voxel coordinate (x y z), 45–60 6; peak Z score = 4.33; Table 2 and Fig. 2A). Meanwhile, analysis of group differences for the self-serving (NE + PI) contrast was found in the frontal lobe. The right inferior frontal lobe showed a greater activation in the DPs for the self-serving versus controls (p < 0.05, corrected: peak voxel coordinate (x y z), 48 39 3; peak Z score = 5.08; Fig. 2B). The significant threshold was set at p < 0.05 (combination voxel threshold at p < 0.005 and cluster size more than 100 voxels) using AlphaSim correction. 3.3.2. Depression causal attribution analysis in negative events Analysis of group differences was conducted between the two subgroups in the negative condition showed differential activations in the self-serving attribution and non-self-serving

Fig. 1. The behavioral result of attribution scores (ASQ) displays a biased causal attribution in 22 healthy controls (HCs) and balanced causal attribution in 20 depressed patients (DPs).

attribution. The region of the IPL showed a greater activation in the subgroup of DPs versus HCs for the non-self-serving attribution (p < 0.05, corrected: peak voxel coordinate (x y z), 57–36 48; peak Z score = 4.54; Fig. 3). The region of the inferior frontal gyrus (IFG) showed a greater activation in the subgroup of depressed versus controls for the self-serving attribution (p < 0.05, corrected: peak voxel coordinate (x y z), 51 42 3; peak Z score = 4.39). In addition, the BOLD signal change of the IPL and the episode duration of depression reveal a significantly positive correlation (r = 0.58, p < 0.05) regarding the ‘‘non-self-serving attribution’’ contrasts (see Fig. 3). 4. Discussion Despite the previous findings in self-serving or non-self-serving attribution of DPs, the neural correlates in negative events remain sparsely investigated. Therefore, in the current study, we presented positive and negative social scenarios to a sample of DPs and asked them to make causal attributions while measuring their brain activity with fMRI. Similar to previous study, contrast of group differences for the self-serving and non-self-serving attribution were specifically conducted in the IFG and middle temporal gyrus in DPs compared to normal controls (Seidel et al., 2012). Most importantly, we found that the region of the IPL showed a greater activation in the subgroup of depressed versus control for the non-selfserving attribution in negative events, and the BOLD signal change of the region of the IPL was positively related to the duration of the episode of the patients.

4.1. Group difference in causal attribution With respect to the behavioral results of attribution scores (ASQ), causal attributions in controls were biased in a self-serving manner. DPs showed a more balanced attribution pattern (Moritz et al., 2007). This group difference showed that DPs might not have the self-protection motivation associated with self-serving tendencies in controls (Butler and Mathews, 1983; Seidel et al., 2012). On fMRI analysis, both groups showed similar patterns of greater activation with the self-serving attribution in temporal lobe and frontal cortex than non-self-serving attribution, similar to previous findings (Seidel et al., 2010, 2012). The group difference in activation of the right middle temporal gyrus and the IFG was specifically driven by non-self-serving attribution and self-serving attribution in DPs compared to controls, which might be related to self-related process or introspective self-referential processing and conflict monitoring or cognitive control. In particular, the temporal gyrus is quite consistently implicated in inference of representational mental states such as beliefs or internal states (Saxe and Powell, 2006; Sommer et al., 2007). Moreover, the IFG might be involved in conflict monitoring and cognitive control (Amodio et al., 2006; Etkin et al., 2006). As Seidel et al. (2012) demonstrated, ‘‘the activation of fronto-temporal network during causal attributions due to the emotional context of their own negative self-concept, resulting in greater conflict when a self-serving judgment is made.’’

1335

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337

4.2. Depressive non-self-serving attribution in negative events An interesting result of our study was that the region of the IPL showed a greater activation in the depressed versus control for the non-self-serving attribution. Generally, the IPL including the supramarginal gyrus and the angular gyrus might be concerned with detection of salient new events and sustaining attention on task goals (Clark et al., 2000; Kiehl et al., 2005; Williams et al., 2007), which indicate that DPs easily direct more attention on negative events and sustain this state. Depressed subjects showed more activation than did controls in the inferior parietal lobule for sad emotional stimuli, given the role of the inferior parietal lobule in attentional processing of emotional stimuli (Canli et al., 2004; Davidson et al., 1999), suggesting that DPs display enhanced responsiveness to negative stimuli, whereas control subjects may engage in an emotion-regulation strategy that minimizes the processing of such input. In addition, mindfulness training resulted in an increased engagement of inferior parietal lobule in trained participants about momentary experience self-reference, suggesting that this region is involved in switching between first- and third-person perspectives (Ruby and Decety, 2004). Moreover, meditation practice is associated with cortical thickening in the inferior parietal lobule cortices (Lazar et al., 2005), suggesting these regions may be altered with extended daily focused attention

to moment-to-moment experience, and thus may represent the neural underpinnings of self-reference in the psychological present. Thus, a stronger activation of the IPL in depression may demonstrate that DPs always pay more attention on self-reference in negative events. An early study revealed that bilateral IPL was more activated by beliefs, but equally activated by fixation (Zaitchik et al., 2010). This finding raises the possibility that belief attribution is possibly similar to the resting state, when the brain is not engaged in any directed task. The resting state represents that a subject is not performing an explicit task within functional brain imaging (Biswal, 2012). Previous resting-state functional connectivity research has revealed a number of networks represent specific patterns of synchronous activity (Cole et al., 2010; Rosazza and Minati, 2011). A previous study has demonstrated that the DMN is a group of areas in the human brain characterized by functions of a self-referential nature (Sheline et al., 2009). Meanwhile, the DMN is involved in the evaluation of potentially survival-salient information from the body and the world: beliefs, perspective taking of the desires (Buckner et al., 2008). Thus, bilateral IPL activation in the depressed versus control represent that the DP is easily involved in self-related thought in negative events. According to Buckner and Carroll (2007), this undirected mode of processing, sometimes referred to as the default mode (Raichle

Table 2 Results of between group analysis of attributional bias, including MNI coordinates and z-values of the peak as well as cluster size (k). Groups

Cluster

MNI x

y

z-Value

k

z

DP (n = 20) vs. HC (n = 22)

Self-serving DP vs. HC Non-self-serving DP vs. HC

R. inferior frontal lobe R. middle temporal gyrus

48 45

39 60

3 6

5.08 4.33

128 162

Subsample DP (n = 12) vs. HC (n = 12)

Self-serving DP vs. HC Non-self-serving DP vs. HC

R. inferior frontal gyrus R. inferior parietal lobe

51 57

42 36

3 48

4.39 4.54

111 163

DP = Depressed patients; HC = healthy controls.

Fig. 2. fMRI results: Illustration of the group difference regarding non-self-serving versus self-serving attributions. Panel A shows increased activation in the middle temporal gyrus relates to non-self-serving attributions in 20 depressed patients versus 22 healthy controls. Panel B shows increased activation in inferior frontal gyrus relates to selfserving attributions in depressed patients versus healthy controls.

1336

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337

Fig. 3. Illustration of subgroup difference regarding nonself-serving versus self-serving attributions. The left panel shows an increased activation in IPL relates to nonselfserving in negative events between 12 depressed patients and 12 healthy controls. The right panel shows that the BOLD signal change of the IPL was positively related to the duration of the episode of the patients.

et al., 2001), involves a highly stereotypic pattern of activity that characterizes a state when people passively think about themselves and are relatively inattentive to their immediate surroundings. In addition, we found that the duration of depression illness was correlated with the activation of the IPL. This may demonstrate that the duration of the illness of the patients was correlated the abnormal function of the region. Therefore, as discussed above, the activation of IPL might be associated with sustained self-refection on negative events or direct attention to passive stimulates.

Acknowledgments This research was supported by the National Natural Science Foundation of China (31271087), the Program for New Century Excellent Talents in University (2011) by the Ministry of Education, the Program for the Top Young Talents by Chongqing, the Fundamental Research Funds for the Central Universities (SWU1209101), China and Chongqing Postdoctoral Science Foundation funded project (2012M510098; XM2012006). Conflict of interest: None of the authors have potential conflicts of interest to be disclosed.

4.3. Depression self-serving attribution in negative events References Furthermore, we also found that DPs revealed a greater activation on the IFG during self-serving attribution in negative events than controls, which is consistent with our above results in DPs (N = 20) versus control (N = 22) participants. As we discussed before, the IFG was involved in conflict monitoring or cognitive control (Amodio et al., 2006; Etkin et al., 2006). In addition, previous studies also demonstrated that the IFG was correlated with self-reflection or self-concept (D’Argembeau et al., 2005; Fossati et al., 2003). Therefore, DPs show activation in the IFG during causal attributions due to the emotional context of their own negative self-concept, resulting in a greater conflict when a self-serving judgment is made. 5. Conclusion In the current study, we contrast the group of DPs and controls in non-self-serving or self-serving attribution. Group differences in causal attribution were specifically conducted in the IFG and middle temporal gyrus in depressive participants compared to normal controls. Moreover, our study using fMRI revealed that region of the IPL showed a greater activation in the group of depressed versus control for the non-self-serving attribution in negative events. Furthermore, interestingly, the duration of the episode of the patients was correlated with the BOLD signal change of this region. These interesting findings might provide a biomarker for subtle differences in brain signal alterations associated with depressive cognitive characteristics.

Abramson LY, Sackheim HA. A paradox in depression: uncontrollability and selfblame. Psychol Bull 1977;84:838–51. Abramson LY, Seligman ME, Teasdale JD. Learned helplessness in humans: critique and reformulation. J Abnorm Psychol 1978;87:49–74. Amodio DM, Frith CD. Meeting of minds: the medial frontal cortex and social cognition. Nat Rev Neurosci 2006;7:268–77. Anand A, Li Y, Wang Y. Activity and connectivity of brain mood regulating circuit in depression: a functional magnetic resonance study. Biol Psychiat 2005;57:1079–88. Beck AT. Cognitive therapy and the emotional disorders, New York: International Universities Press; 1976. Beck AT. Depression: clinical, experimental, and theoretical aspects. New York: Harper & Row; 1967. Beevers CG, Clasen P, Stice E. Depression symptoms and cognitive control of emotion cues: a functional magnetic resonance imaging study. Neuroscience 2010;167:97–103. Biswal BB. Resting state fMRI: a personal history. Neuroimage 2012;62:938–44. Brass M, Schmitt RM, Spengler S. Investigating action understanding: inferential processes versus action simulation. Curr Biol 2007;17:2117–21. Broyd SJ, Demanuele C, Debener S. Default-mode brain dysfunction in mental disorders: a systematic review. Neurosci Biobehav Rev 2009;33:279–96. Buckner RL, Andrews-Hanna JR, Schacter DL. The brain’s default network. Ann N Y Acad Sci 2008;1124:1–38. Buckner RL, Bandettini PA, O’Craven KM. Detection of cortical activation during averaged single trials of a cognitive task using functional magnetic resonance imaging. Proc Nat Acad Sci 1996;93:14878–83. Buckner RL, Carroll DC. Self-projection and the brain. Trends Cogn Sci 2007;11:49–57. Butler G, Mathews A. Cognitive processes in anxiety. Adv Behav Res Ther 1983;5:51–62. Canli T, Sivers H, Thomason ME. Brain activation to emotional words in depressed vs healthy subjects. Neuroreport 2004;15:2585–8. Chen Z, Lei X, Ding C. The neural mechanisms of semantic and response conflicts: an fMRI study of practice-related effects in the Stroop task. Neuroimage 2013;66:577–84.

L. Hao et al. / Clinical Neurophysiology 126 (2015) 1331–1337 Clark VP, Fannon S, Lai S. Responses to rare visual target and distractor stimuli using event-related fMRI. J Neurophysiol 2000;83:3133–9. Cole DM, Smith SM, Beckmann CF. Advances and pitfalls in the analysis and interpretation of resting-state fMRI data. Front Syst Neurosci 2010;4:8. D’Argembeau A, Collette F, Van der Linden M. Self-referential reflective activity and its relationship with rest: a PET study. Neuroimage 2005;25:616–24. Davidson RJ, Abercrombie H, Nitschke JB. Regional brain function, emotion and disorders of emotion. Curr Opin Neurobiol 1999;9:228–34. Diener C, Kuehner C, Brusniak W. A meta-analysis of neurofunctional imaging studies of emotion and cognition in major depression. Neuroimage 2012;61:677–85. Diez-Alegría C, Vázquez C, Nieto-Moreno M. Personalizing and externalizing biases in deluded and depressed patients: are attributional biases a stable and specific characteristic of delusions? Br J Clin Psychol 2006;45:531–44. Elliott R, Dolan RJ, Frith CD. Dissociable functions in the medial and lateral orbitofrontal cortex: evidence from human neuroimaging studies. Cereb Cortex 2000;10:308–17. Etkin A, Egner T, Peraza DM. Resolving emotional conflict: a role for the rostral anterior cingulate cortex in modulating activity in the amygdala. Neuron 2006;51:871–82. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for Axis-I DSM-IV Disorders-Patient Edition (SCID-I/P). In: Biometrics research department, New York: Psychiatric Institute; 1995. Fitzgerald DA, Angstadt M, Jelsone LM. Beyond threat: amygdala reactivity across multiple expressions of facial affects. Neuroimage 2006;30:1441–8. Fossati P, Hevenor SJ, Graham SJ. In search of the emotional self: an fMRI study using positive and negative emotional words. Am J Psychiatry 2003;160:1938–45. Friston KJ, Holmes AP, Worsley KJ. Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Mapp 1994;2:189–210. Frith CD, Frith U. Interacting minds – a biological basis. Science 1999;286:1692–5. Gallagher HL, Frith CD. Functional imaging of ‘theory of mind’. Trends Cogn Sci 2003;7:77–83. Golin S, Sweeney PD, Shaeffer DE. The causality of causal attributions in depression: a cross-lagged panel correlational analysis. J Abnorm Psychol 1981;90:14–22. Gong-Guy E, Hammen C. Causal perceptions of stressful events in depressed and nondepressed outpatients. J Abnorm Psychol 1980;89:662–9. Gotlib IH, Joormann J. Cognition and depression: current status and future directions. Annu Rev Clin Psychol 2010;6:285–312. Hamilton JP, Furman DJ, Gotlib IH. The neural foundations of major depression: classical approaches and new frontiers. In: Lopez-Munoz FF, Alamo C, editors. Neurobiology of depression. Boca Raton, FL: Taylor & Francis Group; 2011. p. 57–73. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56–62.

1337

Harvey DM. Depression and attributional style: interpretations of important personal events. J Abnorm Psychol 1981;90:134–42. Harvey JH, Ickes WE, Kidd RF. New directions in attribution research: II, Hillsdale. New Jersey: Lawrence Erlbaum Associates; 1978. Kiehl KA, Stevens MC, Laurens KR. An adaptive reflexive processing model of neurocognitive function: supporting evidence from a large scale (n = 100) fMRI study of an auditory oddball task. Neuroimage 2005;25:899–915. Lazar SW, Kerr CE, Wasserman RH. Meditation experience is associated with increased cortical thickness. Neuroreport 2005;16:1893–7. Miller DT, Ross M. Self-serving biases in the attribution of causality: fact or fiction? Psychol Bull 1975;82:213–25. Moritz S, Woodward TS, Burlon M. Attributional style in schizophrenia: evidence for a decreased sense of self-causation in currently paranoid patients. Cogn Ther Res 2007;31:371–83. Ochsner KN, Knierim K, Ludlow DH. Reflecting upon feelings: an fMRI study of neural systems supporting the attribution of emotion to self and other. J Cogn Neurosci 2004;16:1746–72. Peterson C, Semmel A, von Baeyer C. The attributional style questionnaire. Cogn Ther Res 1982;6:287–99. Raichle ME, MacLeod AM, Snyder AZ. A default mode of brain function. Proc Nat Acad Sci 2001;98:676–82. Rosazza C, Minati L. Resting-state brain networks: literature review and clinical applications. Neurol Sci 2011;32:773–85. Ruby P, Decety J. How would you feel versus how do you think she would feel? A neuroimaging study of perspective-taking with social emotions. J Cogn Neurosci 2004;16:988–99. Saxe R, Powell LJ. It’s the thought that counts specific brain regions for one component of theory of mind. Psychol Sci 2006;17:692–9. Seidel E-M, Eickhoff SB, Kellermann T. Who is to blame? Neural correlates of causal attribution in social situations. Soc Neurosci 2010;5:335–50. Seidel E-M, Satterthwaite TD, Eickhoff SB. Neural correlates of depressive realism – An fMRI study on causal attribution in depression. J Affect Disord 2012;138: 268–76. Sheline YI, Barch DM, Price JL. The default mode network and self-referential processes in depression. Proc Nat Acad Sci 2009;106:1942–7. Sommer M, Döhnel K, Sodian B. Neural correlates of true and false belief reasoning. Neuroimage 2007;35:1378–84. Wager TD, Rilling JK, Smith EE. Placebo-induced changes in fMRI in the anticipation and experience of pain. Science 2004;303:1162–7. Wang L, Hermens DF, Lagopoulos J. A systematic review of resting-state functionalMRI studies in major depression. J Affect Disord 2012;142:6–12. Williams LM, Felmingham K, Kemp AH. Mapping frontal-limbic correlates of orienting to change detection. Neuroreport 2007;18:197–202. Zaitchik D, Walker C, Miller S. Mental state attribution and the temporoparietal junction: an fMRI study comparing belief, emotion, and perception. Neuropsychologia 2010;48:2528–36.