Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Imbalance of default mode and regulatory networks during externally focused processing in depression

Imbalance of default mode and regulatory networks during externally focused processing in depression [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 doi:10.1093/scan/nsu117 SCAN (2015) 10, 744 ^751 Imbalance of default mode and regulatory networks during externally focused processing in depression Emily L. Belleau, Lauren E. Taubitz, and Christine L. Larson Department of Psychology, University of Wisconsin-Milwaukee, 2441 East Hartford Avenue, Milwaukee, WI 53211, USA Attentional control difficulties likely underlie rumination, a core cognitive vulnerability in major depressive disorder (MDD). Abnormalities in the default mode, executive and salience networks are implicated in both rumination and attentional control difficulties in MDD. In the current study, individuals with MDD (n¼ 16) and healthy controls (n¼ 16) completed tasks designed to elicit self-focused (ruminative) and externally-focused thinking during fMRI scanning. The MDD group showed greater default mode network connectivity and less executive and salience network connectivity during the external- focus condition. Contrary to our predictions, there were no differences in connectivity between the groups during the self-focus condition. Thus, it appears that when directed to engage in self-referential thinking, both depressed and non-depressed individuals similarly recruit networks supporting this process. In contrast, when instructed to engage in non-self-focused thought, non-depressed individuals show a pattern of network connectivity indicative of minimized self-referential processing, whereas depressed individuals fail to reallocate neural resources in a manner consistent with effective down regulation of self-focused thought. This is consistent with difficulties in regulating self-focused thinking in order to engage in more goal-directed behavior that is seen in individuals with MDD. Keywords: depression; rumination; default mode network; executive network; salience network A core cognitive feature of major depressive disorder (MDD) is the information processing in both those with MDD and healthy controls tendency to ruminatethe persistent and passive focus on one’s (Nejad et al., 2013). Individuals with MDD exhibit stronger functional negative mood and the adverse consequences of being in a negative connections between the subgenual anterior cingulate cortex mood (Nolen-Hoeksema, 1991). Rumination is related to a host of (subACC) and other areas within the default mode network unfavorable outcomes including more severe and prolonged depressive (Greicius et al., 2007; Berman et al., 2011). Moreover, greater resting symptoms (Nolen-Hoeksema and Morrow, 1991; Nolen-Hoeksema state functional connectivity between the subACC and posterior et al., 1993; Just and Alloy, 1997), a poorer response to treatment cingulate cortex has been tied to greater trait rumination in individuals (Jones et al., 2008) and an increased risk for relapse after remission with MDD and healthy controls (Berman et al., 2011). However, when of a depressive episode (Michalak et al., 2011). Additionally, there is directly comparing depressed and non-depressed individuals during a evidence showing that rumination serves as a risk factor for the devel- task that engages self-focus, depressed individuals showed greater ac- opment of MDD (Just and Alloy, 1997; Nolen-Hoeksema, 2000). tivation in a number of default mode network structures compared Attentional control difficulties, another central trait in MDD (e.g. with non-depressed individuals (Cooney et al., 2010). Gohier et al., 2009; De Lissnyder et al., 2012), may underlay ruminative On the other hand, when a task is cognitively demanding or requires thinking by undermining the ability to disengage from negative self- focus on the external environment, the default mode network normally relevant material (Koster et al., 2011). A recent study demonstrated deactivates (Vanhaudenhuyse et al., 2010). An inability to deactivate the that rumination partially mediated the relationship between atten- default mode network during such tasks has been positively associated tional control difficulties and future depressive symptoms among with greater attentional control difficulties among healthy volunteers MDD patients (Demeyer et al., 2012). (Weissman et al., 2006; Mason et al., 2007; Christoff et al., 2009). One The neural underpinnings of both rumination and attentional study demonstrated that compared with healthy controls, depressed dyscontrol in MDD are thought to implicate three neural networks individuals were unable to deactivate the default mode network when (see Hamilton et al., 2013): the default mode network (Raichle et al., instructed to disengage from negative stimuli (Sheline et al.,2009). 2001), the executive network (Seeley et al., 2007) and the salience MDD participants have also shown difficulties deactivating an inferior network (Seeley et al., 2007). The default mode network is composed portion of the anterior medial prefrontal cortex, a central hub of the of a number of midline regions including the medial prefrontal cortex, default mode network, during a task designed to evoke external focus. the anterior cingulate cortex, the precuneus, the posterior cingulate This impaired deactivation was also associated with self-reported ru- cortex, the retrosplenial cortex, lateralized areas of the parietal cortex mination in the MDD group (Johnson et al., 2009). (Raichle and Snyder, 2007), as well as the medial temporal cortex and In addition to the default mode network, altered coordinated activ- hippocampal formation (Buckner et al., 2008). The default mode net- ity within the executive network, which encompasses areas in the work is thought to subserve a number of internally focused attentional lateral prefrontal cortex and the posterior parietal cortex (Seeley processes (Vanhaudenhuyse et al., 2010) including self-referential et al., 2007; Dosenbach et al., 2008), has been implicated in MDD (Zhou et al., 2010; Wei et al., 2013). Activation in this network is anticorrelated with the default mode network (Fox et al., 2005) and Received 22 March 2014; Revised 19 July 2014; Accepted 1 September 2014 is essential for paying attention to external environmental demands Advance Access publication 1 October 2014 The authors gratefully acknowledge the adults that took the time to participate in the study. The authors also (Vanhaudenhuyse et al., 2010) as well as engaging in goal-directed want to thank Michael Duke and Kim Lewis for help with participant recruitment as well as Judi Zaferos-Pylant and or cognitively demanding activities (Dosenbach et al., 2008). Yu Liu for their role in acquiring the fMRI data. The work was supported by the University of Wisconsin-Milwaukee Deficient activations within executive network regions have been Research Growth Initiative (C.L.L.) and the National Institute of Mental Health K01 (MH086809 to C.L.L.). linked with greater trait rumination amongst healthy individuals Correspondence should be addressed to Christine L. Larson, Department of Psychology, University of Wisconsin Milwaukee, 2441 East Hartford Ave., Milwaukee, WI, 53211, USA. E-mail: larsoncl@uwm.edu (Kuhn et al., 2012), as well as poor attentional control in MDD The Author (2014). Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 745 (Fales et al., 2008; Wang et al., 2008) and dysphoric participants without MDD on levels of neural activity in some default mode, sali- (Beevers et al., 2010). Recently, it was shown that MDD individuals ence and executive network structures (Cooney et al., 2010). We exam- recruited executive network regions to a lesser extent when instructed ined group connectivity differences by separately correlating central to inhibit both ruminative and emotionally neutral thoughts (Carew nodes of each network shown to be aberrant in MDD with other et al., 2013). However, some studies have found elevated activity in regions contained within that specific network. While there have areas within the executive network when instructed to engage in self- been some conflicting findings, based on the weight of the extant evi- focused thinking (e.g. Lemogne et al., 2009; Cooney et al., 2010), which dence, we expected that during the self-focused thought induction, the may potentially be a compensatory mechanism for counteracting cog- MDD group would show greater connectivity within the default mode nitive vulnerabilities in MDD (e.g. Harvey et al., 2005). network and decreased connectivity within the executive network Another network that is associated with aberrant coordinated activ- compared to those without MDD. Within the salience network, we ity in individuals with MDD compared with healthy controls is the specifically predicted decreased connectivity between the RFIC and salience network (Wei et al., 2013; Manoliu et al., 2014), which is dACC in participants with MDD compared with healthy controls, composed of the anterior insula/frontal operculum, dorsal anterior given the role these structures play in attentional control abilities cingulate cortex/paracingulate cortex and the superior temporal pole (Eckert et al., 2009; Shenhav et al., 2013). (Seeley et al., 2007; Dosenbach et al., 2008). Activation in the salience network is thought to be associated with monitoring and drawing METHODS attention to stimuli that are subjectively relevant (Seeley et al., 2007). Participants Additionally, the salience network plays a role in switching mental sets, Eighteen individuals diagnosed with MDD and 18 healthy controls with damage to the anterior insula being associated with poor switch- (HC) were administered the Structured Clinical Interview for the ing ability (Hodgson et al., 2007). In regards to switching behavior, it DSM-IV (First et al., 1995). MDD participants had to be free of any has been shown that the right frontal insular cortex (RFIC) and dorsal history of bipolar disorder, schizophrenia or psychosis and current anterior cingulate cortex (dACC) play a role in neural network switch- alcohol or substance abuse. Anxiety disorder comorbidity and the ing, initiating deactivation of the default mode and activation in the use of psychotropic medication were permitted in the MDD group. executive network (Sridharan et al., 2008). Evidence suggests that The HC group was free of any current Axis I psychopathology or individuals with MDD have reduced connectivity between the RFIC lifetime history of a mood disorder. Two participants from the with other regions in the salience network (Manoliu et al., 2014). MDD group and two participants from the HC group were excluded Specifically, compared with healthy individuals, MDD participants from further analysis due to excessive head motion in the scanner, show deficient recruitment of salience network structures when resulting in a final sample of 16 in the HC group (8 males, mean directed to inhibit ruminative and emotionally neutral thoughts age¼ 31.13; s.d.¼ 10.70) and 16 in the MDD group (5 males, mean (Carew et al., 2013). Furthermore, individuals with MDD show an age¼ 33.19, s.d.¼ 11.49; mean duration of current MDD abnormal relationship between activation in the RFIC and the onset episode¼ 21.63 months, s.d.¼ 49.49). of increased activation in the executive network versus the default Additionally, ten MDD participants reported that they were experi- mode network. Individuals with MDD show a rise in RFIC activation encing their first episode or have had one prior episode, three indi- at the onset of increased executive network activation, whereas healthy viduals reported experiencing three to six prior episodes, and another individuals show greater RFIC activation during a rise in default mode three individuals reported experiencing 10 or more prior major network activation (Hamilton et al., 2011). Given this evidence, the depressive episodes. Five MDD participants were currently taking psy- salience network likely plays a role in attentional dyscontrol and ru- chotropic medication (n¼ 2 duloxetine hydrochloride/Cymbalta, mination in MDD. n¼ 1 bupropion hydrochloride/Wellbutrin XL, n¼ 1 bupropion Thus, existing research has shown a potential link between activity hydrochloride and fluoxetine hydrochloride/Prozac, n¼ 1 risperi- in these three networks with rumination and the attentional difficulties done/Risperdal). Additionally, eleven MDD participants were diag- that may underlie rumination in MDD. However, these studies contain nosed with comorbid anxiety disorders (n¼ 2 social phobia, n¼ 2 limitations. While resting state functional connectivity studies have social and specific phobias, n¼ 2 generalized anxiety disorder, n¼ 1 paved the way for understanding of aberrant intrinsic connectivity in generalized anxiety disorder and social phobia, n¼ 1 generalized these core networks in MDD, the lack of constraints placed on behav- anxiety disorder and panic disorder, n¼ 1 posttraumatic stress dis- ior makes it difficult to clarify what role these networks are playing in order, n¼ 1 specific phobia, n¼ 1 anxiety disorder not otherwise spe- the cognitive difficulties in MDD. On the other hand, while task-based cified). At the scanning session, participants completed the Beck designs have allowed examination of aberrant brain activity when Depression Inventory-II (BDI-II; Beck et al., 1979) and the rumination individuals with MDD are instructed to engage in self-referential pro- subscale of the Response Styles Questionnaire (Nolen-Hoeksema, cessing (e.g. Cooney et al., 2010), researchers have typically not focused 1991). The study was approved by the University of Wisconsin- on how brain regions work together to contribute to cognitive Milwaukee Institutional Review Board, and participants provided writ- problems in MDD. It has been proposed that characterizing functional ten informed consent according to the Declaration of Helsinki. connections amongst neural regions may be a useful approach for Participants were paid for completion of the interview and scanning understanding the cognitive, behavioral and emotional difficulties in session. depression (Hamilton et al., 2013). The current study aimed to examine functional connections within Self-focused and externally focused thought induction tasks the default mode, executive and salience networks among individuals with and without MDD during a task in which participants were While in the scanner, participants completed an 8-min self-focus explicitly instructed to think about topics designed to elicit self-focused thought induction task and an 8-min external-focus thought induction and externally-focused thinking. Previous studies have found that task (adapted from Lyubomirsky and Nolen-Hoeksema, 1993). The compared with the external-focus condition, the self-focus condition self-focus thought induction statements were designed to generate used in this paradigm effectively generates rumination in depressed more abstract, internally focused, open-ended conceptual mental rep- individuals (Lyubomirsky and Nolen-Hoeksema, 1993; Nolen- resentations. On the other hand, the external-focus statements were Hoeksema and Morrow, 1993) and differentiates those with and meant to evoke more concrete, outwardly-focused, and stereotypical, [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 746 SCAN (2015) E. L. Belleau et al. visually dependent mental representations of external objects or scenes. (Hamilton et al., 2012). Lastly, the RFIC is a fundamental region in In each of the thought induction tasks, participants were visually the salience network that has shown to be abnormal in depression and presented with up to 45 statements. The self-focused statements were related to attentional control and rumination (see Slitz and Hayley, neither inherently negative nor positive, but open to individual inter- 2012 for a review). Pearson’s r maps were generated by correlating pretation. An example of a statement from the self-focused thought each seed with other regions belonging to each of their respective induction was ‘Think about why you react the way you do’ and ‘Think neural networks. The time series of the subACC seed was correlated about sitting down and analyzing your personality.’ The externally with every other voxel within the default mode network and the RFIC focused thought induction included more visually dependent seed was correlated with every other voxel in the salience network. statements such as ‘Think about and imagine a boat slowly crossing Similarly, the right DLPFC time series was correlated with regions the Atlantic.’ For a full list of the thought induction statements, please within the right executive network and the left DLPFC time series see the Supplementary Materials. At the beginning of each thought was correlated with regions in the left executive network. The analyses induction task, participants were told to read each statement slowly were restricted to each of the respective network regions by applying and silently to themselves. Participants were asked to use their imagin- previous established spatial templates representing each of the three ation and concentration to focus on each of the ideas being presented. networks (Franco et al., 2009; Laird et al., 2011). The r statistics were Participants were instructed to proceed with each statement at their then normalized using a Fischer’s z transformation and then resampled own pace. In between each statement, subjects were presented with a to 1-mm voxels. This normalized data were used to calculate all group- one second fixation cross. The order of the two thought induction level statistics. tasks were counterbalanced within each of the groups. To determine whether network connectivity differed as a function of group and task, a Group (MDD, HC) Thought Induction (self- focused, externally-focused) ANOVA was calculated separately for Imaging parameters each of the three networks. Participant was included as a random Whole brain imaging was conducted using a 3-T short bore Signa factor in each model. To correct for multiple comparisons, cluster Excite system. Functional images were collected using a T2* weighted thresholding using Monte Carlo simulations was applied to each of gradient-echo, echoplanar pulse sequence (TR¼2s; TE¼ 25 ms; the three separate neural network templates. Using a voxel-based FOV¼ 24 cm, flip angle¼ 848; voxel size 3.75 3.75 4 mm, sagittal P < 0.01 and nearest-neighbor selection criteria, accounting for spatial orientation). The self-focused and externally-focused thought correlation, clusters greater than 570 mm for the default mode network induction tasks consisted of 232 whole brain scans. High resolution and the salience network, and 567 mm for the right and left executive spoiled gradient recalled (SPGR) images (1 mm sagittal slices; network achieved a corrected P < 0.05. Because our hypotheses were TR¼ 9.6 s; TE¼ 3.9 ms; field of view¼ 24 cm; flip angle¼ 128; voxel centered on Group Thought Induction interaction influences on size¼ 0.9375 0.9375 1) were collected and served as an anatomical functional connectivity within the three networks, we focused on map for the functional images. these findings in the results section below. For a listing of Group and Thought Induction main effects, see Table 1. Additionally, given that Imaging preprocessing there are gender differences in the tendency to ruminate (Johnson and Reconstruction and image processing was conducted in AFNI (Cox, Whisman, 2013) and that the depressed sample did not have the same 1996). The first seven images for each scan were excluded to allow gender proportions as the control sample, we also conducted secondary for magnetic saturation. The remaining images were despiked, analyses examining whether our Group Thought Induction volume registered to correct for motion, spatially normalized to interaction results were influenced by gender. We also tested whether Talairach space, and spatially smoothed by convolution with an medication effects influenced our ANOVA findings, as we had five isotropic Gaussian kernel (full-width at half-maximum¼ 6 mm) to individuals with MDD on psychotropic medication, including an decrease high spatial frequency noise. For subsequent analyses, the individual taking an antipsychotic medication (not standard for the following nuisance regressors were included: baseline, signal drift, treatment of depression). These results are in the Supplementary estimates from a rigid-body motion correction that was done in Material section. three translational and three rotational directions, and censoring of Finally, we also correlated Significant Group Thought Induction high motion images (cutoff 0.3 mm). Additionally, a band pass filter interaction findings with the Beck Depression Inventory-II (BDI-II; was applied to attenuate frequencies above 0.1 Hz and below 0.01 Hz. Beck et al., 1979) and the rumination subscale of the Response Styles Subjects who exhibited excessive head motion (greater than an average Questionnaire (RSQ-Rumination; Nolen-Hoeksema, 1991). These value of 2.5 mm translation and/or 2.58 rotation) were excluded from results are also presented in the Supplementary Figures S1–S3. further analysis. RESULTS Seed-based functional connectivity analyses Participant characteristics The time series was extracted from four seeds, including a seed placed The two groups did not differ significantly with regard to age (t¼ .53, in the subgenual anterior cingulate cortex (subACC; Cullen et al., 2009; df¼ 30, P¼ 0.60), sex distribution ( ¼ 1.17, df¼ 1, 32; P¼ 0.28) 0, 12, 7 Right Anterior Inferior (RAI) orientation; 6 mm radius), and proportion that had obtained a college degree ( ¼ 2.00, df¼ 1, right dorsolateral prefrontal cortex (DLPFC; Alexopoulos et al., 2012; 31; P¼ 0.16). The groups did significantly differ in depression severity 36, 27, 29 RAI orientation; 8 mm radius), left dorsolateral pre- (t(30)¼ 13.41, P < 0.001) and trait rumination (t(30)¼ 7.32, frontal cortex (Alexopolous et al., 2012; 36, 27, 29 RAI orientation; 8 mm radius) and the right frontal insular cortex (RFIC; Seeley et al., P < 0.001), with the MDD group (Mean BDI-II Score¼ 23.13, 2007; 38, 13 3 RAI orientation; 8 mm radius). The subACC is part s.d.¼ 6.33; Mean RSQ-Rumination Score¼ 56.19, s.d.¼ 11.42) of the default mode network (Raichle and Snyder, 2007), and numer- scoring greater on the BDI-II and RSQ-Rumination than the HC ous studies have reported aberrant functioning in this region in indi- group (Mean BDI-II Score¼ 1.19, s.d.¼ 1.64; Mean viduals with MDD (see Drevets et al., 2008 for a review). The bilateral RSQ-Rumination Score¼ 31.38, s.d.¼ 7.33). There were no significant DLPFC is a central hub of the executive network (Seeley et al., 2007) differences in the percentage of images censored between the MDD and has been consistently reported to be dysfunctional in MDD group (M¼ 3.43%, s.d.¼ (4.99)) and the HC group (M¼ 2.10%, [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 747 Table 1 Significant group (MDD, HC) and thought induction condition (Self-Focus, Induction interaction between the left DLPFC and other regions External-Focus) main effects within the left executive network. Region Side BA xy z mm RFIC-salience network connectivity Group main effects Connectivity between the RFIC and other salience network structures MDD < HC L DLPFC-L executive connectivity was also characterized by a significant Group Thought Induction Inferior frontal gyrus L 9/44 52 7.5 26.5 1511 interaction. As with the right executive network, during the external- MDD < HC RFIC-salience connectivity Insula/precentral gyrus R 13/44 45 7.5 1.5 730 focus condition, the MDD group showed less connectivity between the MDD > HC RFIC-salience connectivity RFIC and the right dACC (10, 35.5, 26.5; BA32; 1189 mm ) Insula L 13 39 3.5 5.5 1103 compared with the HC group (See Figure 3). The MDD and HC Thought induction main effects group did not show significant differences in functional connectivity Self > External R DLPFC-R executive connectivity between the RFIC and other salience network structures during the Supramarginal gyrus/inferior parietal lobule R 40 55 48.5 19.5 3215 Insula/inferior frontal gyrus R 13 38 21.5 8.5 787 self-focused thought induction task. Self > External L DLPFC-L executive connectivity Inferior Parietal Lobule L 40 42 41.5 40.5 990 Self > External RFIC-salience connectivity DISCUSSION Insula L 13 35 14.5 8.5 861 The aim of the study was twofold: first to identify MDD-related Note. Contrasts that are not listed showed no main effects for group or thought induction. differences in functional connectivity in central nodes of the default MDD, major depressive disorder; HC, healthy control; DLPFC, dorsolateral prefrontal cortex; RFIC, right mode, executive and salience networks, and second to examine this frontal insular cortex; R, right; L, left. neural system connectivity in the context of eliciting self-referential thinking and non-self-referential, outwardly focused thinking. Greater activity in the default mode network has been associated s.d.¼ (2.57)), F (1, 30)¼ 0.903, P¼ 0.35,  P¼ 0.029). In addition, with more self-focus, whereas greater activity in the executive and there was no significant Group Thought Induction interaction or salience network has been linked to more attentional control over Thought Induction main effect in the percentage of images censored these thoughts. Given the different functions of these networks and (all P > 0.50). that excessive self-focus is a cognitive hallmark of MDD, we expected that during the self-focus condition, individuals with MDD would show greater default mode connectivity as well as less executive and SubACC-default mode network connectivity salience recruitment to help regulate these thoughts. When comparing differences in subACC connectivity with other struc- However, in contrast to previous work, we found no group differ- tures within the default mode network, there was a significant ences in connectivity within any of the three networks during the Group Thought Induction interaction, with the MDD group show- self-focus condition. Our findings may differ from previous studies ing greater subACC connectivity with the left medial temporal gyrus because research thus far has focused on either trait rumination-related (56, 3.5, 8.5; BA 21; 1141 mm ), left cingulate gyrus (6, 40.5, 33.5; BA resting state functional connectivity or task-based studies of rumin- 31; 946 mm ) and the right posterior cingulate/retrosplenial cortex ation-related activation in single brain structures. The current study (14, 45.5, 15.5; BA 29/31; 788 mm ) during the externally-focused was the first to examine functional connectivity within these networks thought induction compared with the HC group (See Figure 1). There when participants were instructed to engage in self-focused and were no significant differences between the MDD group and HC group externally-focused thinking. This difference in approach likely contrib- in connectivity between the subgenual ACC and other regions within uted to dissimilar findings compared with previous studies. With the default mode network during the self-focused thought induction. regard to resting state and self-referential task studies, it is thought that resting state and explicitly self-focused tasks may be eliciting dif- ferent aspects of self-focus (Lemogne et al., 2012). Specifically, it is Bilateral DLPFC-executive network connectivity thought that resting state studies may be tapping into a more Connectivity between the right DLPFC and other regions in the spontaneous or automatic self-focus whereas task-based studies using right executive network was also characterized by a significant instructed self-referential thought may be tapping into more strategic Group Thought Induction interaction. As with the default mode or controlled self-focus (Lemogne et al., 2012). Both aspects are network findings, the group differences were specific to the thought to be important to MDD, but likely have different underlying externally-focused thought condition; however, the pattern of connect- neural mechanisms (Shestyuk and Deldin, 2010). Shestyuk and ivity was in the opposite direction. During the externally-focused Deldin’s (2010) results suggest that automatic negative self-focus is thought induction, the right DLPFC showed decreased connectivity associated with a trait-like cognitive vulnerability in MDD that is with other right executive network regions among MDD participants even present amongst those who have remitted from MDD whereas compared with controls (See Figure 2). Specifically, the MDD group strategic negative self-focus is more state-related and mood-dependent. showed decreased connectivity between the right DLPFC and the right Given these results, it is possible that the addition of more explicitly middle frontal gyrus (35.0, 3.5, 47.5; BA6/8; 3885 mm ), an area negative mood-congruent self-focused statements may have better within the right frontopolar prefrontal cortex (42, 57.5, 1.5; BA10; captured differences in strategic self-focus in MDD versus healthy 1230 mm ) and the right posterior parietal cortex (41, 52.5, 26.5; BA participants. Additionally, using other tools with greater temporal 39/40; 696 mm ) compared with the HC group. The MDD participants resolution such as electroencephalography, which was employed by did not exhibit any increased right DLPFC-executive network connect- Shestyuk and Deldin (2010), may better capture differences in ivity. There were also no significant differences between the MDD and automatic versus strategic and elaborative self-focus. Further research controls in functional connectivity between the right DLPFC and other is needed to clarify what conditions may best elucidate the neural regions within the right executive network during the self-focused mechanisms underlying problematic strategic versus automatic condition. Additionally, there was no significant Group Thought self-focus in individuals with MDD. [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 748 SCAN (2015) E. L. Belleau et al. Fig. 1 Significant Group Thought Induction interactions in the default mode network. Specifically, individuals with MDD compared with HCs showed greater connectivity between the subACC and left medial temporal gyrus (coordinates: 56, 3.5, 8.5), left cingulate gyrus (coordinates: 6, 40.5, 33.5) and right posterior cingulate cortex (coordinates: 14, 45.5, 15.5) during the external-focus condition. There were no significant group differences in the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; SUBACC, subgenual anterior cingulate cortex. On the other hand, we found robust and consistent differences during the external thought condition an explicit trigger to engage within all three networks for the external-focus condition. in self-referential thought was not provided, it may be that this pattern Specifically, the MDD group showed greater default mode and less of findings is more representative of the automatic aspects of self-focus right executive and salience network connectivity compared with the that are characteristic of MDD. HC group. This pattern of results suggest that when trying to engage in One hypothesis is that for MDD participants, the external-focus task did not sufficiently engage working memory resources to effectively externally focused thought, individuals with MDD fail to adequately recruit the salience network, which is needed to switch on the executive compete with the intrusion of ruminative or self-referential thoughts. network in the service of down-regulating self-focused-related default This is in line with a number of studies suggesting that in order for mode network activity. Altogether, previous studies suggest that a attentional deployment from negative information to be successful, it pattern of hyperactive connectivity in the default mode network and must occupy enough working memory capacity so that insufficient resources are available for processing of negative information (e.g. hypoactive connectivity in the executive and salience networks are indicative of difficulties in disengaging from habitual ruminative Van Dillen and Koole, 2007; Van Dillen et al., 2009). Additionally, a thought (e.g. Berman et al., 2011; Carew et al., 2013). Given that negative mood has been shown to narrow one’s attentional scope [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 749 Fig. 2 Significant Group Thought Induction interactions in the right executive network. Specifically, individuals with MDD compared with HCs showed less connectivity between the right DLPFC and the right middle frontal gyrus (coordinates: 35, 3.5, 47.5), right frontopolar prefrontal cortex (coordinates: 42, 57.5, 1.5) and the right posterior parietal cortex (41, 52.5, 26.5) during the external-focus condition. There were no significant group differences during the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; DLPFC, dorsolateral prefrontal cortex. toward mood-congruent thoughts, which may preclude entry of infor- Our study has some notable strengths, particularly that our task mation into working memory that is irrelevant to negative self-focus design allowed us to examine MDD-related functional connectivity (see Whitmer and Gotlib, 2013). Because MDD participants are prone differences within important nodes of the default mode, executive to ruminative self-referential thinking, it is possible that during the and salience networks during self-referential and externally-focused external-focus task, the MDD participants were more likely to rumin- conditions. That said, there are a number of limitations to consider. ate (despite being task-irrelevant), thus further narrowing attentional We had a small heterogeneous MDD sample with five participants that scope and interfering with processing of incompatible externally were on psychotropic medication. While the sample contained focused information. In summary, our pattern of findings suggest variability in terms of number of prior episodes, symptom severity, that during the external-focus condition, individuals with MDD anxiety comorbidity and psychotropic medication usage, our findings were unable to recruit enough regulatory systems (the salience and appear to be in line with studies that contained similar heterogeneity executive networks) to help attenuate default mode network activity within their MDD sample (Greicius et al., 2007; Berman et al., 2011)as that is associated with the entry of intrusive self-focused thoughts. well as others that included a more homogenous MDD sample [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 750 SCAN (2015) E. L. Belleau et al. Fig. 3 Significant Group Thought Induction Interaction in the salience network. Specifically, individuals with MDD compared with HCs showed less connectivity between the RFIC and the right dACC (coordinates: 10, 35.5, 26.5) during the external-focus condition. There were no significant group differences in the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; RFIC, right frontal insular cortex; DACC, dorsal anterior cingulate cortex. (Hamilton et al., 2011). Additionally, while we cannot rule out medi- default mode network and switch on the executive network via the cation influences, when removing the five participants on psychotropic salience network. This pattern of neural network activity is consistent medications, the pattern of results remained the same. Furthermore, with an inability to disengage from self-focused thought processes, within the MDD group, there were largely no significant differences particularly when self-focus is incompatible with engaging in task between those who were taking psychotropic medications from those goals. This attentional control difficulty is thought to contribute to who were not (see Supplemental Material). The fact that we still found the development of rumination in those with MDD. Future studies aberrant functional connectivity within critical regions among each of would benefit from a more detailed analysis of these core neural net- the three neural networks suggests that the findings are not entirely works during tasks that vary in attentional demands and in content to explained by medication effects. determine what factors are most effective at disrupting rumination. Additionally, with regards to our task design, the external-focus condition differs from the self-focus condition on a number of other SUPPLEMENTARY DATA dimensions besides content, including level of abstractness and Supplementary data are available at SCAN. complexity. While this same paradigm has been used in a number of studies (e.g. Lyubomirsky and Nolen-Hoeksema, 1993; Nolen- Conflict of Interest Hoeksema and Morrow, 1993), the current study could have benefited None declared. from including an ‘abstract’ control condition to further understand neural substrates specific to self-focus, as was done by Cooney et al. REFERENCES (2010). Also, of note with respect to the thought induction task, we did Alexopoulos, G.S., Hoptmann, M.J., Kanellopoulos, D., Murphy, C.F., Lim, K.O., not include a means of verifying that the participants were indeed Gunning, F.M. (2012). Functional connectivity in the cognitive control network and generating the type of thoughts each condition is designed to elicit. the default mode network in late-life depression. Journal of Affective Disorders, 139, Nor did we assess the number of statements that each participant 56–65. completed. Beck, A.T., Rush, A.J., Shaw, B.F., Emery, G. (1979). Cognitive Therapy of Depression. In addition to some limitations with the design, we also chose to New York: Guilford Publications. Beevers, C.G., Clasen, P., Stice, E., Schnyer, D. (2010). Depression symptoms and cognitive focus on a small part of within-network functional connectivity in the control of emotion cues: a functional magnetic resonance imaging study. Neuroscience, default, executive and salience networks. In addition to the subgenual 167,97–103. anterior cingulate cortex, the central hub that we chose to examine Berman, M.G., Peltier, S., Nee, D.E., Kross, E., Deldin, P.J., Jonides, J. (2011). Depression, default mode network functional connectivity, the precuneus and dor- rumination and the default network. Social Cognitive and Affective Neuroscience, 6, 548–55. somedial prefrontal cortex are also critical to self-focused thought Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L. (2008). The brain’s default network: processes including rumination in both individuals with and without anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences, MDD (Johnson et al., 2009; Freton et al., 2014; Nejad et al., 2013). We 1124,1–38. also did not examine differences in connectivity between the three Carew, C.L., Miline, A.M., Tatham, E.L., MacQueen, G.M., Hall, G.B.C. (2013). Neural networks. Some studies have found differences in connectivity between systems underlying thought suppression in young women with, and at-risk, for depres- sion. Behavioral Brain Research, 257,13–24. these networks in those with and without MDD during a resting state Christoff, K., Gordon, A.M., Smalllwood, J., Smith, S.M., Schooler, J.W. (2009). Experience (Sheline et al., 2010), as well as during a self-focus task (Lemogne et al., sampling during fMRI reveals default network and executive system contributions to 2009). Given the complexity of the psychological processes being mind wandering. Proceedings of the National Academy of Sciences of the USA, 106, tapped, future research should incorporate more comprehensive 8719–24. Cooney, R.E., Joorman, J., Eugene, F., Dennis, E.L., Gotlib, I.H. (2010). Neural correlates of within and cross-network connectivity assessments. rumination in depression. Cognitive, Affective, and Behavioral Neuroscience, 10,470–8. In sum, our study provides novel evidence for dysregulated default Cox, R. (1996). AFNI: software for analysis and visualization of functional magnetic res- mode, executive and salience network connectivity in MDD partici- onance neuroimages. Computers and Biomedical Research, 29,162–73. pants during non-self-referential thought. The results suggest Cullen, K.R., Gee, D.G., Klimes-Dougan, B., et al. (2009). A preliminary study of functional that individuals with MDD are unable to adequately switch off the connectivity in comorbid adolescent depression. Neuroscience Letters, 460,227–31. [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 751 De Lissnyder, E., Koster, E.H.W., Everaert, J., Schacht, R., Van den Abeele, D., De Raedt, R. Lyubomirsky, S., Nolen-Hoeksema, S. (1993). Self-perpetuating properties of dysphoric (2012). Internal cognitive control in clinical depression: general but no emotion-specific rumination. Journal of Personality and Social Psychology, 65,339–49. impairments. Psychiatry Research, 199, 124–30. Manoliu, A., Meng, C., Brandl, F., et al. (2014). Insular dysfunction within the salience Demeyer, I., De Lissnyder, E., Koster, E.H.W., De Raedt, R. (2012). Rumination mediates network is associated with severity of symptoms and aberrant inter-network connectivity the relationship between impaired cognitive control for emotional information and in major depressive disorder. Frontiers in Human Neuroscience, 7(930), 1–17. Mason, M.F., Norton, M.I., Van Horn, J.D., et al. (2007). Wandering minds: the default depressive symptoms: a prospective study in remitted depressed adults. Behavior network and stimulus-independent thought. Science, 315, 393–5. Research and Therapy, 50, 292–7. Michalak, J., Holz, A., Teismann, T. (2011). Rumination as a predictor of relapse in mind- Dosenbach, N.U.F., Fair, D.A., Cohen, A.L., Schlaggar, B.L., Petersen, S.E. (2008). A dual- fulness-based cognitive therapy for depression. Psychology and Psychotherapy: Theory, networks architecture of top-down control. Trends in Cognitive Sciences, 12,99–105. Research, and Practice, 840,230–6. Drevets, W.C., Savitz, J., Trimble, M. (2008). The subgenual anterior cingulate cortex in Nejad, A.B., Fossati, P., Lemogne, C. (2013). Self-referential processing, rumination, and mood disorders. CNS Spectrums, 13(8), 663–81. cortical midline structures in major depression. Frontiers in Human Neuroscience, Eckert, M.A., Menon, V., Walczak, A., et al. (2009). At the heart of the ventral attention 7(666), 1–9. system: the right anterior insula. Human Brain Mapping, 30, 2530–41. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of Fales, C.L., Barch, D.M., Rundle, M.M., et al. (2008). Altered emotional interference pro- depressive episodes. Journal of Abnormal Psychology, 100,569–82. cessing in affective and cognitive-control brain circuitry in major depression. Biological Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders and mixed Psychiatry, 63, 377–84. anxiety/depressive symptoms. Journal of Abnormal Psychology, 109,504–11. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B. (1997). Structured Clinical Interview for Nolen-Hoeksema, S., Morrow, J. (1991). A prospective-study of depression and posttrau- the DSM-IV Axis I Disorders. Washington, DC: American Psychiatric Press Inc. matic stress symptoms after a natural disaster - the 1989 Loma-Prieta Earthquake. Fox, M.D., Snyder, A.Z., Vincent, J.L., Corbetta, M., Van Essen, D.C., Raichle, M.E. (2005). Journal of Personality and Social Psychology, 61,115–21. The human is intrinsically organized into dynamic, anticorrelated functional networks. Nolen-Hoeksema, S., Morrow, J. (1993). Effects of rumination and distraction on natur- Proceedings of the National Academy of Sciences of the USA, 102, 9673–8. ally-occuring depressed mood. Cognition and Emotion, 7,561–70. Franco, A.R., Pritchard, A., Calhoun, V.D., Mayer, A.R. (2009). Inter-rater and inter- Nolen-Hoeksema, S., Morrow, J., Fredrickson, B.L. (1993). Response styles and the method reliability for selecting the default mode network during data-driven analyses. duration of episodes of depressed mood. Journal of Abnormal Psychology, 102,20–8. Human Brain Mapping, 30, 2293–303. Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., Shulman, G.L. Freton, M., Lemogne, C., Delaveau, P., et al. (2014). The dark side of self-focus: brain (2001). A default mode of brain function. Proceedings of the National Academy of Sciences activity during self-focus in low and high brooders. Social Cognitive and Affective of the USA, 98, 676–82. Neuroscience, 9(11), 1808–13. Raichle, M.E., Snyder, A.Z. (2007). A default mode of brain function: a brief history of an Gohier, B., Ferracci, L., Surguladze, S.A., et al. (2009). Cognitive inhibition and working evolving idea. NeuroImage, 37, 1083–90. memory in unipolar depression. Journal of Affective Disorders, 116, 100–5. Seeley, W.W., Menon, V., Schatzberg, A.F., et al. (2007). Dissociable intrinsic connectivity Greicius, M.D., Flores, B.H., Menon, V., et al. (2007). Resting-state functional connectivity networks for salience processing and executive control. The Journal of Neuroscience, 27, in major depression: abnormal contributions from the subgenual cingulate cortex and 2349–56. thalamus. Biological Psychiatry, 62, 429–37. Sheline, Y.I., Barch, D.M., Price, J.L., et al. (2009). The default mode network and self- Hamilton, J.P., Chen, M.C., Gotlib, I.H. (2013). Neural systems approaches to understand- referential processes in depression. Proceedings of the National Academy of Sciences of the ing major depressive disorder: an intrinsic functional organization perspective. USA, 106, 1942–7. Neurobiology of Disease, 52,4–11. Sheline, Y.I., Price, J.L., Yan, Z., Mintun, M.A. (2010). Resting state functional MRI Hamilton, J.P., Etkin, A., Furman, D.J., Lemus, M.G., Johnson, R.F., Gotlib, I.H. (2012). in depression unmasks increased connectivity between networks via the dorsal nexus. Functional neuroimaging of major depressive disorder: a meta-analysis and new inte- Proceesings of the National Academy of Sciences of the USA, 107, 11020–5. gration of baseline activation and neural response data. American Journal of Psychiatry, Shenhav, A., Botvinick, M.M., Cohen, J.D. (2013). The expected value of control: an in- 169, 693–703. tegrative theory of anterior cingulate cortex function. Neuron, 79,217–40. Hamilton, J.P., Furman, D.J., Chang, C., Thomason, M.E., Dennis, E., Gotlib, I.H. (2011). Shestyuk, A.Y., Deldin, P.J. (2010). Automatic and strategic representation of the self in Default-mode and task-positive network activity in major depressive disorder: implica- major depression: trait and state abnormalities. American Journal of Psychiatry, 167, tions for adaptive and maladaptive rumination. Biological Psychiatry, 70, 327–33. 536–44. Harvey, P.O., Fossati, P., Pochon, J.B., et al. (2005). Cognitive control and brain resources Slitz, D., Hayley, S. (2012). Major depressive disorder and alterations in insular cortical in major depression: an fMRI study using the n-back task. Neuroimage, 26, 860–9. activity: a review of current functional magnetic imaging research. Frontiers in Human Hodgson, T., Chamberlain, M., Parris, B., et al. (2007). The role of the ventrolateral frontal Neuroscience, 6(323), 1–14. cortex in inhibitory oculomotor control. Brain, 130, 1525–37. Sridharan, D., Levitin, D.J., Menon, V. (2008). A critical role for the right fronto-insular Johnson, D.P., Whisman, M.A. (2013). Gender differences in rumination: a meta-analysis. cortex in switching between central-executive and default-mode networks. Proceedings of Personality and Individual Differences, 55, 367–74. the National Academy of Sciences of the USA, 105, 12569–74. Johnson, M.K., Nolen-Hoeksema, S., Mitchell, K.J., Levin, Y. (2009). Medial cortex activity, Van Dillen, L., Heslenfeld, D.J., Koole, S.L. (2009). Tuning down the emotional brain: an self-reflection and depression. Social Cognitive and Affective Neuroscience, 4, 313–27. fMRIstudy of the effects of cognitive load on the processing of affective images. Jones, N.P., Siegle, G.J., Thase, M.E. (2008). Effects of rumination and initial severity on NeuroImage, 45, 1212–19. remission to cognitive therapy for depression. Cognitive Therapy and Research, 32, Van Dillen, L.F., Koole, S.L. (2007). Clearing the mind: a working memory model of 591–604. distraction from negative feelings. Emotion, 7, 715–23. Just, N., Alloy, L.B. (1997). The response styles theory of depression: tests and an extension Vanhaudenhuyse, A., Demertzi, A., Schabus, M., et al. (2010). Two distinct neural networks of the theory. Journal of Abnormal Psychology, 106, 221–9. mediate the awareness of environment and of self. Journal of Cognitive Neuroscience, 23, Koster, E.H.W., De Lissnyder, E., Derakashan, N., De Raedt, R. (2011). Understanding 570–8. depressive rumination from a cognitive science perspective: the impaired disengagement Wang, L., LaBar, K.S., Smoski, M., et al. (2008). Prefrontal mechanisms for executive hypothesis. Clinical Psychology Review, 31, 138–45. control over emotional distraction are altered in major depression. Psychiatry Kuhn, S., Vanderhasselt, M.A., De Raedt, R., Gallinat, J. (2012). Why ruminators won’t Research: Neuroimaging, 163,143–55. stop: the structural and resting state correlates of rumination and its relation to depres- Wei, M., Qin, J., Yan, R., Li, H., Yao, Z., Qing, L. (2013). Identifying major depressive sion. Journal of Affective Disorders, 141, 352–60. disorder using hurst exponent of resting-state brain networks. Psychiatry Research: Laird, A.R., Fox, P.M., Eickhoff, S.B., et al. (2011). Behavioral interpretations of intrinsic Neuroimaging, 214, 306–12. connectivity networks. Journal of Cognitive Neuroscience, 23, 4022–37. Weissman, D.H., Roberts, K.C., Visscher, K.M., Woldorff, M.G. (2006). The neural basis of Lemogne, C., Delaveau, P., Freton, M., Guionnett, S., Fossati, P. (2012). Medial prefrontal momentary lapses in attention. Nature Neuroscience, 9,971–8. cortex and the self in major depression. Journal of Affective Disorders, 136,e1–e11. Whitmer, A.J., Gotlib, I.H. (2013). An attentional scope model of rumination. Psychological Lemogne, C., LeBastard, G., Mayberg, H., et al. (2009). In search of the depressive self: Bulletin, 139, 1036–61. extended medial prefrontal network during self-referential processing in major depres- Zhou, Y., Yu, C., Zheng, H., et al. (2010). Increased neural resources recruitment in the sion. Social Cognitive and Affective Neuroscience, 4, 305–12. intrinsic organization in major depression. Journal of Affective Disorders, 121,220–30. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Social Cognitive and Affective Neuroscience Oxford University Press

Imbalance of default mode and regulatory networks during externally focused processing in depression

Download PDF
8 pages

Loading next page...
 
/lp/oxford-university-press/imbalance-of-default-mode-and-regulatory-networks-during-externally-ebtuUaG4Ls
Publisher
Oxford University Press
Copyright
© The Author (2014). Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com
ISSN
1749-5016
eISSN
1749-5024
DOI
10.1093/scan/nsu117
pmid
25274576
Publisher site
See Article on Publisher Site

Abstract

[28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 doi:10.1093/scan/nsu117 SCAN (2015) 10, 744 ^751 Imbalance of default mode and regulatory networks during externally focused processing in depression Emily L. Belleau, Lauren E. Taubitz, and Christine L. Larson Department of Psychology, University of Wisconsin-Milwaukee, 2441 East Hartford Avenue, Milwaukee, WI 53211, USA Attentional control difficulties likely underlie rumination, a core cognitive vulnerability in major depressive disorder (MDD). Abnormalities in the default mode, executive and salience networks are implicated in both rumination and attentional control difficulties in MDD. In the current study, individuals with MDD (n¼ 16) and healthy controls (n¼ 16) completed tasks designed to elicit self-focused (ruminative) and externally-focused thinking during fMRI scanning. The MDD group showed greater default mode network connectivity and less executive and salience network connectivity during the external- focus condition. Contrary to our predictions, there were no differences in connectivity between the groups during the self-focus condition. Thus, it appears that when directed to engage in self-referential thinking, both depressed and non-depressed individuals similarly recruit networks supporting this process. In contrast, when instructed to engage in non-self-focused thought, non-depressed individuals show a pattern of network connectivity indicative of minimized self-referential processing, whereas depressed individuals fail to reallocate neural resources in a manner consistent with effective down regulation of self-focused thought. This is consistent with difficulties in regulating self-focused thinking in order to engage in more goal-directed behavior that is seen in individuals with MDD. Keywords: depression; rumination; default mode network; executive network; salience network A core cognitive feature of major depressive disorder (MDD) is the information processing in both those with MDD and healthy controls tendency to ruminatethe persistent and passive focus on one’s (Nejad et al., 2013). Individuals with MDD exhibit stronger functional negative mood and the adverse consequences of being in a negative connections between the subgenual anterior cingulate cortex mood (Nolen-Hoeksema, 1991). Rumination is related to a host of (subACC) and other areas within the default mode network unfavorable outcomes including more severe and prolonged depressive (Greicius et al., 2007; Berman et al., 2011). Moreover, greater resting symptoms (Nolen-Hoeksema and Morrow, 1991; Nolen-Hoeksema state functional connectivity between the subACC and posterior et al., 1993; Just and Alloy, 1997), a poorer response to treatment cingulate cortex has been tied to greater trait rumination in individuals (Jones et al., 2008) and an increased risk for relapse after remission with MDD and healthy controls (Berman et al., 2011). However, when of a depressive episode (Michalak et al., 2011). Additionally, there is directly comparing depressed and non-depressed individuals during a evidence showing that rumination serves as a risk factor for the devel- task that engages self-focus, depressed individuals showed greater ac- opment of MDD (Just and Alloy, 1997; Nolen-Hoeksema, 2000). tivation in a number of default mode network structures compared Attentional control difficulties, another central trait in MDD (e.g. with non-depressed individuals (Cooney et al., 2010). Gohier et al., 2009; De Lissnyder et al., 2012), may underlay ruminative On the other hand, when a task is cognitively demanding or requires thinking by undermining the ability to disengage from negative self- focus on the external environment, the default mode network normally relevant material (Koster et al., 2011). A recent study demonstrated deactivates (Vanhaudenhuyse et al., 2010). An inability to deactivate the that rumination partially mediated the relationship between atten- default mode network during such tasks has been positively associated tional control difficulties and future depressive symptoms among with greater attentional control difficulties among healthy volunteers MDD patients (Demeyer et al., 2012). (Weissman et al., 2006; Mason et al., 2007; Christoff et al., 2009). One The neural underpinnings of both rumination and attentional study demonstrated that compared with healthy controls, depressed dyscontrol in MDD are thought to implicate three neural networks individuals were unable to deactivate the default mode network when (see Hamilton et al., 2013): the default mode network (Raichle et al., instructed to disengage from negative stimuli (Sheline et al.,2009). 2001), the executive network (Seeley et al., 2007) and the salience MDD participants have also shown difficulties deactivating an inferior network (Seeley et al., 2007). The default mode network is composed portion of the anterior medial prefrontal cortex, a central hub of the of a number of midline regions including the medial prefrontal cortex, default mode network, during a task designed to evoke external focus. the anterior cingulate cortex, the precuneus, the posterior cingulate This impaired deactivation was also associated with self-reported ru- cortex, the retrosplenial cortex, lateralized areas of the parietal cortex mination in the MDD group (Johnson et al., 2009). (Raichle and Snyder, 2007), as well as the medial temporal cortex and In addition to the default mode network, altered coordinated activ- hippocampal formation (Buckner et al., 2008). The default mode net- ity within the executive network, which encompasses areas in the work is thought to subserve a number of internally focused attentional lateral prefrontal cortex and the posterior parietal cortex (Seeley processes (Vanhaudenhuyse et al., 2010) including self-referential et al., 2007; Dosenbach et al., 2008), has been implicated in MDD (Zhou et al., 2010; Wei et al., 2013). Activation in this network is anticorrelated with the default mode network (Fox et al., 2005) and Received 22 March 2014; Revised 19 July 2014; Accepted 1 September 2014 is essential for paying attention to external environmental demands Advance Access publication 1 October 2014 The authors gratefully acknowledge the adults that took the time to participate in the study. The authors also (Vanhaudenhuyse et al., 2010) as well as engaging in goal-directed want to thank Michael Duke and Kim Lewis for help with participant recruitment as well as Judi Zaferos-Pylant and or cognitively demanding activities (Dosenbach et al., 2008). Yu Liu for their role in acquiring the fMRI data. The work was supported by the University of Wisconsin-Milwaukee Deficient activations within executive network regions have been Research Growth Initiative (C.L.L.) and the National Institute of Mental Health K01 (MH086809 to C.L.L.). linked with greater trait rumination amongst healthy individuals Correspondence should be addressed to Christine L. Larson, Department of Psychology, University of Wisconsin Milwaukee, 2441 East Hartford Ave., Milwaukee, WI, 53211, USA. E-mail: larsoncl@uwm.edu (Kuhn et al., 2012), as well as poor attentional control in MDD The Author (2014). Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 745 (Fales et al., 2008; Wang et al., 2008) and dysphoric participants without MDD on levels of neural activity in some default mode, sali- (Beevers et al., 2010). Recently, it was shown that MDD individuals ence and executive network structures (Cooney et al., 2010). We exam- recruited executive network regions to a lesser extent when instructed ined group connectivity differences by separately correlating central to inhibit both ruminative and emotionally neutral thoughts (Carew nodes of each network shown to be aberrant in MDD with other et al., 2013). However, some studies have found elevated activity in regions contained within that specific network. While there have areas within the executive network when instructed to engage in self- been some conflicting findings, based on the weight of the extant evi- focused thinking (e.g. Lemogne et al., 2009; Cooney et al., 2010), which dence, we expected that during the self-focused thought induction, the may potentially be a compensatory mechanism for counteracting cog- MDD group would show greater connectivity within the default mode nitive vulnerabilities in MDD (e.g. Harvey et al., 2005). network and decreased connectivity within the executive network Another network that is associated with aberrant coordinated activ- compared to those without MDD. Within the salience network, we ity in individuals with MDD compared with healthy controls is the specifically predicted decreased connectivity between the RFIC and salience network (Wei et al., 2013; Manoliu et al., 2014), which is dACC in participants with MDD compared with healthy controls, composed of the anterior insula/frontal operculum, dorsal anterior given the role these structures play in attentional control abilities cingulate cortex/paracingulate cortex and the superior temporal pole (Eckert et al., 2009; Shenhav et al., 2013). (Seeley et al., 2007; Dosenbach et al., 2008). Activation in the salience network is thought to be associated with monitoring and drawing METHODS attention to stimuli that are subjectively relevant (Seeley et al., 2007). Participants Additionally, the salience network plays a role in switching mental sets, Eighteen individuals diagnosed with MDD and 18 healthy controls with damage to the anterior insula being associated with poor switch- (HC) were administered the Structured Clinical Interview for the ing ability (Hodgson et al., 2007). In regards to switching behavior, it DSM-IV (First et al., 1995). MDD participants had to be free of any has been shown that the right frontal insular cortex (RFIC) and dorsal history of bipolar disorder, schizophrenia or psychosis and current anterior cingulate cortex (dACC) play a role in neural network switch- alcohol or substance abuse. Anxiety disorder comorbidity and the ing, initiating deactivation of the default mode and activation in the use of psychotropic medication were permitted in the MDD group. executive network (Sridharan et al., 2008). Evidence suggests that The HC group was free of any current Axis I psychopathology or individuals with MDD have reduced connectivity between the RFIC lifetime history of a mood disorder. Two participants from the with other regions in the salience network (Manoliu et al., 2014). MDD group and two participants from the HC group were excluded Specifically, compared with healthy individuals, MDD participants from further analysis due to excessive head motion in the scanner, show deficient recruitment of salience network structures when resulting in a final sample of 16 in the HC group (8 males, mean directed to inhibit ruminative and emotionally neutral thoughts age¼ 31.13; s.d.¼ 10.70) and 16 in the MDD group (5 males, mean (Carew et al., 2013). Furthermore, individuals with MDD show an age¼ 33.19, s.d.¼ 11.49; mean duration of current MDD abnormal relationship between activation in the RFIC and the onset episode¼ 21.63 months, s.d.¼ 49.49). of increased activation in the executive network versus the default Additionally, ten MDD participants reported that they were experi- mode network. Individuals with MDD show a rise in RFIC activation encing their first episode or have had one prior episode, three indi- at the onset of increased executive network activation, whereas healthy viduals reported experiencing three to six prior episodes, and another individuals show greater RFIC activation during a rise in default mode three individuals reported experiencing 10 or more prior major network activation (Hamilton et al., 2011). Given this evidence, the depressive episodes. Five MDD participants were currently taking psy- salience network likely plays a role in attentional dyscontrol and ru- chotropic medication (n¼ 2 duloxetine hydrochloride/Cymbalta, mination in MDD. n¼ 1 bupropion hydrochloride/Wellbutrin XL, n¼ 1 bupropion Thus, existing research has shown a potential link between activity hydrochloride and fluoxetine hydrochloride/Prozac, n¼ 1 risperi- in these three networks with rumination and the attentional difficulties done/Risperdal). Additionally, eleven MDD participants were diag- that may underlie rumination in MDD. However, these studies contain nosed with comorbid anxiety disorders (n¼ 2 social phobia, n¼ 2 limitations. While resting state functional connectivity studies have social and specific phobias, n¼ 2 generalized anxiety disorder, n¼ 1 paved the way for understanding of aberrant intrinsic connectivity in generalized anxiety disorder and social phobia, n¼ 1 generalized these core networks in MDD, the lack of constraints placed on behav- anxiety disorder and panic disorder, n¼ 1 posttraumatic stress dis- ior makes it difficult to clarify what role these networks are playing in order, n¼ 1 specific phobia, n¼ 1 anxiety disorder not otherwise spe- the cognitive difficulties in MDD. On the other hand, while task-based cified). At the scanning session, participants completed the Beck designs have allowed examination of aberrant brain activity when Depression Inventory-II (BDI-II; Beck et al., 1979) and the rumination individuals with MDD are instructed to engage in self-referential pro- subscale of the Response Styles Questionnaire (Nolen-Hoeksema, cessing (e.g. Cooney et al., 2010), researchers have typically not focused 1991). The study was approved by the University of Wisconsin- on how brain regions work together to contribute to cognitive Milwaukee Institutional Review Board, and participants provided writ- problems in MDD. It has been proposed that characterizing functional ten informed consent according to the Declaration of Helsinki. connections amongst neural regions may be a useful approach for Participants were paid for completion of the interview and scanning understanding the cognitive, behavioral and emotional difficulties in session. depression (Hamilton et al., 2013). The current study aimed to examine functional connections within Self-focused and externally focused thought induction tasks the default mode, executive and salience networks among individuals with and without MDD during a task in which participants were While in the scanner, participants completed an 8-min self-focus explicitly instructed to think about topics designed to elicit self-focused thought induction task and an 8-min external-focus thought induction and externally-focused thinking. Previous studies have found that task (adapted from Lyubomirsky and Nolen-Hoeksema, 1993). The compared with the external-focus condition, the self-focus condition self-focus thought induction statements were designed to generate used in this paradigm effectively generates rumination in depressed more abstract, internally focused, open-ended conceptual mental rep- individuals (Lyubomirsky and Nolen-Hoeksema, 1993; Nolen- resentations. On the other hand, the external-focus statements were Hoeksema and Morrow, 1993) and differentiates those with and meant to evoke more concrete, outwardly-focused, and stereotypical, [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 746 SCAN (2015) E. L. Belleau et al. visually dependent mental representations of external objects or scenes. (Hamilton et al., 2012). Lastly, the RFIC is a fundamental region in In each of the thought induction tasks, participants were visually the salience network that has shown to be abnormal in depression and presented with up to 45 statements. The self-focused statements were related to attentional control and rumination (see Slitz and Hayley, neither inherently negative nor positive, but open to individual inter- 2012 for a review). Pearson’s r maps were generated by correlating pretation. An example of a statement from the self-focused thought each seed with other regions belonging to each of their respective induction was ‘Think about why you react the way you do’ and ‘Think neural networks. The time series of the subACC seed was correlated about sitting down and analyzing your personality.’ The externally with every other voxel within the default mode network and the RFIC focused thought induction included more visually dependent seed was correlated with every other voxel in the salience network. statements such as ‘Think about and imagine a boat slowly crossing Similarly, the right DLPFC time series was correlated with regions the Atlantic.’ For a full list of the thought induction statements, please within the right executive network and the left DLPFC time series see the Supplementary Materials. At the beginning of each thought was correlated with regions in the left executive network. The analyses induction task, participants were told to read each statement slowly were restricted to each of the respective network regions by applying and silently to themselves. Participants were asked to use their imagin- previous established spatial templates representing each of the three ation and concentration to focus on each of the ideas being presented. networks (Franco et al., 2009; Laird et al., 2011). The r statistics were Participants were instructed to proceed with each statement at their then normalized using a Fischer’s z transformation and then resampled own pace. In between each statement, subjects were presented with a to 1-mm voxels. This normalized data were used to calculate all group- one second fixation cross. The order of the two thought induction level statistics. tasks were counterbalanced within each of the groups. To determine whether network connectivity differed as a function of group and task, a Group (MDD, HC) Thought Induction (self- focused, externally-focused) ANOVA was calculated separately for Imaging parameters each of the three networks. Participant was included as a random Whole brain imaging was conducted using a 3-T short bore Signa factor in each model. To correct for multiple comparisons, cluster Excite system. Functional images were collected using a T2* weighted thresholding using Monte Carlo simulations was applied to each of gradient-echo, echoplanar pulse sequence (TR¼2s; TE¼ 25 ms; the three separate neural network templates. Using a voxel-based FOV¼ 24 cm, flip angle¼ 848; voxel size 3.75 3.75 4 mm, sagittal P < 0.01 and nearest-neighbor selection criteria, accounting for spatial orientation). The self-focused and externally-focused thought correlation, clusters greater than 570 mm for the default mode network induction tasks consisted of 232 whole brain scans. High resolution and the salience network, and 567 mm for the right and left executive spoiled gradient recalled (SPGR) images (1 mm sagittal slices; network achieved a corrected P < 0.05. Because our hypotheses were TR¼ 9.6 s; TE¼ 3.9 ms; field of view¼ 24 cm; flip angle¼ 128; voxel centered on Group Thought Induction interaction influences on size¼ 0.9375 0.9375 1) were collected and served as an anatomical functional connectivity within the three networks, we focused on map for the functional images. these findings in the results section below. For a listing of Group and Thought Induction main effects, see Table 1. Additionally, given that Imaging preprocessing there are gender differences in the tendency to ruminate (Johnson and Reconstruction and image processing was conducted in AFNI (Cox, Whisman, 2013) and that the depressed sample did not have the same 1996). The first seven images for each scan were excluded to allow gender proportions as the control sample, we also conducted secondary for magnetic saturation. The remaining images were despiked, analyses examining whether our Group Thought Induction volume registered to correct for motion, spatially normalized to interaction results were influenced by gender. We also tested whether Talairach space, and spatially smoothed by convolution with an medication effects influenced our ANOVA findings, as we had five isotropic Gaussian kernel (full-width at half-maximum¼ 6 mm) to individuals with MDD on psychotropic medication, including an decrease high spatial frequency noise. For subsequent analyses, the individual taking an antipsychotic medication (not standard for the following nuisance regressors were included: baseline, signal drift, treatment of depression). These results are in the Supplementary estimates from a rigid-body motion correction that was done in Material section. three translational and three rotational directions, and censoring of Finally, we also correlated Significant Group Thought Induction high motion images (cutoff 0.3 mm). Additionally, a band pass filter interaction findings with the Beck Depression Inventory-II (BDI-II; was applied to attenuate frequencies above 0.1 Hz and below 0.01 Hz. Beck et al., 1979) and the rumination subscale of the Response Styles Subjects who exhibited excessive head motion (greater than an average Questionnaire (RSQ-Rumination; Nolen-Hoeksema, 1991). These value of 2.5 mm translation and/or 2.58 rotation) were excluded from results are also presented in the Supplementary Figures S1–S3. further analysis. RESULTS Seed-based functional connectivity analyses Participant characteristics The time series was extracted from four seeds, including a seed placed The two groups did not differ significantly with regard to age (t¼ .53, in the subgenual anterior cingulate cortex (subACC; Cullen et al., 2009; df¼ 30, P¼ 0.60), sex distribution ( ¼ 1.17, df¼ 1, 32; P¼ 0.28) 0, 12, 7 Right Anterior Inferior (RAI) orientation; 6 mm radius), and proportion that had obtained a college degree ( ¼ 2.00, df¼ 1, right dorsolateral prefrontal cortex (DLPFC; Alexopoulos et al., 2012; 31; P¼ 0.16). The groups did significantly differ in depression severity 36, 27, 29 RAI orientation; 8 mm radius), left dorsolateral pre- (t(30)¼ 13.41, P < 0.001) and trait rumination (t(30)¼ 7.32, frontal cortex (Alexopolous et al., 2012; 36, 27, 29 RAI orientation; 8 mm radius) and the right frontal insular cortex (RFIC; Seeley et al., P < 0.001), with the MDD group (Mean BDI-II Score¼ 23.13, 2007; 38, 13 3 RAI orientation; 8 mm radius). The subACC is part s.d.¼ 6.33; Mean RSQ-Rumination Score¼ 56.19, s.d.¼ 11.42) of the default mode network (Raichle and Snyder, 2007), and numer- scoring greater on the BDI-II and RSQ-Rumination than the HC ous studies have reported aberrant functioning in this region in indi- group (Mean BDI-II Score¼ 1.19, s.d.¼ 1.64; Mean viduals with MDD (see Drevets et al., 2008 for a review). The bilateral RSQ-Rumination Score¼ 31.38, s.d.¼ 7.33). There were no significant DLPFC is a central hub of the executive network (Seeley et al., 2007) differences in the percentage of images censored between the MDD and has been consistently reported to be dysfunctional in MDD group (M¼ 3.43%, s.d.¼ (4.99)) and the HC group (M¼ 2.10%, [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 747 Table 1 Significant group (MDD, HC) and thought induction condition (Self-Focus, Induction interaction between the left DLPFC and other regions External-Focus) main effects within the left executive network. Region Side BA xy z mm RFIC-salience network connectivity Group main effects Connectivity between the RFIC and other salience network structures MDD < HC L DLPFC-L executive connectivity was also characterized by a significant Group Thought Induction Inferior frontal gyrus L 9/44 52 7.5 26.5 1511 interaction. As with the right executive network, during the external- MDD < HC RFIC-salience connectivity Insula/precentral gyrus R 13/44 45 7.5 1.5 730 focus condition, the MDD group showed less connectivity between the MDD > HC RFIC-salience connectivity RFIC and the right dACC (10, 35.5, 26.5; BA32; 1189 mm ) Insula L 13 39 3.5 5.5 1103 compared with the HC group (See Figure 3). The MDD and HC Thought induction main effects group did not show significant differences in functional connectivity Self > External R DLPFC-R executive connectivity between the RFIC and other salience network structures during the Supramarginal gyrus/inferior parietal lobule R 40 55 48.5 19.5 3215 Insula/inferior frontal gyrus R 13 38 21.5 8.5 787 self-focused thought induction task. Self > External L DLPFC-L executive connectivity Inferior Parietal Lobule L 40 42 41.5 40.5 990 Self > External RFIC-salience connectivity DISCUSSION Insula L 13 35 14.5 8.5 861 The aim of the study was twofold: first to identify MDD-related Note. Contrasts that are not listed showed no main effects for group or thought induction. differences in functional connectivity in central nodes of the default MDD, major depressive disorder; HC, healthy control; DLPFC, dorsolateral prefrontal cortex; RFIC, right mode, executive and salience networks, and second to examine this frontal insular cortex; R, right; L, left. neural system connectivity in the context of eliciting self-referential thinking and non-self-referential, outwardly focused thinking. Greater activity in the default mode network has been associated s.d.¼ (2.57)), F (1, 30)¼ 0.903, P¼ 0.35,  P¼ 0.029). In addition, with more self-focus, whereas greater activity in the executive and there was no significant Group Thought Induction interaction or salience network has been linked to more attentional control over Thought Induction main effect in the percentage of images censored these thoughts. Given the different functions of these networks and (all P > 0.50). that excessive self-focus is a cognitive hallmark of MDD, we expected that during the self-focus condition, individuals with MDD would show greater default mode connectivity as well as less executive and SubACC-default mode network connectivity salience recruitment to help regulate these thoughts. When comparing differences in subACC connectivity with other struc- However, in contrast to previous work, we found no group differ- tures within the default mode network, there was a significant ences in connectivity within any of the three networks during the Group Thought Induction interaction, with the MDD group show- self-focus condition. Our findings may differ from previous studies ing greater subACC connectivity with the left medial temporal gyrus because research thus far has focused on either trait rumination-related (56, 3.5, 8.5; BA 21; 1141 mm ), left cingulate gyrus (6, 40.5, 33.5; BA resting state functional connectivity or task-based studies of rumin- 31; 946 mm ) and the right posterior cingulate/retrosplenial cortex ation-related activation in single brain structures. The current study (14, 45.5, 15.5; BA 29/31; 788 mm ) during the externally-focused was the first to examine functional connectivity within these networks thought induction compared with the HC group (See Figure 1). There when participants were instructed to engage in self-focused and were no significant differences between the MDD group and HC group externally-focused thinking. This difference in approach likely contrib- in connectivity between the subgenual ACC and other regions within uted to dissimilar findings compared with previous studies. With the default mode network during the self-focused thought induction. regard to resting state and self-referential task studies, it is thought that resting state and explicitly self-focused tasks may be eliciting dif- ferent aspects of self-focus (Lemogne et al., 2012). Specifically, it is Bilateral DLPFC-executive network connectivity thought that resting state studies may be tapping into a more Connectivity between the right DLPFC and other regions in the spontaneous or automatic self-focus whereas task-based studies using right executive network was also characterized by a significant instructed self-referential thought may be tapping into more strategic Group Thought Induction interaction. As with the default mode or controlled self-focus (Lemogne et al., 2012). Both aspects are network findings, the group differences were specific to the thought to be important to MDD, but likely have different underlying externally-focused thought condition; however, the pattern of connect- neural mechanisms (Shestyuk and Deldin, 2010). Shestyuk and ivity was in the opposite direction. During the externally-focused Deldin’s (2010) results suggest that automatic negative self-focus is thought induction, the right DLPFC showed decreased connectivity associated with a trait-like cognitive vulnerability in MDD that is with other right executive network regions among MDD participants even present amongst those who have remitted from MDD whereas compared with controls (See Figure 2). Specifically, the MDD group strategic negative self-focus is more state-related and mood-dependent. showed decreased connectivity between the right DLPFC and the right Given these results, it is possible that the addition of more explicitly middle frontal gyrus (35.0, 3.5, 47.5; BA6/8; 3885 mm ), an area negative mood-congruent self-focused statements may have better within the right frontopolar prefrontal cortex (42, 57.5, 1.5; BA10; captured differences in strategic self-focus in MDD versus healthy 1230 mm ) and the right posterior parietal cortex (41, 52.5, 26.5; BA participants. Additionally, using other tools with greater temporal 39/40; 696 mm ) compared with the HC group. The MDD participants resolution such as electroencephalography, which was employed by did not exhibit any increased right DLPFC-executive network connect- Shestyuk and Deldin (2010), may better capture differences in ivity. There were also no significant differences between the MDD and automatic versus strategic and elaborative self-focus. Further research controls in functional connectivity between the right DLPFC and other is needed to clarify what conditions may best elucidate the neural regions within the right executive network during the self-focused mechanisms underlying problematic strategic versus automatic condition. Additionally, there was no significant Group Thought self-focus in individuals with MDD. [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 748 SCAN (2015) E. L. Belleau et al. Fig. 1 Significant Group Thought Induction interactions in the default mode network. Specifically, individuals with MDD compared with HCs showed greater connectivity between the subACC and left medial temporal gyrus (coordinates: 56, 3.5, 8.5), left cingulate gyrus (coordinates: 6, 40.5, 33.5) and right posterior cingulate cortex (coordinates: 14, 45.5, 15.5) during the external-focus condition. There were no significant group differences in the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; SUBACC, subgenual anterior cingulate cortex. On the other hand, we found robust and consistent differences during the external thought condition an explicit trigger to engage within all three networks for the external-focus condition. in self-referential thought was not provided, it may be that this pattern Specifically, the MDD group showed greater default mode and less of findings is more representative of the automatic aspects of self-focus right executive and salience network connectivity compared with the that are characteristic of MDD. HC group. This pattern of results suggest that when trying to engage in One hypothesis is that for MDD participants, the external-focus task did not sufficiently engage working memory resources to effectively externally focused thought, individuals with MDD fail to adequately recruit the salience network, which is needed to switch on the executive compete with the intrusion of ruminative or self-referential thoughts. network in the service of down-regulating self-focused-related default This is in line with a number of studies suggesting that in order for mode network activity. Altogether, previous studies suggest that a attentional deployment from negative information to be successful, it pattern of hyperactive connectivity in the default mode network and must occupy enough working memory capacity so that insufficient resources are available for processing of negative information (e.g. hypoactive connectivity in the executive and salience networks are indicative of difficulties in disengaging from habitual ruminative Van Dillen and Koole, 2007; Van Dillen et al., 2009). Additionally, a thought (e.g. Berman et al., 2011; Carew et al., 2013). Given that negative mood has been shown to narrow one’s attentional scope [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 749 Fig. 2 Significant Group Thought Induction interactions in the right executive network. Specifically, individuals with MDD compared with HCs showed less connectivity between the right DLPFC and the right middle frontal gyrus (coordinates: 35, 3.5, 47.5), right frontopolar prefrontal cortex (coordinates: 42, 57.5, 1.5) and the right posterior parietal cortex (41, 52.5, 26.5) during the external-focus condition. There were no significant group differences during the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; DLPFC, dorsolateral prefrontal cortex. toward mood-congruent thoughts, which may preclude entry of infor- Our study has some notable strengths, particularly that our task mation into working memory that is irrelevant to negative self-focus design allowed us to examine MDD-related functional connectivity (see Whitmer and Gotlib, 2013). Because MDD participants are prone differences within important nodes of the default mode, executive to ruminative self-referential thinking, it is possible that during the and salience networks during self-referential and externally-focused external-focus task, the MDD participants were more likely to rumin- conditions. That said, there are a number of limitations to consider. ate (despite being task-irrelevant), thus further narrowing attentional We had a small heterogeneous MDD sample with five participants that scope and interfering with processing of incompatible externally were on psychotropic medication. While the sample contained focused information. In summary, our pattern of findings suggest variability in terms of number of prior episodes, symptom severity, that during the external-focus condition, individuals with MDD anxiety comorbidity and psychotropic medication usage, our findings were unable to recruit enough regulatory systems (the salience and appear to be in line with studies that contained similar heterogeneity executive networks) to help attenuate default mode network activity within their MDD sample (Greicius et al., 2007; Berman et al., 2011)as that is associated with the entry of intrusive self-focused thoughts. well as others that included a more homogenous MDD sample [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 750 SCAN (2015) E. L. Belleau et al. Fig. 3 Significant Group Thought Induction Interaction in the salience network. Specifically, individuals with MDD compared with HCs showed less connectivity between the RFIC and the right dACC (coordinates: 10, 35.5, 26.5) during the external-focus condition. There were no significant group differences in the self-focus condition. MDD, major depressive disorder group; HC, healthy control group; RFIC, right frontal insular cortex; DACC, dorsal anterior cingulate cortex. (Hamilton et al., 2011). Additionally, while we cannot rule out medi- default mode network and switch on the executive network via the cation influences, when removing the five participants on psychotropic salience network. This pattern of neural network activity is consistent medications, the pattern of results remained the same. Furthermore, with an inability to disengage from self-focused thought processes, within the MDD group, there were largely no significant differences particularly when self-focus is incompatible with engaging in task between those who were taking psychotropic medications from those goals. This attentional control difficulty is thought to contribute to who were not (see Supplemental Material). The fact that we still found the development of rumination in those with MDD. Future studies aberrant functional connectivity within critical regions among each of would benefit from a more detailed analysis of these core neural net- the three neural networks suggests that the findings are not entirely works during tasks that vary in attentional demands and in content to explained by medication effects. determine what factors are most effective at disrupting rumination. Additionally, with regards to our task design, the external-focus condition differs from the self-focus condition on a number of other SUPPLEMENTARY DATA dimensions besides content, including level of abstractness and Supplementary data are available at SCAN. complexity. While this same paradigm has been used in a number of studies (e.g. Lyubomirsky and Nolen-Hoeksema, 1993; Nolen- Conflict of Interest Hoeksema and Morrow, 1993), the current study could have benefited None declared. from including an ‘abstract’ control condition to further understand neural substrates specific to self-focus, as was done by Cooney et al. REFERENCES (2010). Also, of note with respect to the thought induction task, we did Alexopoulos, G.S., Hoptmann, M.J., Kanellopoulos, D., Murphy, C.F., Lim, K.O., not include a means of verifying that the participants were indeed Gunning, F.M. (2012). Functional connectivity in the cognitive control network and generating the type of thoughts each condition is designed to elicit. the default mode network in late-life depression. Journal of Affective Disorders, 139, Nor did we assess the number of statements that each participant 56–65. completed. Beck, A.T., Rush, A.J., Shaw, B.F., Emery, G. (1979). Cognitive Therapy of Depression. In addition to some limitations with the design, we also chose to New York: Guilford Publications. Beevers, C.G., Clasen, P., Stice, E., Schnyer, D. (2010). Depression symptoms and cognitive focus on a small part of within-network functional connectivity in the control of emotion cues: a functional magnetic resonance imaging study. Neuroscience, default, executive and salience networks. In addition to the subgenual 167,97–103. anterior cingulate cortex, the central hub that we chose to examine Berman, M.G., Peltier, S., Nee, D.E., Kross, E., Deldin, P.J., Jonides, J. (2011). Depression, default mode network functional connectivity, the precuneus and dor- rumination and the default network. Social Cognitive and Affective Neuroscience, 6, 548–55. somedial prefrontal cortex are also critical to self-focused thought Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L. (2008). The brain’s default network: processes including rumination in both individuals with and without anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences, MDD (Johnson et al., 2009; Freton et al., 2014; Nejad et al., 2013). We 1124,1–38. also did not examine differences in connectivity between the three Carew, C.L., Miline, A.M., Tatham, E.L., MacQueen, G.M., Hall, G.B.C. (2013). Neural networks. Some studies have found differences in connectivity between systems underlying thought suppression in young women with, and at-risk, for depres- sion. Behavioral Brain Research, 257,13–24. these networks in those with and without MDD during a resting state Christoff, K., Gordon, A.M., Smalllwood, J., Smith, S.M., Schooler, J.W. (2009). Experience (Sheline et al., 2010), as well as during a self-focus task (Lemogne et al., sampling during fMRI reveals default network and executive system contributions to 2009). Given the complexity of the psychological processes being mind wandering. Proceedings of the National Academy of Sciences of the USA, 106, tapped, future research should incorporate more comprehensive 8719–24. Cooney, R.E., Joorman, J., Eugene, F., Dennis, E.L., Gotlib, I.H. (2010). Neural correlates of within and cross-network connectivity assessments. rumination in depression. Cognitive, Affective, and Behavioral Neuroscience, 10,470–8. In sum, our study provides novel evidence for dysregulated default Cox, R. (1996). AFNI: software for analysis and visualization of functional magnetic res- mode, executive and salience network connectivity in MDD partici- onance neuroimages. Computers and Biomedical Research, 29,162–73. pants during non-self-referential thought. The results suggest Cullen, K.R., Gee, D.G., Klimes-Dougan, B., et al. (2009). A preliminary study of functional that individuals with MDD are unable to adequately switch off the connectivity in comorbid adolescent depression. Neuroscience Letters, 460,227–31. [28.4.2015–12:41pm] [744–751] Paper: OP-SCAN140123 Network imbalance in depression SCAN (2015) 751 De Lissnyder, E., Koster, E.H.W., Everaert, J., Schacht, R., Van den Abeele, D., De Raedt, R. Lyubomirsky, S., Nolen-Hoeksema, S. (1993). Self-perpetuating properties of dysphoric (2012). Internal cognitive control in clinical depression: general but no emotion-specific rumination. Journal of Personality and Social Psychology, 65,339–49. impairments. Psychiatry Research, 199, 124–30. Manoliu, A., Meng, C., Brandl, F., et al. (2014). Insular dysfunction within the salience Demeyer, I., De Lissnyder, E., Koster, E.H.W., De Raedt, R. (2012). Rumination mediates network is associated with severity of symptoms and aberrant inter-network connectivity the relationship between impaired cognitive control for emotional information and in major depressive disorder. Frontiers in Human Neuroscience, 7(930), 1–17. Mason, M.F., Norton, M.I., Van Horn, J.D., et al. (2007). Wandering minds: the default depressive symptoms: a prospective study in remitted depressed adults. Behavior network and stimulus-independent thought. Science, 315, 393–5. Research and Therapy, 50, 292–7. Michalak, J., Holz, A., Teismann, T. (2011). Rumination as a predictor of relapse in mind- Dosenbach, N.U.F., Fair, D.A., Cohen, A.L., Schlaggar, B.L., Petersen, S.E. (2008). A dual- fulness-based cognitive therapy for depression. Psychology and Psychotherapy: Theory, networks architecture of top-down control. Trends in Cognitive Sciences, 12,99–105. Research, and Practice, 840,230–6. Drevets, W.C., Savitz, J., Trimble, M. (2008). The subgenual anterior cingulate cortex in Nejad, A.B., Fossati, P., Lemogne, C. (2013). Self-referential processing, rumination, and mood disorders. CNS Spectrums, 13(8), 663–81. cortical midline structures in major depression. Frontiers in Human Neuroscience, Eckert, M.A., Menon, V., Walczak, A., et al. (2009). At the heart of the ventral attention 7(666), 1–9. system: the right anterior insula. Human Brain Mapping, 30, 2530–41. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of Fales, C.L., Barch, D.M., Rundle, M.M., et al. (2008). Altered emotional interference pro- depressive episodes. Journal of Abnormal Psychology, 100,569–82. cessing in affective and cognitive-control brain circuitry in major depression. Biological Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders and mixed Psychiatry, 63, 377–84. anxiety/depressive symptoms. Journal of Abnormal Psychology, 109,504–11. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B. (1997). Structured Clinical Interview for Nolen-Hoeksema, S., Morrow, J. (1991). A prospective-study of depression and posttrau- the DSM-IV Axis I Disorders. Washington, DC: American Psychiatric Press Inc. matic stress symptoms after a natural disaster - the 1989 Loma-Prieta Earthquake. Fox, M.D., Snyder, A.Z., Vincent, J.L., Corbetta, M., Van Essen, D.C., Raichle, M.E. (2005). Journal of Personality and Social Psychology, 61,115–21. The human is intrinsically organized into dynamic, anticorrelated functional networks. Nolen-Hoeksema, S., Morrow, J. (1993). Effects of rumination and distraction on natur- Proceedings of the National Academy of Sciences of the USA, 102, 9673–8. ally-occuring depressed mood. Cognition and Emotion, 7,561–70. Franco, A.R., Pritchard, A., Calhoun, V.D., Mayer, A.R. (2009). Inter-rater and inter- Nolen-Hoeksema, S., Morrow, J., Fredrickson, B.L. (1993). Response styles and the method reliability for selecting the default mode network during data-driven analyses. duration of episodes of depressed mood. Journal of Abnormal Psychology, 102,20–8. Human Brain Mapping, 30, 2293–303. Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., Shulman, G.L. Freton, M., Lemogne, C., Delaveau, P., et al. (2014). The dark side of self-focus: brain (2001). A default mode of brain function. Proceedings of the National Academy of Sciences activity during self-focus in low and high brooders. Social Cognitive and Affective of the USA, 98, 676–82. Neuroscience, 9(11), 1808–13. Raichle, M.E., Snyder, A.Z. (2007). A default mode of brain function: a brief history of an Gohier, B., Ferracci, L., Surguladze, S.A., et al. (2009). Cognitive inhibition and working evolving idea. NeuroImage, 37, 1083–90. memory in unipolar depression. Journal of Affective Disorders, 116, 100–5. Seeley, W.W., Menon, V., Schatzberg, A.F., et al. (2007). Dissociable intrinsic connectivity Greicius, M.D., Flores, B.H., Menon, V., et al. (2007). Resting-state functional connectivity networks for salience processing and executive control. The Journal of Neuroscience, 27, in major depression: abnormal contributions from the subgenual cingulate cortex and 2349–56. thalamus. Biological Psychiatry, 62, 429–37. Sheline, Y.I., Barch, D.M., Price, J.L., et al. (2009). The default mode network and self- Hamilton, J.P., Chen, M.C., Gotlib, I.H. (2013). Neural systems approaches to understand- referential processes in depression. Proceedings of the National Academy of Sciences of the ing major depressive disorder: an intrinsic functional organization perspective. USA, 106, 1942–7. Neurobiology of Disease, 52,4–11. Sheline, Y.I., Price, J.L., Yan, Z., Mintun, M.A. (2010). Resting state functional MRI Hamilton, J.P., Etkin, A., Furman, D.J., Lemus, M.G., Johnson, R.F., Gotlib, I.H. (2012). in depression unmasks increased connectivity between networks via the dorsal nexus. Functional neuroimaging of major depressive disorder: a meta-analysis and new inte- Proceesings of the National Academy of Sciences of the USA, 107, 11020–5. gration of baseline activation and neural response data. American Journal of Psychiatry, Shenhav, A., Botvinick, M.M., Cohen, J.D. (2013). The expected value of control: an in- 169, 693–703. tegrative theory of anterior cingulate cortex function. Neuron, 79,217–40. Hamilton, J.P., Furman, D.J., Chang, C., Thomason, M.E., Dennis, E., Gotlib, I.H. (2011). Shestyuk, A.Y., Deldin, P.J. (2010). Automatic and strategic representation of the self in Default-mode and task-positive network activity in major depressive disorder: implica- major depression: trait and state abnormalities. American Journal of Psychiatry, 167, tions for adaptive and maladaptive rumination. Biological Psychiatry, 70, 327–33. 536–44. Harvey, P.O., Fossati, P., Pochon, J.B., et al. (2005). Cognitive control and brain resources Slitz, D., Hayley, S. (2012). Major depressive disorder and alterations in insular cortical in major depression: an fMRI study using the n-back task. Neuroimage, 26, 860–9. activity: a review of current functional magnetic imaging research. Frontiers in Human Hodgson, T., Chamberlain, M., Parris, B., et al. (2007). The role of the ventrolateral frontal Neuroscience, 6(323), 1–14. cortex in inhibitory oculomotor control. Brain, 130, 1525–37. Sridharan, D., Levitin, D.J., Menon, V. (2008). A critical role for the right fronto-insular Johnson, D.P., Whisman, M.A. (2013). Gender differences in rumination: a meta-analysis. cortex in switching between central-executive and default-mode networks. Proceedings of Personality and Individual Differences, 55, 367–74. the National Academy of Sciences of the USA, 105, 12569–74. Johnson, M.K., Nolen-Hoeksema, S., Mitchell, K.J., Levin, Y. (2009). Medial cortex activity, Van Dillen, L., Heslenfeld, D.J., Koole, S.L. (2009). Tuning down the emotional brain: an self-reflection and depression. Social Cognitive and Affective Neuroscience, 4, 313–27. fMRIstudy of the effects of cognitive load on the processing of affective images. Jones, N.P., Siegle, G.J., Thase, M.E. (2008). Effects of rumination and initial severity on NeuroImage, 45, 1212–19. remission to cognitive therapy for depression. Cognitive Therapy and Research, 32, Van Dillen, L.F., Koole, S.L. (2007). Clearing the mind: a working memory model of 591–604. distraction from negative feelings. Emotion, 7, 715–23. Just, N., Alloy, L.B. (1997). The response styles theory of depression: tests and an extension Vanhaudenhuyse, A., Demertzi, A., Schabus, M., et al. (2010). Two distinct neural networks of the theory. Journal of Abnormal Psychology, 106, 221–9. mediate the awareness of environment and of self. Journal of Cognitive Neuroscience, 23, Koster, E.H.W., De Lissnyder, E., Derakashan, N., De Raedt, R. (2011). Understanding 570–8. depressive rumination from a cognitive science perspective: the impaired disengagement Wang, L., LaBar, K.S., Smoski, M., et al. (2008). Prefrontal mechanisms for executive hypothesis. Clinical Psychology Review, 31, 138–45. control over emotional distraction are altered in major depression. Psychiatry Kuhn, S., Vanderhasselt, M.A., De Raedt, R., Gallinat, J. (2012). Why ruminators won’t Research: Neuroimaging, 163,143–55. stop: the structural and resting state correlates of rumination and its relation to depres- Wei, M., Qin, J., Yan, R., Li, H., Yao, Z., Qing, L. (2013). Identifying major depressive sion. Journal of Affective Disorders, 141, 352–60. disorder using hurst exponent of resting-state brain networks. Psychiatry Research: Laird, A.R., Fox, P.M., Eickhoff, S.B., et al. (2011). Behavioral interpretations of intrinsic Neuroimaging, 214, 306–12. connectivity networks. Journal of Cognitive Neuroscience, 23, 4022–37. Weissman, D.H., Roberts, K.C., Visscher, K.M., Woldorff, M.G. (2006). The neural basis of Lemogne, C., Delaveau, P., Freton, M., Guionnett, S., Fossati, P. (2012). Medial prefrontal momentary lapses in attention. Nature Neuroscience, 9,971–8. cortex and the self in major depression. Journal of Affective Disorders, 136,e1–e11. Whitmer, A.J., Gotlib, I.H. (2013). An attentional scope model of rumination. Psychological Lemogne, C., LeBastard, G., Mayberg, H., et al. (2009). In search of the depressive self: Bulletin, 139, 1036–61. extended medial prefrontal network during self-referential processing in major depres- Zhou, Y., Yu, C., Zheng, H., et al. (2010). Increased neural resources recruitment in the sion. Social Cognitive and Affective Neuroscience, 4, 305–12. intrinsic organization in major depression. Journal of Affective Disorders, 121,220–30.

Journal

Social Cognitive and Affective Neuroscience – Oxford University Press

Published: May 1, 2015

References

  • Functional connectivity in the cognitive control network and the default mode network in late-life depression
    Alexopoulos