Prefrontal cortex activation during cognitive interference in nonsuicidal self-injury

Accepted Manuscript

Prefrontal Cortex Activation During Cognitive Interference in Nonsuicidal Self-Injury M. Kathryn Dahlgren M.S. , Jill M. Hooley JM, D.Phil. , Stephanie G. Best Ph.D. , Kelly A. Sagar M.S. , Atilla Gonenc Ph.D. , Staci A. Gruber Ph.D. PII: DOI: Reference:

S0925-4927(17)30239-1 10.1016/j.pscychresns.2018.04.006 PSYN 10809

To appear in:

Psychiatry Research: Neuroimaging

Received date: Revised date: Accepted date:

18 August 2017 10 April 2018 27 April 2018

Please cite this article as: M. Kathryn Dahlgren M.S. , Jill M. Hooley JM, D.Phil. , Stephanie G. Best Ph.D. , Kelly A. Sagar M.S. , Atilla Gonenc Ph.D. , Staci A. Gruber Ph.D. , Prefrontal Cortex Activation During Cognitive Interference in Nonsuicidal Self-Injury, Psychiatry Research: Neuroimaging (2018), doi: 10.1016/j.pscychresns.2018.04.006

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

Highlights Women with NSSI have altered prefrontal activation during cognitive interference



NSSI is associated with increased cingulate cortex (CC) activation



NSSI is associated with decreased dorsolateral prefrontal cortex (DLPFC) activation



Decreased DLPFC activation correlates with poorer emotional control and

CR IP T



AC

CE

PT

ED

M

AN US

impulsivity

1

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

2

Prefrontal Cortex Activation During Cognitive Interference in Nonsuicidal Self-Injury M. Kathryn Dahlgren, M.S.a,b, Jill M. Hooley JM, D.Phil.c, Stephanie G. Best, Ph.D.d,e, Kelly A. Sagar, M.S.a,d, Atilla Gonenc, Ph.D.a,d, & Staci A. Gruber, Ph.D.a,d,*

Cognitive and Clinical Neuroimaging Core, McLean Imaging Center, McLean Hospital, 115

CR IP T

a

Mill Street, Belmont, MA 02478, USA b

c

Department of Psychology, Tufts University, Medford, MA 02155, USA

Clinical Research Laboratory, Department of Psychology, William James Hall, Harvard

d

e

AN US

University, Cambridge, MA 02138, USA

Department of Psychiatry, Harvard Medical School, Boston, MA 02215, USA

Behavioral Health Partial Program, McLean Hospital, 115 Mill Street, Belmont, MA 02478,

M

USA

ED

*Correspondence concerning this article should be addressed to Dr. Staci Gruber, Director, Cognitive and Clinical Neuroimaging Core, McLean Hospital, 115 Mill Street, Belmont, MA

CE

PT

02478, Email: [email protected], Phone: (617) 855-2762, Fax: (617) 855-3713

Keywords: Nonsuicidal self-injury (NSSI), self-harm, functional magnetic resonance imaging

AC

(fMRI), cognitive interference, Multi-Source Interference Task (MSIT), women

Abstract: 199 words; Article Body: 4,976; Figures: 3; Tables: 3; Supplemental Information: 0 Abstract

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

3

Nonsuicidal self-injury (NSSI), deliberate behavior resulting in self-inflicted damage to oneself, is common, particularly among female adolescents, and may be a form of maladaptive emotion regulation. Cognitive interference, a specific type of processing associated with inhibiting prepotent responses in favor of less automatic ones, is utilized in treatment strategies to shift

CR IP T

patients’ thoughts and behaviors away from maladaptive responses and replace them with more adaptive ones. We examined cognitive interference processing using the Multi-Source

Interference Task (MSIT) in females with NSSI behavior (n=15) and healthy control females (n=15). Functional magnetic resonance imaging (fMRI) data were collected concurrently.

AN US

Results revealed similar between-group performance on the MSIT; however, women with NSSI behavior exhibited altered patterns of neural activation during the MSIT. Specifically, the NSSI group demonstrated increased cingulate cortex (CC) and decreased dorsolateral prefrontal cortex

M

(DLPFC) activation compared to the control group. Further, within the NSSI group, DLPFC activation inversely correlated with emotional reactivity and self-reported impulsivity, suggesting

ED

that decreased DLPFC activation is associated with poorer emotional control and increased impulsivity. Taken together, these results indicate that women with NSSI behavior utilize

PT

different cortical areas during cognitive interference processing, which may have broader

CE

implications regarding the treatment efficacy of cognitive-based therapies.

AC

Keywords: Nonsuicidal self-injury (NSSI), self-harm, functional magnetic resonance imaging (fMRI), cognitive interference, Multi-Source Interference Task (MSIT), women

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

4

1. Introduction Nonsuicidal self-injury (NSSI) is deliberate behavior resulting in self-inflicted damage without suicidal intent. Although skin cutting is most common (Nock, 2009a), many engage in multiple methods of NSSI (Victor & Klonsky, 2014), including scratching, picking, burning,

CR IP T

bruising, etc. (Hooley, 2008; Nock, 2009b). NSSI is common, with 5.9% of adults (Klonsky, 2011) and up to 23% of adolescents reporting a history of NSSI (Jacobson & Gould, 2007). Adolescent females are three times more likely to engage in NSSI than males (Barrocas et al., 2012). Additionally, although NSSI occurs without suicidal intent, it is a salient risk factor for

AN US

suicidal behavior (Ribiero et al., 2016), which is especially troubling given that suicide is the 10th leading cause of death in the US (Kochanek et al., 2016). Acute negative affect (e.g. anxiety, worthlessness) typically precedes NSSI, and self-injurers report reduced negative affect

M

afterward (Klonsky, 2007) suggesting that these behaviors are somehow reinforcing and may represent maladaptive coping/emotion regulation strategies (Hooley & Franklin, 2017; Nock &

ED

Prinstein, 2004; McKenzie & Gross, 2014).

Deficits in executive function have been observed in NSSI populations including

PT

increased perseverations on the Wisconsin Card Sorting Task (WCST; Claes et al., 2015);

CE

increased risky decision-making on the Iowa Gambling Task (Claes et al., 2015; Oldershaw et al., 2009); more errors on the Stop Signal Task (Allen & Hooley, 2015; Fikke et al., 2011); and less

AC

efficient strategy scores during a Spatial Working Memory task (Fikke et al., 2011). Although several studies have not reported impaired executive function in NSSI relative to healthy control (HC) participants (Andover et al., 2011; Claes et al., 2012; Glenn & Klonsky, 2010; Janis & Nock, 2009; McCloskey et al., 2012; Ohmann et al., 2008; Vega et al., 2015), the absence of significant findings may be related to a variety of factors. For example, increased severity

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

5

(Fikke et al., 2011; Williams et al., 2015) and more recent NSSI behaviors (Oldershaw et al., 2009) are associated with greater impairment. Additionally, higher levels of impulsivity likely play a role; increased self-reported impulsivity has been well-documented in NSSI (Hamza et al., 2015), and reduced childhood response inhibition predicts NSSI behavior later in life (Meza et

CR IP T

al., 2016). However, only one study to date reported increased impulsivity as measured by both self-report and neuropsychological assessments. Claes and colleagues (2015) found that NSSI patients performed worse on the WCST relative to HC participants, and that perseverative errors positively correlated with self-reported impulsivity.

AN US

Cognitive interference, a core component of executive function associated with

attentional control and inhibitory processing, requires attention-shifting by inhibiting prepotent responses in favor of less automatic ones (Lezak et al., 2004). Aspects of this process are

M

utilized in many therapeutic strategies, including cognitive behavioral therapy (CBT), in order to shift patients’ thoughts and behaviors away from their natural maladaptive responses and replace

ED

them with more adaptive ones (Mennin & Fresco, 2014). Legris and van Reekum (2006) suggest that attentional control helps override and inhibit prepotent negative thought patterns and can act

PT

as an “emotional buffer.” As attentional control may facilitate proper emotional regulation, it is

CE

critical to clarify the pathophysiology of executive function impairment in NSSI patients. Accordingly, the current study examined cognitive interference processing using the Multi-

AC

Source Interference Task (MSIT) while collecting functional magnetic resonance imaging (fMRI) data in NSSI patients and HC participants. The MSIT assesses cognitive interference processing and activates cingulo-frontal-

parietal (CFP) attentional network circuitry (Bush et al., 2003). The CFP network is also involved in emotion regulation; specifically, the prefrontal cortex (PFC) is important for explicit

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

6

emotion regulation (e.g., reappraisal) while the cingulate cortex (CC), particularly the anterior cingulate (ACC), is associated with emotional reactivity (Etkin et al., 2015). MSIT performance and neural activation have previously been assessed in anxiety/mood disorders including general anxiety disorder (GAD; Fitzgerald et al., 2013), posttraumatic stress disorder (PTSD; Shin et al.,

CR IP T

2011), obsessive compulsive disorder (OCD; Cocchi et al., 2012; Fitzgerald et al., 2010; 2013; Yucel et al., 2007), major depressive disorder (MDD; Davey et al., 2012), and bipolar disorder (Gruber et al., 2017). Overall, findings provide evidence for increased medial cingulate activity, particularly the dorsal ACC, (Fitzgerald et al., 2010; Shin et al., 2011; Yucel et al., 2007)

AN US

decreased lateral frontal cortex activity, particularly the dorsolateral PFC (DLPFC; Fitzgerald et al., 2013); and abnormal inter-network connectivity (Cocchi et al., 2012; Davey et al., 2012; Fitzgerald et al., 2010) during MSIT interference processing.

M

Despite activation differences during the MSIT, differences in task performance are not typically observed in mood/anxiety disorders. To date, only one study demonstrated

ED

significantly lower MSIT accuracy, specifically, increased omission errors, in bipolar patients compared to HC participants (Gruber et al., 2017). Interestingly, this study reported decreased

PT

CC and increased PFC activation in bipolar patients relative to HC participants, the opposite

CE

pattern typically observed in mood/anxiety disorders. Trends have been reported for slower response times in patients with PTSD (Shin et al., 2011) and lower accuracy in patients with

AC

OCD (Yucel et al., 2007), but neither of these findings reached statistical significance. As the majority of MSIT studies report increased CC and decreased PFC activation without significant differences in task performance between clinical and non-clinical samples, researchers have suggested that MSIT-related dysfunction of the CFP network is potentially neurocompensatory (e.g., Gruber et al., 2012; Yucel et al., 2007). This neurocompensatory theory suggests that

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

7

alterations in brain activation patterns (particularly increased CC activation) may be necessary for clinical patients to complete the task as successfully as HC participants. There is a paucity of research concerning the neurobiology of NSSI behavior with scientific literature primarily concentrating on borderline personality disorder (BPD) or suicidal

CR IP T

behavior, both of which share common features with NSSI. However, a limited number of

studies have reported that individuals who engage in NSSI behaviors exhibit enhanced activation during emotional, social, and reward processing in frontal regions, including the anterior

cingulate, orbitofrontal cortex, and additional regions within the prefrontal cortex (Brown et al.,

AN US

2017; Groschwitz et al., 2016; Osuch et al., 2014; Plener et al., 2012; Vega et al., 2018). In

addition, both the CC and DLPFC have been implicated in emotional dysregulation, and suicidal planning in patients with BPD or a history of suicidal behavior suggesting that these regions of

M

interest (ROIs) may also be relevant to the pathophysiology of NSSI (reviewed in Tatnell & Hasking, 2015).

ED

A single previous study examined task performance using the MSIT in patients with NSSI and HCs, and reported no between-group performance differences (Allen & Hooley, 2017).

PT

As no studies thus far have examined brain activation patterns in patients with NSSI during tasks

CE

requiring cognitive control, the current investigation acquired fMRI data from NSSI and HC participants during the completion of the MSIT. Given previous findings from samples of

AC

patients with mood and anxiety disorders, we hypothesized that NSSI patients would exhibit increased CC and decreased DLPFC activation relative to HC participants, and that neural activation alterations would be observed without significant, between-group performance differences. Further, as increased impulsivity has been well-documented in NSSI participants (e.g. Hamza et al., 2015) and NSSI behavior can be employed as a maladaptive emotion

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

8

regulation strategy, (e.g. Hooley & Franklin, 2017; Nock & Prinstein, 2004; McKenzie & Gross, 2014) we explored the relationship between MSIT-related neural activation and self-reported measures of emotional reactivity and impulsivity. 2. Methods and materials

CR IP T

2.1. Participants

Fifteen women reporting current skin-cutting behaviors (≥10 lifetime episodes) and

fifteen HC women without NSSI were recruited from the Greater Boston area via flyers and electronic postings at local colleges/universities, McLean Hospital, and on Craigslist. Control

AN US

participants were primarily recruited from a pool generated from ongoing studies at McLean Hospital. Forty-four potential NSSI patients referred themselves for participation by contacting study staff. Prior to participation, patients completed a brief phone screen to determine

M

eligibility for the study including demographic information, MRI contraindications, NSSI behavior, and psychiatric symptoms. A total of 37 participants (HC=20, NSSI=17) met

ED

eligibility criteria and were invited to complete the experimental session; however, 3 failed to respond to scheduling calls and were considered lost to follow-up; 3 were determined to be

PT

ineligible during the study visit due to significant marijuana use; and 1 decided she no longer

CE

wanted to participate due to time constraints. All participants were right-handed women aged between 18-31 years. Exclusion criteria

AC

included head injury with loss of consciousness (≥10 min); history of medical illness affecting cognition; neurological disorders; non-native English speaking (required for the assessments); MRI contraindications (e.g., claustrophobia, pregnancy); and, for HC participants, any current psychiatric diagnosis. Additionally, all participants completed the Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 1999) to assess whether the clinical and control groups were

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

9

matched for general intelligence (IQ). This study was approved by the McLean Hospital Institutional Review Board (IRB), and prior to participation, participants reviewed and signed an informed consent form. Study procedures were fully explained, including risks and benefits of participation and the voluntary nature of the study.

CR IP T

2.2. Assessments and procedures 2.2.1. Clinical Interview Measures

All participants were administered the Structured Clinical Interview for DSM-IV

Disorders I (First et al., 2002) & II (First et al., 1997). The SCID is the most extensively-

AN US

validated clinical interview assessing psychiatric and personality disorders with high test-retest reliability (κ=0.65-0.68; Zanarini et al., 2000) and inter-rater reliability (κ=0.71-0.84; Lobbestael et al., 2011). Within the NSSI group, current disorders included: mood disorders (n=12;

M

80.00%), anxiety disorders (n=8; 53.33%), eating disorders (n=1; 6.67%), and alcohol dependence (n=1; 6.67%). Additionally, 13 NSSI participants (86.67%) met diagnostic criteria

ED

for BPD, and 4 NSSI participants (26.67%) reported a history of at least one suicide attempt. None of the HC participants had any psychiatric disorders. Additionally, participants with NSSI

PT

also completed the Self-Injurious Thoughts and Behaviors Interview (SITBI), a 169-item clinical

CE

interview that evaluates the prevalence, frequency, and severity of five types of self-injurious behaviors: suicidal ideation, suicide plans, suicide gestures, suicide attempts, and NSSI (Nock et

AC

al., 2007). The SITBI assesses lifetime, past year, past month, and past week prevalence as well as age of onset of each thought/behavior. The SITBI has high test-retest reliability (κ=1.00), inter-rater reliability (κ=1.00), and inter-informant agreement (κ=0.91) for NSSI, as well as strong construct validity when compared to the Functional Assessment of Self-Mutilation scale (κ=1.00 for presence of NSSI and all rs≥.64 across comparisons of subscales).

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

10

2.2.2. Clinical Self-Report Measures Participants completed a comprehensive battery of clinical rating scales comprised of well-validated measures designed to assess mood, emotional reactivity, and impulsivity. The Beck Depression Inventory (BDI) is a reliable (α=0.86; Beck et al., 1988) scale assessing

CR IP T

severity of depression over the past 2 weeks, and includes 21 items rated from 0-3 according to intensity of symptoms (Beck et al., 1961). The State-Trait Anxiety Inventory (STAI) is a 22item scale (α=0.86-0.95) that measures current anxiety levels (state) and general anxiety level (trait); items have a range of four response options depending on frequency of symptoms

AN US

(Spielberger et al., 1983). The short form of the Mood and Anxiety Symptom Questionnaire (MASQ) is a 62-item assessment (α=0.76; Watson et al., 1995) that provides measures of general distress from depression and anxiety-based symptoms as well as anxious arousal and anhedonic

M

depression; items are rated from 1-5 based on how much each symptom was experienced in the past week (Watson & Clark, 1991). The Positive and Negative Affect Schedule (PANAS) is a

ED

20-item scale (α=0.84-0.90) that assesses positive affect associated with pleasurable engagement and negative affect associated with arousing aversive states; items are rated from 1-5 depending

PT

on how participants generally feel (Watson et al., 1988). The Profile of Mood States (POMS) is

CE

a 65-item measurement (α=0.63-0.96) of current mood state for the individual domains of vigor, anger, confusion, tension, and depression, and a composite measure of total mood disturbance

AC

(TMD); items are rated from 0-4 based on how participants generally feel (McNair et al., 1971). The Emotion Reactivity Scale (ERS) is a 21-item measure (α=0.94) that evaluates the extent to which participants experience emotions on a regular basis, and contains three subscales regarding how emotion is experienced: sensitivity, arousal/intensity, and persistence (Nock et al., 2008). Items are rated from 0-4 depending on how much the participant feels the statement is

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

11

“like me.” The UPPS Impulsive Behavior Scale (UPPS; Whiteside & Lynam, 2001) is a 59-item measurement (α=0.82-0.91) assessing impulsivity using five subscales: lack of premeditation, lack of perseverance, negative urgency, positive urgency, and sensation seeking; items are rated

2.2.3. The Multi-Source Interference Task (MSIT)

CR IP T

1-4 from strongly disagree to strongly agree.

While undergoing fMRI scanning, participants completed the MSIT (Bush & Shin, 2006), which reliably and robustly activates the CFP network using two types of cognitive interference, spatial and flanker (Bush et al., 2003; Bush & Shin, 2006). During the MSIT (Figure 1), three-

AN US

digit stimuli sets (comprised of numbers 1, 2, 3, or 0) are presented briefly on a screen. Each stimulus set contains two identical numbers (distractors) and one target number, which is always different from the other two. Participants are instructed to report the identity of the target

M

number regardless of its position using a button box. During the control condition, distractor numbers are zeros and the position of the target number is congruent to its corresponding

ED

position on the button box (i.e., 100, 020, or 003). During the interference condition, distractors are numbers other than zero, and the position of the target number is incongruent to its button

PT

box position (e.g., 211, 232, 331, etc.). Immediately before scanning, participants completed a

CE

practice MSIT, and scanning did not commence until participants correctly completed at least three consecutive practice trials. Task parameters were identical to those described and utilized

AC

in previous studies (e.g., Gruber et al., 2012; 2017). MSIT performance was recorded for both task conditions, and a derived contrast

(interference-control) was calculated to assess the specific contribution of the interference condition. Dependent variables included average response time for correct responses (ms), percent accuracy, and errors. Errors were subdivided into two types: 1) omission errors (no

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

12

response), which reflect slower or overloaded cognitive processing; and 2) commission errors (incorrect responses), which indicate difficulty inhibiting inappropriate responses. 2.3. Statistical methods and analyses Two-tailed, univariate analyses of variance (ANOVAs) were used to compare the two

CR IP T

groups on demographic variables as well as MSIT performance variables. As the NSSI group was expected to have more severe clinical symptomatology than the HC group, one-tailed

ANOVAs were used for those analyses. All MSIT performance data was screened for outliers (

̅

; subsequently, performance data from one HC was excluded. In order to ensure

AN US

results were not influenced by this outlier, fMRI analyses were completed both including and excluding data from this subject which did not significantly impact the results. 2.3.1. Imaging methods

M

Imaging was performed on a Siemens Trio whole body 3T MRI scanner (Siemens Corporation, Erlangen, Germany) using a quadrature RF head coil; 40 contiguous coronal slices

ED

were acquired, providing whole brain coverage (5mm, 0mm skip). Images were collected every 3s using a single shot, 64x64 acquisition matrix, gradient pulse echo sequence (TR=3000ms,

PT

TE=30ms, flip angle=90, FOV=20cm) in plane resolution 3.125x3.125x3.125mm3, equaling total

CE

of 132 images per slice.

Functional MRI images were analyzed using Statistical Parametric Mapping (SPM8,

AC

version 4290, Wellcome Department of Imaging Neuroscience, University College, London, UK). First, blood oxygen level dependent (BOLD) fMRI data were corrected for motion using a 2-step, intra-run realignment algorithm, which used the mean image created after the first realignment as a reference. Root mean square of x, y, and z translation was calculated and ≥3mm of head motion was exclusionary. Motion was further assessed as frame-wise

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

13

displacement, root mean square of translation, and the absolute euler angle; t-tests indicated that none of these motion variables were significantly different between groups (all ps≥.58). Additionally, signal to noise ratios (SNR) were calculated for each participant, and SNR below 40 was exclusionary. Realigned images were then normalized in Montreal Neurological Institute

CR IP T

stereotactic space, resampled into 2mm3 voxels, and spatially smoothed using an isotropic

Gaussian kernel (8mm full width at half maximum) without global scaling. High-pass temporal filtering (cutoff=128s) was applied, and serial autocorrelations were modeled with SPM8’s AR(1) model.

AN US

Statistical parametric images were calculated individually for each subject, using a

general linear model that accounted for task-related changes, with each condition modeled as a block design with a boxcar waveform. At the first level, three regressors were fit to the data:

M

baseline fixation, control condition, and interference condition. Activation was averaged across condition blocks without adjustments for individual item performances. Direct contrast images

ED

(interference-control) were calculated for each subject, entered into second level models, and subjected to voxel-wise, 1-sample t-tests. Then the NSSI and HC groups were compared using

PT

between-group t-tests. Analyses were performed including the movement parameters as

CE

regressors. ROI masks (CC and DLPFC) were created using the Wake Forest University PickAtlas utility (Maldijan et al., 2003). The statistical threshold was set at uncorrected p≤0.01

AC

and a minimum cluster extent k≥10 contiguous voxels. 2.3.2. Correlation Analyses: fMRI Activation with Emotional Reactivity and Impulsivity Individual participants’ fMRI activation values from the CC and DLPFC were calculated

as voxel counts extracted using the MarsBaR SPM toolbox (Brett et al., 2002). Pearson

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

14

correlations (2-tailed) were performed to examine the associations between individual fMRI activation and measures of emotional reactivity and impulsivity. 3. Results 3.1. Demographics & clinical state

CR IP T

The NSSI and HC groups were well-matched for age. While a trend emerged for the HC group to have slightly more years of education (p=.063), IQ did not differ significantly between the groups (Table 1). All NSSI participants reported skin cutting as their primary mode of NSSI with an average of 6.00 years (SD=3.91) of NSSI behaviors and 124.09 (SD=118.74) lifetime

AN US

NSSI episodes. A total of 6 NSSI patients (40.00%) reported receiving outpatient therapy within the past month, while the remaining 9 (60.00%) had not received any past month psychiatric care. Seven of the NSSI patients (46.67%) reported at least one previous psychiatric hospitalization; of

M

those patients, the average number of hospitalizations was 1.71 (SD=0.95). With regard to clinical state, the NSSI group reported significantly greater mood disturbance than the HC group

ED

across all rating scales (Table 1). Similar results were noted for emotion regulation, with the NSSI group reporting more maladaptive emotion reactivity compared to the HC group (Table 1).

PT

NSSI participants endorsed significantly higher levels of self-reported impulsivity than the HC

CE

group; however, the UPPS sensation seeking subscale was the only clinical rating on which the groups did not significantly differ (Table 1).

AC

3.2. MSIT performance & neuroimaging The NSSI and HC group demonstrated similar performance on the control and

interference conditions of the MSIT as well as the interference-control contrast, with no significant differences noted for any variable (Table 2).

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

15

FMRI analyses indicated that despite similar MSIT performance, the groups exhibited significantly different patterns of brain activation for the interference-control contrast. Single sample t-test analyses revealed robust activation of the CC and DLPFC in both groups. The HC group had three different local maxima within the CC (k=336) and six different local maxima

CR IP T

within the DLPFC (k=667) activated above threshold measures (Table 3, Figure 2A). The NSSI group had two different local maxima within the CC (k=1181) and one local maxima within the DLPFC (k=113) activated above threshold measures (Table 3, Figure 2A).

Within the CC, between-group t-tests revealed three local maxima (k=77) with the NSSI

AN US

group showing more activation relative to the HC group (Table 3, Figure 2B); no local maxima revealed greater activation in the HC relative to the NSSI group. Within the DLPFC, one local maxima (k=10) emerged in which the NSSI group activated more than the HC group, and two

M

local maxima (k=172) emerged in which the HC group activated more than the NSSI group (Table 3, Figure 2B). These results demonstrate a clear pattern of greater CC activation coupled

ED

with decreased DLPFC activation in the NSSI group relative to the HC group during the interference-control contrast of the MSIT.

PT

3.2.1. Correlation Results: fMRI Activation with Emotional Reactivity and Impulsivity

CE

Within the NSSI group, emotional reactivity measured by the ERS inversely correlated with DLPFC activation. Specifically, higher Emotional Sensitivity ratings (r(12)=-.620, p=.018)

AC

and Total Emotional Reactivity scores (r(12)=-.569, p=.034) were associated with less DLPFC activation (Figure 3A). Additionally, a trend was noted between higher Emotional Persistence ratings and lower DLPFC activation (r(12)=-.463, p=.096). Impulsivity measured by the UPPS inversely correlated with DLPFC activation during the MSIT; Negative Urgency scores were inversely correlated with DLPFC activation (r(13)=-.539, p=.038), and a trend was noted for

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

16

Lack of Perseverance ratings to negatively correlate with DLPFC activation (r(13)=-.481, p=.070; Figure 3B). Within the HC sample, ERS and UPPS scores did not significantly correlate with DLPFC activation. CC activation did not significantly correlate with emotional

4. Discussion

CR IP T

reactivity or impulsivity in either sample.

As hypothesized, NSSI patients demonstrated altered CFP attentional network activation, specifically increased CC and decreased DLPFC activation, during the MSIT relative to HC participants. These findings are consistent with previous MSIT neuroimaging studies in anxiety

AN US

disorders, which have reported increased medial CFP network activation, particularly in the dorsal ACC, and provide some evidence for decreased lateral frontal CFP network activation in the DLPFC (Cocchi et al., 2012; Fitzgerald et al., 2010; 2013; Shin et al., 2011; Yucel et al.,

M

2007). Findings are also consistent with previous studies of NSSI patients reporting increased activation within frontal circuitry related to social, emotional and reward processing (Brown et

ED

al., 2017; Groschwitz et al., 2016; Osuch et al., 2014; Plener et al., 2012; Vega et al., 2018). With regard to MSIT performance, similar to previous MSIT studies of NSSI (Allen & Hooley,

PT

2017) and mood/anxiety disorders (Cocchi et al., 2012; Davey et al., 2012; Fitzgerald et al.,

CE

2010; 2013; Shin et al., 2011; Yucel et al., 2007), the current study found no evidence of significant between-group differences, providing further support to the neurocompensatory

AC

theory, which posits that increased neural activation is necessary for clinical populations to successfully complete the MSIT. Additionally, the current study found significantly greater mood disturbance, maladaptive emotional reactivity, and impulsivity levels in the NSSI group relative to the HC group. Interestingly, decreased DLPFC activation during the MSIT

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

17

significantly correlated with increased self-reported impulsivity and emotional reactivity in NSSI patients only; no such correlations were observed in the HC sample. Despite evidence that NSSI patients have significantly higher levels of self-reported impulsivity compared to HC participants (Hamza et al., 2015), most NSSI research assessing

CR IP T

behavioral impulsivity with cognitive tasks has yielded non-significant results (Claes et al., 2012; Glenn & Klonsky, 2010; Janis and Nock, 2009; McCloskey et al., 2012). In fact, only Claes and colleagues (2015) reported increased impulsivity as measured by both self-report and cognitive assessments. However, a meta-analysis indicated that self-report measures of impulsivity tend to

AN US

only weakly correlate with cognitive measures (Cyders & Coskunpinar, 2011), suggesting that these assessments may not measure the same constructs of impulsivity. McCloskey and colleagues (McCloskey et al., 2012) posit that self-report scales measure self-perceived overall

M

impulsivity, while cognitive measures assess impulsivity/executive functioning related to individual task demands at specific time points, which may not account for emotional stressors

ED

that drive impulsive behaviors outside of laboratory environments. These findings highlight the importance of utilizing multimodal research strategies to assess impulsivity.

PT

With regard to the neuroimaging results, Dosenbach and colleagues (2007) proposed that

CE

the CFP attentional network is composed of two synergistic sub-networks: the cingulo-opercular and frontoparietal systems. The cingulo-opercular system is comprised of medial regions of the

AC

cingulate and subcortex, particularly the ACC, and is associated with overall maintenance of task performance and set maintenance. Interestingly, previous research has observed that increased dorsal ACC activation during the MSIT is correlated with poorer task performance and longer response times in OCD, suggesting that increased ACC activation is associated with more difficulty successfully completing the task (Fitzgerald et al., 2010; Yucel et al., 2007). The

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

18

frontoparietal system is comprised of lateral regions of the frontal and parietal cortices, particularly the DLPFC, and is associated with moment-to-moment task processing and setshifting. Accordingly, one potential interpretation of the current fMRI findings is that NSSI may be linked to impaired moment-to-moment processing during the MSIT, as indicated by reduced

CR IP T

DLPFC activation, and therefore greater activation of the CC is necessary to maintain task

performance. However, additional research is necessary to clarify the specific impact of the observed fMRI alterations in patients with NSSI.

Further, dorsal ACC activation during the MSIT positively correlated with clinical

AN US

symptoms in patients with PTSD (Shin et al., 2011). While the current study did not report any significant correlations between CC activation and clinical symptomatology, both emotional reactivity and impulsivity measurements inversely correlated with DLFPC activation in NSSI

M

patients but not HC participants. Fitzgerald and colleagues (2013) suggested that decreased DLPFC activation in OCD patients may reflect impairment of the capacity to adjust repetitive

ED

anxious thoughts. In the current study, decreased DLPFC activation associated with poorer emotional reactivity and increased impulsivity may help drive NSSI behaviors.

PT

Numerous studies using other cognitive interference tasks in clinical and HC samples

CE

have demonstrated altered CC activation during non-emotional tasks of cognitive interference (Bush et al., 1998; 2000). Previous findings suggest that increased CC activation during

AC

cognitive interference is associated with higher anxiety. For example, PTSD patients showed increased dorsal ACC activation on the Stroop compared to HC participants, and clinical improvement was significantly associated with reduced ACC activation (Thomaes et al., 2012). Further, Comte and colleagues (Comte et al., 2015) recently found that increased ACC activation during cognitive interference processing significantly correlated with increased ratings of self-

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

19

reported anxiety in HC participants, and connectivity between the ACC and DLPFC inversely correlated with measures of anxiety. The authors suggest that poor clinical state results in inefficient higher-order cognitive control, characterized by impaired connectivity between the ACC and DLPFC, resulting in increased ACC and decreased DLPFC activation (Comte et al.,

CR IP T

2015). These finds are consistent with the current results, and provide evidence for a specific pattern of increased CC and decreased DLPFC activation during cognitive interference processing in psychiatric samples, which is related to higher clinical symptomatology. 4.1. Limitations and future directions

AN US

Results should be interpreted in the context of several limitations. As females are three times more likely to engage in NSSI than males (Barrocas et al., 2012), the current study only included female participants. However, it is possible that sex differences may exist among NSSI

M

patients. Additionally, participants referred themselves for this study, which may have biased the sample; however this decision was made to facilitate recruitment and, in the case of NSSI

ED

participants, ensure voluntary participation without pressure from treatment providers. Further, the current study was comprised of young adults, but a broader age range may yield different

PT

results. Furthermore, the current study had modest sample sizes, which reduces statistical power

CE

to detect small effect sizes; however similar sample sizes are typical for neuroimaging studies (e.g., Cocchi et al., 2012; Fitzgerald et al., 2010; Shin et al., 2011), and smaller sample sizes

AC

have previously been reported in neuroimaging studies using the MSIT (e.g., Ikuta et al., 2012). NSSI is associated with several psychiatric disorders including mood and anxiety

disorders, therefore entry criteria for the current study did not include diagnosis of a specific psychiatric disorder. However, NSSI behavior may be related to severity of psychiatric symptomatology (e.g., increased suicidal behavior; Ribiero et al., 2016), which may mediate the

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

20

observed results. Future studies should further clarify the unique impact of NSSI behavior in clinical samples by including a non-NSSI, psychiatrically-matched control group. Additionally, previous research has indicated that medication, particularly antipsychotics, may impair MSIT performance (Ikuta et al., 2014) and may alter CFP network activation (e.g., Strakowski et al.,

CR IP T

2005). In the current study, only five of the NSSI participants reported current psychiatric

medication use (antidepressants: n=5; antipsychotics: n=2; anxiolytics: n=2; stimulants: n=1). As relatively few participants reported use of psychiatric medication, it was not controlled for in subsequent analyses. However, the single study demonstrating medication effects during the

AN US

MSIT reported impaired performance associated with antipsychotic treatment (Ikuta et al., 2014). Given the lack of significant, between-group performance differences in the current study, it is unlikely that the five NSSI participants taking psychiatric medications significantly impacted

M

overall group performance. Previous studies examining the impact of psychiatric medication report attenuated activation of the CFP network (e.g., Bell et al., 2005). While this may have

ED

mediated some of the decreased DLPFC activation observed in the NSSI group, we report

medication.

PT

increased (not attenuated) CC activation, minimizing concern regarding the impact of psychiatric

CE

Interestingly, the only previous study to examine the MSIT in NSSI and HC participants reported no significant between-group performance differences at baseline or following negative

AC

mood induction (Allen & Hooley, 2017); the authors suggested that increased negative mood did not correspond with increased behavioral impulsivity in NSSI. This interpretation is consistent with previous studies reporting no significant MSIT performance differences between HC and clinical samples (Cocchi et al., 2012; Davey et al., 2012; Fitzgerald et al., 2010; 2013; Shin et al., 2011; Yucel et al., 2007). However, as the MSIT was specifically designed as a neuroimaging

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

21

paradigm to assess CFP attentional network activation (Bush et al., 2003), data from imaging studies may provide a more sensitive and specific measure of cognitive interference processing during the MSIT. Future studies should further examine the effect of negative mood on MSIT neural activation patterns in NSSI patients.

CR IP T

Finally, data from the current study underscores the importance of identifying

pathophysiological processes associated with cognitive processing in NSSI. Future studies

should examine the potential impact that impaired cognitive function may have on efficacy of NSSI treatment. A recent review found that although standard cognitive-based therapies for

AN US

NSSI (e.g., CBT) were associated with decreased NSSI behavior, the impact of these

interventions was comparable to treatment as usual, suggesting that adding specialized treatment interventions to traditional therapy did not improve NSSI (Gonzalez & Bergstrom, 2013).

M

Interestingly, some research has suggested that neural activation patterns during executive function tasks may predict treatment efficacy. For example, increased medial frontal gyrus

ED

activation and decreased inferior frontal triangle activation during WCST were associated with improved CBT outcomes in elderly, depressed patients; these results were observed even though

PT

WCST performance itself was not a significant predictor of treatment efficacy (Thompson et al.,

CE

2015). Similarly, patients with substance use disorders exhibited decreased ACC and inferior frontal gyrus activation during the Stroop after CBT relative to pre-treatment activation levels

AC

(DeVito et al., 2012). Future research should continue to explore the impact of cognitive function on cognitive-based therapies for NSSI using functional imaging, as neural activation patterns may be sensitive to these treatment strategies. 4.2. Conclusions

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

22

Results from the current study demonstrate that, despite similar between-group task performance, NSSI patients demonstrated significantly increased CC activation and significantly decreased DLPFC activation during the MSIT relative to HC participants. Further, DLPFC activation inversely correlated with emotional reactivity and impulsivity within the NSSI sample,

CR IP T

suggesting that decreased DLPFC activation was related to poorer emotional control and

increased impulsivity. These results provide evidence that women who engage in NSSI utilize different neural circuitry during cognitive interference processing compared to healthy

AC

CE

PT

ED

M

AN US

individuals, which may be related to impaired impulse control.

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

23

Acknowledgements Support for this project was provided by a grant from the Harvard University William F. Milton Fund awarded to Dr. Hooley. The authors would like to thank Rosemary Smith, Ashley Lambros, and Madeline Kuppe for their assistance in preparing this manuscript. Preliminary

MA.

Contributors

CR IP T

findings from this study were previously reported as a poster at McLean Research Day, Belmont,

AN US

JH and SG developed the study concept with input from MKD and SB. Clinical interviews and neurocognitive testing were primarily performed by SB with some assistance from JH and MKD. Neuroimaging data were primarily acquired by MKD with some assistance from JH and SB. AG

M

performed the neuroimaging analyses. MKD performed the all other statistical analyses and interpreted the data under the supervision of SG. MKD wrote the manuscript, and JH, KS, and

PT

ED

SG provided critical revisions.

Conflicts of Interest

CE

All authors report no direct or indirect financial or personal relationships, interests, or affiliations relevant to the subject matter of the manuscript. All authors report no biomedical financial

AC

interests or potential conflicts of interest.

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

24

References Allen, K.J.D., Hooley, J.M. (2015). Inhibitory control in people who self-injure: Evidence for impairment and enhancement. Psychiatry Res. 225, 631-637. Allen, K.J.D., Hooley, J.M., 2017. Negative mood and interference control in non-suicidal self-

CR IP T

injury. Compr. Psychiatry. 73, 35-42. doi:10.1016/j.comppsych.2016.10.011

Andover, M.S., Schatten, H.T., Crossman, D.M., Donovick, P.J., 2011. Neuropsychological functioning in prisoners with and without self-injurious behaviors: implications for the

doi:10.1177/0093854811420670

AN US

criminal justice system. Crim. Justice Behav. 38, 1103-1114.

Barrocas, A.L., Hankin, B.L., Young, J.F., Abela, J.R.Z., 2012. Rates of nonsuicidal self-injury in youth: age, sex, and behavioral methods in a community sample. Pediatr. 130, 39-45.

M

doi: 10.1542/peds.2011-2094

Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., Erbaugh, J., 1961. An inventory for

ED

measuring depression. Arch. Gen. Psychiatr. 4, 53-63. doi:10.1001/archpsyc.1961.01710120031004

PT

Beck, A. T., Steer, R. A., Garbin, M. G., 1988. Psychometric properties of the Beck Depression

CE

Inventory: Twenty-five years of evaluation. Clin. Psychol. Rev. 8, 77-100. doi: 10.1016/0272-7358(88)90050-5

AC

Bell, E.C., Willson, M.C., Wilman, A.H., Dave, S., Silverstone, P.H., 2005. Differential effects of chronic lithium and valproate on brain activation in healthy volunteers. Hum. Psychopharmacol. 20, 415–424. doi:10.1002/hup.710

Brett, M., Anton, J.L., Valabregue, R., Poline, J.B., 2002. Region of interest analysis using an SPM toolbox. Presented at: 8th International Conference on Functional Mapping of the

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

25

Human Brain; June 2-6, 2002; Sendai, Japan. Available on CD-ROM in Neuroimage. 16(2). Brown, R.C., Plener, P.L., Groen, G., Neff, D., Bonenberger, M., Abler, B., 2017. Differential Neural Processing of Social Exclusion and Inclusion in Adolescents with Non-Suicidal

CR IP T

Self-Injury and Young Adults with Borderline Personality Disorder. Front. Psychiatry. 8, 267. doi:10.3389/fpsyt.2017.00267

Bush, G., Shin, L.M., 2006. The multi-source interference task: an fMRI task that reliability activates the cingulo-frontal-parietal cognitive/attention network. Nat. Protoc. 1, 308-313.

AN US

doi:10.1038/nprot.2006.48

Bush, G., Shin, L.M., Holmes, J., Rosen, B.R., Vogt, B.A., 2003. The multi-source interference task: validation study with fMRI in individual subjects. Mol. Psychiatry. 8, 60-70.

M

doi:10.1038/sj.mp.4001217

Bush, G., Luu, P., Posner, M.I., 2000. Cognitive and emotional influences in anterior cingulate

ED

cortex. Trends Cogn. Sci. 4, 215-222. doi:10.1016/S1364-6613(00)01483-2 Bush, G., Whalen, P.J., Rosen, B.R., Jenike, M.A., McInerney, S.C., Rauch, S.L., 1998. The

PT

counting stroop: An interference task specialized for functional neuroimaging--validation

CE

study with functional MRI. Hum. Brain Mapp, 6, 270-282. doi:10.1002/(SICI)10970193(1998)6:4<270::AID-HBM6>3.0.CO;2-0

AC

Claes, L., Fagundo, A.B., Jiménez-Murcia, S., Agüera, Z., Giner-Bartolome, C., Granero, R., et al., 2015. Is Non-suicidal self-injury related to impulsivity in anorexia nervosa? Results from self-report and performance-based tasks. Eur. Eat. Disord. Rev. 23, 28-33. doi:10.1002/erv.2329

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

26

Claes, L., Van den Eynde, F., Guillaume, S., Vogels, C., Audenaert, K., 2012. Executive functioning in borderline personality disorder with and without self-harming behaviors. Adv. Psychol. Res. 89, 177–189. Cocchi, L., Harrison, B.J., Pujol, J., Harding, I.H., Fornito, A., Pantelis, C., et al., 2012.

CR IP T

Functional alterations of large-scale brain networks related to cognitive control in obsessive-compulsive disorder. Hum. Brain Mapp. 33, 1089-1106. doi:10.1002/hbm.21270

Comte, M., Cancel, A., Coull, J.T., Schön, D., Reynaud, E., Boukezzi, S., et al., 2015. Effect of

AN US

trait anxiety on prefrontal control mechanisms during emotional conflict. Hum. Brain. Mapp. 36, 2207-2214. doi:10.1002/hbm.22765

Cyders, M.A., Coskunpinar, A., 2011. Measurement of constructs using self-report and

M

behavioral lab tasks: Is there overlap in nomothetic span and construct representation for impulsivity? Clin. Psychol. Rev. 31, 965-982. doi:10.1016/j.cpr.2011.06.001

ED

Davey, C.G., Yucel, M., Allen, N.B., Harrison, B.J., 2012. Task-related deactivation and functional connectivity of the subgenual cingulate cortex in major depressive disorder.

PT

Front. Psychiatry. 3. doi:10.3389/fpsyt.2012.00014.

CE

DeVito, E.E., Worhunsky, P.D., Carroll, K.M., Rounsaville, B.J., Kober, H., Potenza, M.N., 2012. A preliminary study of the neural effects of behavioral therapy for substance use

AC

disorders. Drug Alcohol Depend. 122, 228-235. doi:10.1016/j.drugalcdep.2011.10.002 Dosenbach, R.A.T., Dosenbach, N.U.F., Fair, D.A., Miezin, F. M., Cohen, A.L., Wenger, K. K., et al., 2007. Distinct brain networks for adaptive and stable task control in humans. Proc. Natl. Acad. Sci. USA. 104, 11073-11078. doi:10.1073/pnas.0704320104

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

27

Etkin, A., Büchel, C., Gross, J.J., 2015. The neural bases of emotion regulation. Nature Rev. Neurosci. 16, 693-700. doi:10.1038/nrn4044 Fikke, L.T., Melinder, A., Landrø, N.I., 2011. Executive functions are impaired in adolescents engaging in non-suicidal self-injury. Psychol. Med. 41, 601-610.

CR IP T

doi:10.1017/S0033291710001030

First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2002. Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition (SCID-I/P). Biometrics Research, New York State Psychiatric Institute. New York, NY.

AN US

First, M.B., Gibbon, M., Spitzer, R.L., Williams, J.B.V., Benjamin, L.S., 1997. Structured clinical interview for DSM-IV axis II personality disorders (SCID-II). American Psychiatric Press, Inc. Washington, DC.

M

Fitzgerald, K.D., Liu, Y., Stern E.R., Welsh, R.C. Hanna, G.L., Monk, C.S., et al., 2013. Reduced error-related activation of dorsolateral prefrontal cortex across pediatric anxiety

ED

disorders. J. Am. Acad. Chil. Adolesc. Psychiatry. 52, 1183-1191. doi:10.1016/j.jaac.2013.09.002

PT

Fitzgerald, K.D., Stern, E.R., Angstadt, M., Nicholson-Muth, K.C., Maynor, M.R., Welsh, R.C.,

CE

et al., 2010. Altered function and connectivity of the medial frontal cortex in pediatric obsessive-compulsive disorder. Biol. Psychiatry. 68, 1039-1047.

AC

doi:10.1016/j.biopsych.2010.08.018 Glenn, C.R., Klonsky, E.D., 2010. A multimethod analysis of impulsivity in nonsuicidal selfinjury. Personal. Disord. 1, 67-75. doi:10.1037/a0017427

Gonzalez, A.H., Bergstrom, L., 2013. Adolescent non-suicidal self-injury (nssi) interventions. J. Child Adolesc. Psychiatr. Nurs. 26, 124-130. doi:10.1111/jcap.12035

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

28

Groschwitz, R.C., Plener, P.L., Groen, G., Bonenberger, M., Abler, B., 2016. Differential neural processing of social exclusion in adolescents with non-suicidal self-injury: An fMRI study. Psychiatry Res. 255, 43-49. doi:10.1016/j.pscychresns.2016.08.001 Gruber S., Dahlgren M.K., Sagar K., Gönenc A., Killgore W., 2012. Age of onset of marijuana

CR IP T

use impacts inhibitory processing. Neurosci. Lett. 511, 89-94. doi:10.1016/j.neulet.2012.01.039

Gruber, S.A., Dahlgren, M.K., Sagar, K.A., Gonenc, A., Norris, L., Cohen, B.M., et al., 2017. Decreased cingulate cortex activation during cognitive control processing in bipolar

AN US

disorder. J. Affect. Disord. 213, 86-95. doi:10.1016/j.jad.2017.02.003

Hamza, C.A., Willoughby, T., Heffer, T., 2015. Impulsivity and nonsuicidal self-injury: A review and meta-analysis. Clin. Psychol. Rev. 38, 13-24. doi:10.1016/j.cpr.2015.02.010

M

Hooley, J.M., 2008. Self-harming behavior: introduction to the special series on non-suicidal self-injury. Appl. Prev. Psychol. 12, 155-158. doi:10.1016/j.appsy.2008.08.001

ED

Hooley, J.M., & Franklin, J.C. (2017). Why do people hurt themselves? A new conceptual model of nonsuicidal self-injury. Clinical Psychol Sci. doi:10.1177/

PT

216770261774564102617745641ttps://doi.org/10.1177/2167702617745641

CE

Ikuta, T., Robinson, D.G., Gallego, J.A., Peters, B.D., Gruner, P., Kane, J., et al., 2014. Subcortical modulation of attentional control by second-generation antipsychotics in

AC

first-episode psychosis. Psychiatry Res. 221, 127-134. doi:10.1016/j.pscychresns.2013.09.010

Ikuta, T., Szeszko, P.R., Gruner, P., DeRosse, P., Gallego, J., Malhotra, A.K., 2012. Abnormal anterior cingulate cortex activity predicts functional disability in schizophrenia. Schizophr. Res. 137, 267-268. doi:10.1016/j.schres.2011.12.021

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

29

Jacobson, C., Gould, M., 2007. The epidemiology and phenomenology of non-suicidal selfinjurious behavior among adolescents: a critical review of the literature. Arch. Suicide Res. 11, 129-147. doi:10.1080/13811110701247602 Janis, I.B., Nock, M.K., 2009. Are self-injurers impulsive?: Results from two behavioral

CR IP T

laboratory studies. Psychiatry Res. 169, 261-267. doi:10.1016/j.psychres.2008.06.041 Klonsky, E.D., 2011. Non-suicidal self-injury in united states adults: Prevalence,

sociodemographics, topography and functions. Psychol. Med. 41, 1981-1986. doi:10.1017/S0033291710002497

AN US

Klonsky, E.D., 2007. The functions of deliberate self-injury: A review of the evidence. Clin. Psychol. Rev. 27, 226-239. doi:10.1016/j.cpr.2006.08.002

Kochanek, K.D., Murphy, S.L., Xu, J.Q., Tejada-Vera, B., 2016. Deaths: final data for 2014.

M

Natl. Vital Stat. Rep. 65. National Center for Health Statistics. Hyattsville, MD. LeGris, J., Van Reekum, R., 2006. The neuropsychological correlates of borderline personality

ED

disorder and suicidal behavior. Can. J. Psychiatry. 51, 131–142. doi:10.1177/070674370605100303

PT

Lezak, M.D., Howieson, D.B., Loring, D.W., Hannay, J., Fischer, J.S., 2004. Orientation and

CE

attention, in: Neuropsychological assessment (4th ed.). Oxford University Press, Inc. New York, NY. pp. 337-374.

AC

Lobbestael, J., Leurgans, M., Arntz, A., 2011. Inter-rater reliability of the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID I) and Axis II Disorders (SCID II). Clin. Psychol. Psychother. 18, 75-79. doi: 10.1002/cpp.693

Maldijan, J.A., Laurienti, P.J., Kraft, R.A., Burdette, J.H., 2003. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets.

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

30

Neuroimage. 19, 1233–1239. doi:10.1016/S1053-8119(03)00169-1 McKenzie, K.C., Gross, J.J., 2014. Nonsuicidal self-injury: An emotion regulation perspective. Psychopathology. 47, 207. doi:10.1159/000358097 McCloskey, M.S., Look, A.E., Chen, E.Y., Pajoumand, G., Berman, M.E., 2012. Nonsuicidal

CR IP T

self-injury: Relationship to behavioral and self-rating measures of impulsivity and selfaggression: Impulsivity and self-aggression. Suicide Life Threat. Behav. 42, 197-209. doi:10.1111/j.1943-278X.2012.0082.x

McNair, D., Lorr, M., Droppleman, L., 1971. Manual for the profile of mood states. Educational

AN US

and Industrial Testing Services. San Diego, CA.

Mennin, D.S., Fresco, D.M., 2014. Emotion Regulation Therapy, in: Gross, J.J. (Ed.), Handbook of emotion regulation (2nd ed.). The Guilford Press, New York, pp. 469-490.

M

Meza, J.I., Owens, E.B., Hinshaw, S.P., 2016. Response inhibition, peer preference and victimization, and self-harm: Longitudinal associations in young adult women with and

ED

without ADHD. J. Abnorm. Child Psych. 44, 323-334. doi:10.1007/s10802-015-0036-5 Nock, M.K., 2009a. Why do people hurt themselves? New insights into the nature and functions

PT

of self-injury. Curr. Dir. in Psychol. Sci. 18, 78-83. doi:10.1111/j.1467-

CE

8721.2009.01613.x

Nock, M.K., 2009b. Understanding nonsuicidal self-injury: origins, assessment, and treatment.

AC

American Psychological Association: Washington, DC. doi:10.1037/11875-000 Nock, M.K., Holmberg, E.B., Photos, V.I., Michel, B.D., 2007. The self-injurious thoughts and behaviors interview: development, reliability, and validity of a new measure. Psychol. Assessment. 19, 309-317. doi:10.1037/1040-3590.19.3.309

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

31

Nock, M.K., Prinstein, M.J., 2004. A functional approach to the assessment of self-mutilative behavior. J. Consult. Clin. Psychol. 72, 885-890. doi:10.1037/0022-006X.72.5.885 Nock M.K., Wedig M.M., Holmberg E.B., and Hooley J.M., 2008. The emotion reactivity scale:

Ther. 39, 107-116. doi:10.1016/j.beth.2007.05.005

CR IP T

development, evaluation, and relation to self-injurious thoughts and behaviors. Behav.

Ohmann, S., Schuch, B., König, M., Blaas, S., Fliri, C., Popow, C., 2008. Self-injurious behavior in adolescent girls. Psychopathology. 41, 226-235. doi:10.1159/000125556

Oldershaw, A., Grima, E., Jollant, F., Richards, C., Simic, M., Taylor, L., et al., 2009. Decision

AN US

making and problem solving in adolescents who deliberately self-harm. Psychol. Med. 39, 95-104. doi:10.1017/S0033291708003693

Osuch, E., Ford, K., Wrath, A., Bartha, R., Neufeld, R., 2014. Functional MRI of pain

M

application in youth who engaged in repetitive non-suicidal self-injury vs. psychiatric controls. Psychiatry Res. 223, 104-112. doi:10.1016/j.pscychresns.2014.05.003

ED

Plener, P. L., Bubalo, N., Fladung, A. K., Ludolph, A.G., Lule, D., 2012. Prone to excitement: adolescent females with Non-suicidal self-injury (NSSI) show altered cortical pattern to

PT

emotional and NSS-related material. Psychiatry Res. 203, 146-152.

CE

doi:10.1016/j.pscychresns.2011.12.012 Ribeiro, J., Franklin, J., Fox, K., Bentley, K., Kleiman, E., Chang, B., et al., 2016. Self-injurious

AC

thoughts and behaviors as risk factors for future suicide ideation, attempts, and death: A meta-analysis of longitudinal studies. Psychol. Med. 46, 225-236. doi:10.1017/S0033291715001804

Strakowski, S.M., Adler, C.M., Holland, S.K., Mills, N.P., DelBello, M.P., Eliassen, J.C., 2005. Abnormal FMRI brain activation in euthymic bipolar disorder patients during a counting

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

32

stroop interference task. Am. J. Psychiatry. 162, 1697-1705. doi:10.1176/appi.ajp.162.9.1697 Shin, L.M., Bush, G., Milad, M.R., Lasko, N.B., Brohawn, K.H., Hughes, K.C, et al., 2011. Exaggerated activation of dorsal anterior cingulate cortex during cognitive interference: a

985. doi:10.1176/appi.ajp.2011.09121812

CR IP T

monozygotic twin study of posttraumatic stress disorder. Am. J. Psychiatry. 168, 978-

Spielberger, C.D., Gorsuch, R.L., Lushene, R., Vagg, P.R., Jacobs, G.A., 1983. Manual for the state-trait anxiety inventory. Consulting Psychologists Press. Palo Alto, CA.

AN US

Tatnell, R.C., Hasking, P.A., 2015. Emotion regulation, the anterior cingulate cortex and nonsuicidal self-injury, in: Haque, S., Sheppard, E. (Eds.), Culture and Cognition: A Collection of Critical Essays. Peter Lang Publishing, Bern, Switzerland, pp. 163-178.

M

doi:10.3726/978-3-0351-0826-2

Thomaes, K., Dorrepaal, E., Draijer, N., de Ruiter, M.B., Elzinga, B.M., van Balkom, A.J., et al.,

ED

2012. Treatment effects on insular and anterior cingulate cortex activation during classic and emotional stroop interference in child abuse-related complex post-traumatic stress

PT

disorder. Psychol. Med. 42, 1-13. doi:10.1017/S0033291712000499

CE

Thompson, D.G., Kesler, S.R., Sudheimer, K., Mehta, K.M., Thompson, L.W., Marquett, R.M., et al., 2015. FMRI activation during executive function predicts response to cognitive

AC

behavioral therapy in older, depressed adults. Am. J. Geriatr. Psychiatry. 23, 13-22. doi:10.1016/j.jagp.2014.02.001

Vega, D., Ripolles, P., Soto, A., Torrubia, R., Ribas, J., Monreal, J.A…Marco-Pallares, J., 2018. Orbitofrontal overactivation in reward processing in borderline personality disorder: the

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

33

role of non-suicidal self-injury. Brain Imaging and Behav. 12, 217-228. doi:10.1007/s11682-017-9687-x Vega, D., Vilà-Balló, A., Soto, À., Amengual, J., Ribas, J., Torrubia, R.. et al., 2015. Preserved error-monitoring in borderline personality disorder patients with and without non-suicidal

CR IP T

self-injury behaviors. PloS One. 10, e0143994. doi:10.1371/journal.pone.0143994

Victor, S.E., Klonsky, E.D., 2014. Correlates of suicide attempts among self-injurers: a metaanalysis. Clin. Psychol. Rev. 34, 282-297. doi:10.1016/j.cpr.2014.03.005

Watson, D., Weber, K., Assenheimer, J. S., Clark, L. A., Strauss, M. E., McCormick, R. A.,

AN US

1995. Testing a Tripartite Model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. J. Abnorm. Psychol 104, 3-14. doi: 10.1037/0021-843X.104.1.3

M

Watson, D., Clark, L. A., 1991. The mood and anxiety symptom questionnaire. University of Iowa, Department of Psychology, Iowa City, IA.

ED

Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Pers. Soc. Psychol. 54, 1063-1070.

PT

Wechsler, D., 1999. Wechsler abbreviated scale of intelligence. The Psychological Corporation,

CE

Hartcourt Brace and Company, San Antonio, TX. Whiteside, S.P., Lynam, D.R., 2001. The five factor model and impulsivity: using a structural

AC

model of personality to understand impulsivity. Pers. Individ. Dif. 30, 669-689. doi:10.1016/S0191-8869(00)00064-7

Williams G.E., Daros A.R., Graves B., McMain S.F., Links P.S., Ruocco A.C., 2015. Executive functions and social cognition in highly lethal self-injuring patients with borderline personality disorder. Personal. Disord. 6, 107-116. doi:10.1037/per0000105

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

34

Yucel, M., Harrison, B.J., Wood, S.J., Fornito, A., Wellard, R.M., Pujol, J., 2007. Functional and biochemical alterations of the medial frontal cortex in obsessive-compulsive disorder. Arch. Gen. Psychiatr. 64, 946-955. doi:10.1001/archpsyc.64.8.946 Zanarini, M., Skodol, A., Bender, D., Dolan, R., Sanislow, C., Schaefer, E., et al., 2000. The

CR IP T

collaborative longitudinal personality disorders study: Reliability of Axis I and II

AC

CE

PT

ED

M

AN US

diagnoses. J. Pers. Disord. 14, 291-299. doi: 10.1521/pedi.2000.14.4.291

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

35

Table 1. Demographic and Clinical State Comparison of Nonsuicidal Self-Injury (NSSI) and Healthy Control Participants NSSI n=15 M±SD 21.27±3.67 13.73±2.60 112.07±12.46 M±SD 15.27±1.98 6.00±3.91

M±SD

124.09±118.74 38.82±54.04 5.00±7.93 1.07±1.59 M±SD

PT

CE

AC

Analyses of Variancea p (η2) 2-tailed .238 (.049) .063 (.118) .311 (.037) 95% CI [14.17, 16.36] [3.84, 8.17]

CR IP T

F 1.455 3.741 1.063

[44.32, 203.86] [2.51, 75.12] [0.42, 9.58] [0.15, 1.99] F p (η2) 1-tailed

AN US

28.93±7.10 29.73±5.84

43.40±9.20 55.60±11.90

23.272 57.105

<.001 (.454) <.001 (.671)

1.07±1.67

20.17±12.63

33.880

<.001 (.575)

13.73±2.58 17.40±2.23 15.20±2.78 46.20±8.77 92.53±12.32

22.73±8.07 23.27±6.54 34.47±12.67 76.20±11.95 156.67±39.90

16.944 10.811 33.085 61.467 56.391

<.001 (.377) .002 (.279) <.001 (.542) <.001 (.687) <.001 (.668)

32.00±5.68 10.36±0.84

21.53±4.21 15.00±4.50

32.077 14.374

<.001 (.543) .001 (.347)

18.73±4.08 1.87±2.03 3.00±2.39 3.73±3.11 2.27±2.63 1.87±2.17 -6.00±10.63 M±SD

14.07±7.07 10.57±8.96 12.79±6.94 14.93±6.04 11.79±11.55 25.07±15.50 61.07±38.71 M±SD

4.818 13.454 26.506 40.181 9.674 32.996 41.766 F

.019 (.151) .001 (.333) <.001 (.495) <.001 (.598) .002 (.264) <.001 (.550) <.001(.607) p (η2) 1-tailed

4.53±2.61 1.67±1.50

14.00±8.11 7.93±4.12

18.449 30.391

<.001 (.406) <.001 (.530)

ED

Age Education (yrs) IQ (WASI) SITBI Age of 1st NSSI Duration of NSSI (yrs) NSSI Episodes Lifetimeb Past Yearb Past Monthc Past Weekc Clinical State STAI State Anxiety Trait Anxiety BDI State Depression MASQ Distress: Anxiety Anxious Arousal Distress: Depression Anhedonic Depression Total PANASd Positive Affect Negative Affect POMSd Vigor Anger Confusion Tension Fatigue Depression Total Mood Disturbance Emotion Regulation ERSd Arousal/Intensity Persistence

Control n=15 M±SD 22.80±3.28 15.47±2.30 116.80±12.68 M±SD -

M

Variable

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

36

AC

CE

PT

ED

M

AN US

CR IP T

Sensitivity 3.60±3.11 16.71±12.15 16.367 <.001 (.377) Total 9.80±6.37 23.72±22.00 21.625 <.001 (.445) Impulsivity M±SD M±SD F p (η2) 1-tailed UPPS Negative Urgency 18.87±5.28 33.87±7.95 37.103 <.001 (.570) Positive Urgency 17.33±5.22 26.53±8.82 12.099 .001 (.302) Premeditation (lack of) 20.13±4.10 25.00±4.78 8.950 .003 (.242) Perseverance (lack of) 16.20±3.91 21.93±5.22 11.595 .001 (.293) Sensation Seeking 32.67±6.96 31.87±6.47 0.106 .374 (.004) Total 17.33±5.22 26.53±8.82 12.099 .001 (.302) Significant Effects in bold a df=1,28 unless otherwise indicated b n=11 c n=14 d df=1,27; Control n=15, NSSI n=14 (one NSSI participant did not complete all scales) Abbreviations: BDI, Beck Depression Inventory; ERS, Emotion Reactivity Scale; MASQ, Mood and Anxiety Symptom Questionnaire; PANAS, Positive and Negative Affect Schedule; POMS, Profile of Mood States; SITBI, Self-Injurious Thoughts and Behaviors Interview; STAI, StateTrait Anxiety Inventory; UPPS, UPPS Impulsive Behavior Scale; WASI, Wechsler Abbreviated Scale of Intelligence

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

37

Table 2. Analyses of Variance (ANOVA) Comparing Nonsuicidal Self-Injury (NSSI) and Healthy Control Participants on Multi-Source Interference Task (MSIT) Performance

F

99.03±1.76 0.79±1.63 0.14±0.36 579.47±106.58

99.38±0.95 0.33±0.62 0.27±0.80 552.22±63.55

0.435 1.008 0.281 0.711

.515 (.016) .324 (.036) .600 (.010) .407 (.026)

91.74±6.53 3.93±4.95 4.00±2.96 834.71±74.17

90.35±9.26 4.33±5.62 4.93±5.06 830.42±76.90

0.216 0.042 0.360 0.023

.646 (.008) .839 (.002) .553 (.013) .880 (.001)

-7.29±5.68 3.14±4.07 3.86±2.83 255.25±80.99

-9.03±8.57 4.00±5.36 4.67±4.75 278.20±61.85

0.407 0.233 0.306 0.742

.529 (.015) .634 (.009) .585 (.011) .397 (.027)

ED PT CE AC

CR IP T

NSSI n=15

M

Control Condition Percent Accuracy Omission Errors Commission Errors Response Time (ms) Interference Condition Percent Accuracy Omission Errors Commission Errors Response Time (ms) Interference-Control Percent Accuracy Omission Errors Commission Errors Response Time (ms) a df=1,26

ANOVAa p (η2) 2-tailed

Control n=14

AN US

Variable

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

38

Table 3. Multi-Source Interference Task (MSIT) Interference–Control Contrast fMRI Local Maxima fMRI Activation from Regions of Interest in the CC and DLPFC

DLPFC

CC DLPFC

x

y

z

t Score

p*

Left Anterior CC Right Middle CC Left Middle CC Right Middle Frontal Gyrus Right Inferior Frontal Gyrus Left Superior Frontal Gyrus Left Superior Frontal Gyrus Right Superior Frontal Gyrus Right Inferior Frontal Gyrus Left Middle CC Right Anterior CC Left Inferior Frontal Gyrus

122 33 181 225 320 12 64 10 36 1048 133 113

-14 12 -2 34 46 -12 -24 24 58 -8 2 -54

46 -16 -40 -2 12 12 -10 44 20 -44 32 34

0 50 54 52 24 48 74 22 2 34 -4 2

4.36 3.93 3.58 6.05 5.19 3.89 3.64 3.57 3.40 7.02 6.65 4.29

<.001 .001 .002 <.001 <.001 .001 .001 .002 .002 <.001 <.001 <.001

None Right Inferior Frontal Gyrus Right Middle Frontal Gyrus Right Middle CC Right Anterior CC Right Middle CC Left Inferior Frontal Gyrus

None 63 109 27 32 18 10

46 12 40 -2 12 -16 14 36 4 -30 -58 22

22 52 50 16 52 22

4.59 3.60 3.76 3.31 2.94 3.06

<.001 .001 <.001 .001 .003 .002

ED

CC

M

2-sample t-tests Control > NSSI CC DLPFC NSSI > Control

Cluster Size k (Voxels)

CR IP T

1-sample t-tests Control > Null CC

NSSI > Null

Coordinate Label

Region

AN US

Comparison

AC

CE

PT

DLPFC *p≤.01 (uncorrected), k≥10 Abbreviations: CC, cingulate cortex; DLPFC, dorsolateral prefrontal cortex; fMRI, functional magnetic resonance imaging; NSSI, nonsuicidal self-injury

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

39

Figure 1. Schematic of the Multi-Source Interference Task (MSIT) illustrating the block design of the control (C) and interference (I) conditions. Image previously published and used with permission (Gruber

CE

PT

ED

M

AN US

CR IP T

et al., 2017).

Figure 2. Functional MRI activations during the Multi-Source Interference task (MSIT) interference-

AC

control contrast in nonsuicidal self-injury (NSSI) patients and healthy control participants from A) 1sample t-tests, and B) 2-sample t-test comparisons. Regions of interest include the cingulate cortex (CC) and dorsolateral prefrontal cortex (DLPFC).

ACCEPTED MANUSCRIPT 40

CE

PT

ED

M

AN US

CR IP T

COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY

Figure 3. Correlation analyses assessing the relationship between functional neuroimaging activation

AC

during the Multi-Source Interference Task (MSIT) and A) emotional reactivity, measured with the Emotion Reactivity Scale (ERS), and B) impulsivity, measured with the UPPS Impulsive Behavior Scale. In patients with nonsuicidal self-injury (NSSI), ERS Emotional Sensitivity ratings (r(12)=-.620, p=.018), ERS Total Emotional Reactivity scores (r(12)=-.569, p=.034), UPPS Negative Urgency scores (r(13)=.539, p=.038), and UPPS Lack of Perseverance ratings (r(13)=-.481, p=.070) all negatively correlated with

ACCEPTED MANUSCRIPT COGNITIVE INTERFERENCE IN NONSUICIDAL SELF-INJURY dorsolateral prefrontal cortex (DLPFC) activation. Within healthy control (HC) individuals, none of the

AC

CE

PT

ED

M

AN US

CR IP T

ERS and UPPS scores significantly correlated with DLPFC activation.

41