A Study Examining Adolescents With Non-Suicidal Self-Injury (RDoC)

A Longitudinal Study Examining Three RDoC Constructs in Adolescents With Non-Suicidal Self-Injury

This study will examine longitudinal brain development in young adolescent girls with a history of Non-Suicidal Self-Injury (NSSI). Specifically, three constructs outlined by the Research Domains Criteria (RDoC) will be examined through self-assessment, MRI, and a cognitive battery.

Study Overview

• Status: Completed
• Conditions: Self-Injurious Behavior; Deliberate Self-Harm

Detailed Description

This study is designed to study brain development in 12-14 year old females with a history of NSSI. Assessments will take place over the course of three years, with three visits per year - a total of nine visits.

The first visit will take place at the Ambulatory Research Center (ARC) at the Fairview Riverside hospital complex. This visit will include a number of questionnaires designed to gather information about the participant's physical and psychological health. Additionally, measures of intelligence quotient (IQ) and demographics will be assessed.

The second visit will again take place at the ARC. For this visit, participants will be asked to partake in the Trier Social Stress Test (TSST) in which they prepare a short, five-minute speech and deliver the speech to an audience with an additional task to follow. The TSST will be videotaped. Following the completion of this task, the participant will complete computer tests in order to measure things like attention and memory. During the second visit, five saliva samples will be collected to measure levels of the hormone Cortisol. The participant will need to collect additional saliva samples at home.

The third visit will take place at the University of Minnesota Center for Magnetic Resonance Research (CMRR) and will involve a brain MRI. Before the MRI the participant will be asked to provide a urine sample and complete a drug and pregnancy test. During the MRI, the participant will be asked to lie quietly in the scanner. The participant will do activities such as resting, listening to music, and playing games using a button box while in the scanner.

The participant will need to come in to complete these assessments twice more at one and two years following the completion of the first set of visits - a total of nine visits overall.

• Study Type: Observational
• Enrollment (Actual): 168

Contacts and Locations

Study Locations

• United States
  • Minnesota
    • Minneapolis, Minnesota, United States, 55455
      • University of Minnesota
    • Minneapolis, Minnesota, United States, 55454
      • Ambulatory Research Center (ARC)
    • Minneapolis, Minnesota, United States, 55455
      • Center for Magnetic Resonance Research

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 12 years to 16 years (Child)
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Adolescents with a history of NSSI will be recruited from community outlets such as flyers, as well as through social media (e.g., Facebook and Pandora). Additionally, these adolescents will be recruited through local clinicians who treat adolescent mood disorders. Healthy controls will be recruited through community outlets and social media.

Description

Inclusion Criteria:

• 12-16 years old
• Female
• Has had first menstruation
• Willingness to have de-identified data shared with RDoC database

Inclusion Criteria - Adolescents with Non-Suicidal Self-Injury:

• Any history of self harm

Inclusion Criteria - Healthy Controls:

• No history of self harm

Exclusion Criteria:

• Male
• Pregnancy
• MRI Incompatibility
• IQ of less than 80
• Major Medical Illness
• Diagnosis of any of the following: Primary Psychotic Disorder, Bipolar Spectrum Disorder, Autism Spectrum Disorder, Current or Past Substance Use Disorder

Study Plan

How is the study designed?

Design Details

• Observational Models: Case-Control
• Time Perspectives: Prospective

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
Adolescents with NSSI; 12-16 year old females who have a history of non-suicidal self-injury are included in this cohort. No interventions will be administered.
Healthy Controls; 12-16 year old females with no history of non-suicidal self-injury are included in this cohort.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Lifetime Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on lifetime frequency -- the total number of episodes throughout the youth's life. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Baseline
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Year Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-year frequency, defined as the number of episodes occurring within the year leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Baseline
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Month Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-month frequency, defined as the number of episodes occurring within the month leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Baseline
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Lifetime Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on lifetime frequency -- the total number of episodes throughout the youth's life. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 12
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Year Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-year frequency, defined as the number of episodes occurring within the year leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 12
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Month Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-month frequency, defined as the number of episodes occurring within the month leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 12
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Lifetime Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on lifetime frequency -- the total number of episodes throughout the youth's life. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 24
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Year Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-year frequency, defined as the number of episodes occurring within the year leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 24
Self-Injurious Thoughts and Behaviors Interview (SITBI) - Past Month Episodes	The Self-Injurious Thoughts and Behaviors Interview (SITBI) is a semi-structured clinical interview providing standardized, comprehensive measurement of self-injury. The interview is organized into modules, each starting with a screening question regarding lifetime presence. If a behavior is denied, the module is skipped; if endorsed, the SITBI assesses frequency (lifetime, past year, month, and week), onset, severity (injuries per episode and tissue damage), function, duration, and methods. Higher scores indicate greater severity of Non-Suicidal Self-Injury (NSSI). This specific outcome focuses solely on past-month frequency, defined as the number of episodes occurring within the month leading up to the interview date. Because the SITBI is an interview-based assessment rather than a rating scale, scores reflect the raw number of episodes; thus, while the minimum value is 0, the maximum value is not predetermined.	Month 24

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Salivary Cortisol Response to Stress	Salivary cortisol response was calculated as the Area Under the Curve with respect to increase (AUCi) across five time points (Baseline, +15, +30, +45, and +60 minutes). This value represents the raw change in cortisol concentration over time relative to the initial baseline using the trapezoidal rule. Because this is a measure of change, positive values indicate a cumulative increase above baseline, while negative values reflect a cumulative decrease below the baseline concentration during the session. Data were also windsorized to minimize the influence of extreme outliers.	Baseline
Salivary Cortisol Response to Stress	Salivary cortisol response was calculated as the Area Under the Curve with respect to increase (AUCi) across five time points (Baseline, +15, +30, +45, and +60 minutes). This value represents the raw change in cortisol concentration over time relative to the initial baseline using the trapezoidal rule. Because this is a measure of change, positive values indicate a cumulative increase above baseline, while negative values reflect a cumulative decrease below the baseline concentration during the session. Data were also windsorized to minimize the influence of extreme outliers.	Month 12
Salivary Cortisol Response to Stress	Salivary cortisol response was calculated as the Area Under the Curve with respect to increase (AUCi) across five time points (Baseline, +15, +30, +45, and +60 minutes). This value represents the raw change in cortisol concentration over time relative to the initial baseline using the trapezoidal rule. Because this is a measure of change, positive values indicate a cumulative increase above baseline, while negative values reflect a cumulative decrease below the baseline concentration during the session. Data were also windsorized to minimize the influence of extreme outliers.	Month 24
Beck Scale for Suicidal Ideation (BSSI)	The Beck Scale for Suicide Ideation (BSSI) is a 19-item semi-structured clinical interview used to assess the presence and intensity of suicidal thoughts, plans, and intent. The interview utilizes a screening process where a score of 0 on both items 4 and 5 (active and passive desire) results in the remaining 14 items being skipped and recorded as 0. Each item is rated on a 3-point scale (0 to 2), and the total score is calculated by summing all 19 items. Total scores range from a minimum of 0 to a maximum of 38, with higher scores representing greater severity of suicidal ideation (a worse clinical outcome). If an adolescent is assessed to be at clinical risk during the interview, a standardized safety plan is developed with the youth and their family.	Baseline
Beck Scale for Suicidal Ideation (BSSI)	The Beck Scale for Suicide Ideation (BSSI) is a 19-item semi-structured clinical interview used to assess the presence and intensity of suicidal thoughts, plans, and intent. The interview utilizes a screening process where a score of 0 on both items 4 and 5 (active and passive desire) results in the remaining 14 items being skipped and recorded as 0. Each item is rated on a 3-point scale (0 to 2), and the total score is calculated by summing all 19 items. Total scores range from a minimum of 0 to a maximum of 38, with higher scores representing greater severity of suicidal ideation (a worse clinical outcome). If an adolescent is assessed to be at clinical risk during the interview, a standardized safety plan is developed with the youth and their family.	Month 12
Beck Scale for Suicidal Ideation (BSSI)	The Beck Scale for Suicide Ideation (BSSI) is a 19-item semi-structured clinical interview used to assess the presence and intensity of suicidal thoughts, plans, and intent. The interview utilizes a screening process where a score of 0 on both items 4 and 5 (active and passive desire) results in the remaining 14 items being skipped and recorded as 0. Each item is rated on a 3-point scale (0 to 2), and the total score is calculated by summing all 19 items. Total scores range from a minimum of 0 to a maximum of 38, with higher scores representing greater severity of suicidal ideation (a worse clinical outcome). If an adolescent is assessed to be at clinical risk during the interview, a standardized safety plan is developed with the youth and their family.	Month 24
Beck Depression Inventory	Description: This is a measure of depression symptom severity. Total scores on this scale are reported. Minimum total score: 0 Maximum total score: 54 Higher total scores mean greater severity of depression.	Baseline
Beck Depression Inventory	Description: This is a measure of depression symptom severity. Total scores on this scale are reported. Minimum total score: 0 Maximum total score: 54 Higher total scores mean greater severity of depression.	Month 12
Beck Depression Inventory	Description: This is a measure of depression symptom severity. Total scores on this scale are reported. Minimum total score: 0 Maximum total score: 54 Higher total scores mean greater severity of depression.	Month 24
Personality Assessment Inventory, Adolescent Form (PAI-A)	A self-report personality indicator for adolescent populations based on 22 scales of psychological well-being. Total scores on this subscale are reported. Minimum total score: 0 Maximum total score: 50 Higher total scores mean greater severity of anxiety.	Baseline
Personality Assessment Inventory, Adolescent Form (PAI-A)	A self-report personality indicator for adolescent populations based on 22 scales of psychological well-being. Total scores on this subscale are reported. Minimum total score: 0 Maximum total score: 50 Higher total scores mean greater severity of anxiety.	Month 12
Personality Assessment Inventory, Adolescent Form (PAI-A)	A self-report personality indicator for adolescent populations based on 22 scales of psychological well-being. Total scores on this subscale are reported. Minimum total score: 0 Maximum total score: 50 Higher total scores mean greater severity of anxiety.	Month 24
SPPA, Global Self-worth	This is a measure of self perceptions in adolescents. Here we focus on the subscale for Global Self Worth. Total scores on this subscale are reported. Minimum score: 1 Maximum score: 4 Higher total scores mean greater sense of self-worth.	Baseline
SPPA, Global Self-worth	This is a measure of self perceptions in adolescents. Here we focus on the subscale for Global Self Worth. Total scores on this subscale are reported. Minimum score: 1 Maximum score: 4 Higher total scores mean greater sense of self-worth.	Month 12
SPPA, Global Self-worth	This is a measure of self perceptions in adolescents. Here we focus on the subscale for Global Self Worth. Total scores on this subscale are reported. Minimum score: 1 Maximum score: 4 Higher total scores mean greater sense of self-worth.	Month 24
Urgency, Premeditation, Perseverance, Sensation Seeking, and Positive Urgency (UPPS-P)	This is a measure of different facets of impulsivity. Here we focus on the subscales for negative urgency, which is the tendency to act impulsively in the context of negative emotion. Total scores on this subscale are reported. Minimum score for negative urgency: 1 Maximum score for negative urgency: 4 Higher total scores mean severity of impulsivity in the context of negative emotion.	Baseline
Urgency, Premeditation, Perseverance, Sensation Seeking, and Positive Urgency (UPPS-P)	This is a measure of different facets of impulsivity. Here we focus on the subscales for negative urgency, which is the tendency to act impulsively in the context of negative emotion. Total scores on this subscale are reported. Minimum score for negative urgency: 1 Maximum score for negative urgency: 4 Higher total scores mean severity of impulsivity in the context of negative emotion.	Month 12
Urgency, Premeditation, Perseverance, Sensation Seeking, and Positive Urgency (UPPS-P)	This is a measure of different facets of impulsivity. Here we focus on the subscales for negative urgency, which is the tendency to act impulsively in the context of negative emotion. Total scores on this subscale are reported. Minimum score for negative urgency: 1 Maximum score for negative urgency: 4 Higher total scores mean severity of impulsivity in the context of negative emotion.	Month 24
D-prime (d') Values Representing Response-inhibition Performance on the Emotional Go/NoGo Task	The Emotional Go/NoGo Task assesses participants' ability to inhibit responses in the presence of emotionally valenced stimuli (e.g., negative emotional faces). Participants are instructed to make a speeded button press to frequent "Go" stimuli and withhold responses to infrequent "NoGo" stimuli. Task performance is summarized using d-prime (d'), a standard signal-detection-theory metric that reflects the participant's ability to discriminate between Go and NoGo trials while accounting for both hits and false alarms. Higher d' values indicate better inhibitory control. Values are unitless. Minimum score: -4.2896 Maximum score: 2.5008	Baseline
D-prime (d') Values Representing Response-inhibition Performance on the Emotional Go/NoGo Task.	The Emotional Go/NoGo Task assesses participants' ability to inhibit responses in the presence of emotionally valenced stimuli (e.g., negative emotional faces). Participants are instructed to make a speeded button press to frequent "Go" stimuli and withhold responses to infrequent "NoGo" stimuli. Task performance is summarized using d-prime (d'), a standard signal-detection-theory metric that reflects the participant's ability to discriminate between Go and NoGo trials while accounting for both hits and false alarms. Higher d' values indicate better inhibitory control. Values are unitless. Minimum score: -4.2896 Maximum score: 2.5008	Month 12
D-prime (d') Values Representing Response-inhibition Performance on the Emotional Go/NoGo Task.	The Emotional Go/NoGo Task assesses participants' ability to inhibit responses in the presence of emotionally valenced stimuli (e.g., negative emotional faces). Participants are instructed to make a speeded button press to frequent "Go" stimuli and withhold responses to infrequent "NoGo" stimuli. Task performance is summarized using d-prime (d'), a standard signal-detection-theory metric that reflects the participant's ability to discriminate between Go and NoGo trials while accounting for both hits and false alarms. Higher d' values indicate better inhibitory control. Values are unitless. Minimum score: -4.2896 Maximum score: 2.5008	Month 24
Self-Network Cortical Thickness (Structural MRI)	This measure represents the global average cortical thickness (in millimeters) of the "Self Network," defined a priori using the Glasser et al. (2016) multimodal parcellation. The network comprises two structural nodes: the Anterior Self Network (ACC, mPFC, and orbitofrontal cortex) and the Posterior Self Network (posterior cingulate cortex and precuneus). Cortical thickness values were extracted from T1-weighted structural MRI processed via HCP pipelines. To produce the single value reported in the data table, a grand average was calculated across all parcels within these nodes, averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Baseline
Self-Network Cortical Thickness (Structural MRI)	This measure represents the global average cortical thickness (in millimeters) of the "Self Network," defined a priori using the Glasser et al. (2016) multimodal parcellation. The network comprises two structural nodes: the Anterior Self Network (ACC, mPFC, and orbitofrontal cortex) and the Posterior Self Network (posterior cingulate cortex and precuneus). Cortical thickness values were extracted from T1-weighted structural MRI processed via HCP pipelines. To produce the single value reported in the data table, a grand average was calculated across all parcels within these nodes, averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Month 12
Self-Network Cortical Thickness (Structural MRI)	This measure represents the global average cortical thickness (in millimeters) of the "Self Network," defined a priori using the Glasser et al. (2016) multimodal parcellation. The network comprises two structural nodes: the Anterior Self Network (ACC, mPFC, and orbitofrontal cortex) and the Posterior Self Network (posterior cingulate cortex and precuneus). Cortical thickness values were extracted from T1-weighted structural MRI processed via HCP pipelines. To produce the single value reported in the data table, a grand average was calculated across all parcels within these nodes, averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Month 24
Cognitive Control Network Cortical Thickness and Subcortical Volume (Structural MRI)	This measure reflects the average cortical thickness of the Cognitive Control Network (CCN), defined using the Glasser parcellation for cortical regions. The CCN includes the bilateral dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). Cortical thickness was derived from HCP-processed vertex-wise maps and calculated as surface-area-weighted averages across all CCN parcels. To provide the single value reported in the data table, these measurements were averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Baseline
Cognitive Control Network Cortical Thickness and Subcortical Volume (Structural MRI)	This measure reflects the average cortical thickness of the Cognitive Control Network (CCN), defined using the Glasser parcellation for cortical regions. The CCN includes the bilateral dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). Cortical thickness was derived from HCP-processed vertex-wise maps and calculated as surface-area-weighted averages across all CCN parcels. To provide the single value reported in the data table, these measurements were averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Month 12
Cognitive Control Network Cortical Thickness and Subcortical Volume (Structural MRI)	This measure reflects the average cortical thickness of the Cognitive Control Network (CCN), defined using the Glasser parcellation for cortical regions. The CCN includes the bilateral dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). Cortical thickness was derived from HCP-processed vertex-wise maps and calculated as surface-area-weighted averages across all CCN parcels. To provide the single value reported in the data table, these measurements were averaged across both the left and right hemispheres (bilateral). Higher values represent a greater physical thickness of the cortical ribbon.	Month 24
Amygdala Volume (Structural MRI)	This structural MRI outcome reflects the total bilateral volume of the amygdala, a central node of the Sustained Threat system within the RDoC Negative Valence domain. Amygdala volumes were extracted from T1-weighted structural scans using FreeSurfer's automated volumetric segmentation (aseg). To produce the single value reported in the data table, the volumes of the right and left amygdala were averaged.All volumetric values are reported in cubic millimeters mm³), where higher values represent a greater physical volume of the structure.	Baseline
Amygdala Volume (Structural MRI)	This structural MRI outcome reflects the total bilateral volume of the amygdala, a central node of the Sustained Threat system within the RDoC Negative Valence domain. Amygdala volumes were extracted from T1-weighted structural scans using FreeSurfer's automated volumetric segmentation (aseg). To produce the single value reported in the data table, the volumes of the right and left amygdala were averaged.All volumetric values are reported in cubic millimeters mm³), where higher values represent a greater physical volume of the structure.	Month 12
Amygdala Volume (Structural MRI)	This structural MRI outcome reflects the total bilateral volume of the amygdala, a central node of the Sustained Threat system within the RDoC Negative Valence domain. Amygdala volumes were extracted from T1-weighted structural scans using FreeSurfer's automated volumetric segmentation (aseg). To produce the single value reported in the data table, the volumes of the right and left amygdala were averaged.All volumetric values are reported in cubic millimeters mm³), where higher values represent a greater physical volume of the structure.	Month 24
Bilateral Self-Network Task fMRI Activation During Negative vs Positive Evaluation Conditions of the Self vs Change Task	This measure reflects task-based functional MRI (fMRI) activation within the Self Network during the Self vs. Change Task. Participants evaluate if positive/negative valence words describe them (Self) or can change in a person (Change). The values represent the Z-score of the Blood-Oxygen-Level-Dependent (BOLD) signal contrast for Self vs. Change, indicating the magnitude of brain activity relative to baseline variability. To produce the reported value, Z-scores were extracted from the Self Network-comprising the bilateral ACC, mPFC, OFC, PCC, and precuneus-and averaged across the four bilateral network nodes (left/right anterior; left/right posterior). A Z-score of 0 indicates no difference in activation; higher positive scores represent greater neural activation during self-referential processing. These research-based scores compare relative magnitude between groups rather than clinical thresholds.	Baseline
Bilateral Self-Network Task fMRI Activation During Negative vs Positive Evaluation Conditions of the Self vs Change Task	This measure reflects task-based functional MRI (fMRI) activation within the Self Network during the Self vs. Change Task. Participants evaluate if positive/negative valence words describe them (Self) or can change in a person (Change). The values represent the Z-score of the Blood-Oxygen-Level-Dependent (BOLD) signal contrast for Self vs. Change, indicating the magnitude of brain activity relative to baseline variability. To produce the reported value, Z-scores were extracted from the Self Network-comprising the bilateral ACC, mPFC, OFC, PCC, and precuneus-and averaged across the four bilateral network nodes (left/right anterior; left/right posterior). A Z-score of 0 indicates no difference in activation; higher positive scores represent greater neural activation during self-referential processing. These research-based scores compare relative magnitude between groups rather than clinical thresholds.	Month 12
Bilateral Self-Network Task fMRI Activation During Negative vs Positive Evaluation Conditions of the Self vs Change Task	This measure reflects task-based functional MRI (fMRI) activation within the Self Network during the Self vs. Change Task. Participants evaluate if positive/negative valence words describe them (Self) or can change in a person (Change). The values represent the Z-score of the Blood-Oxygen-Level-Dependent (BOLD) signal contrast for Self vs. Change, indicating the magnitude of brain activity relative to baseline variability. To produce the reported value, Z-scores were extracted from the Self Network-comprising the bilateral ACC, mPFC, OFC, PCC, and precuneus-and averaged across the four bilateral network nodes (left/right anterior; left/right posterior). A Z-score of 0 indicates no difference in activation; higher positive scores represent greater neural activation during self-referential processing. These research-based scores compare relative magnitude between groups rather than clinical thresholds.	Month 24
Bilateral Cognitive Control Network Task fMRI Activation During Inhibition in Negative Contexts Versus Scrambled Contexts in the Go/No-Go Task	This measure reflects task-based functional MRI (fMRI) activation within the Cognitive Control Network (CCN) during the Emotional Go/NoGo Task. Participants press a button for frequent "Go" stimuli and withhold responses for infrequent "NoGo" stimuli across negative, positive, and scrambled contexts. The reported values are the Z-score of the BOLD signal contrast for inhibition during negative vs. scrambled contexts. This indicates the magnitude of neural inhibition effort relative to baseline variability. To produce the single value, Z-scores were extracted from eight bilateral regions (dACC, rACC, mPFC, and DLPFC) and averaged. A Z-score of 0 represents no significant difference in activation between negative and scrambled inhibition; higher positive scores represent greater neural activation. These research-based scores compare relative magnitude between groups rather than established clinical thresholds.	Baseline
Bilateral Cognitive Control Network Task fMRI Activation During Inhibition in Negative Contexts Versus Scrambled Contexts in the Go/No-Go Task	This measure reflects task-based functional MRI (fMRI) activation within the Cognitive Control Network (CCN) during the Emotional Go/NoGo Task. Participants press a button for frequent "Go" stimuli and withhold responses for infrequent "NoGo" stimuli across negative, positive, and scrambled contexts. The reported values are the Z-score of the BOLD signal contrast for inhibition during negative vs. scrambled contexts. This indicates the magnitude of neural inhibition effort relative to baseline variability. To produce the single value, Z-scores were extracted from eight bilateral regions (dACC, rACC, mPFC, and DLPFC) and averaged. A Z-score of 0 represents no significant difference in activation between negative and scrambled inhibition; higher positive scores represent greater neural activation. These research-based scores compare relative magnitude between groups rather than established clinical thresholds.	Month 12
Bilateral Cognitive Control Network Task fMRI Activation During Inhibition in Negative Contexts Versus Scrambled Contexts in the Go/No-Go Task	This measure reflects task-based functional MRI (fMRI) activation within the Cognitive Control Network (CCN) during the Emotional Go/NoGo Task. Participants press a button for frequent "Go" stimuli and withhold responses for infrequent "NoGo" stimuli across negative, positive, and scrambled contexts. The reported values are the Z-score of the BOLD signal contrast for inhibition during negative vs. scrambled contexts. This indicates the magnitude of neural inhibition effort relative to baseline variability. To produce the single value, Z-scores were extracted from eight bilateral regions (dACC, rACC, mPFC, and DLPFC) and averaged. A Z-score of 0 represents no significant difference in activation between negative and scrambled inhibition; higher positive scores represent greater neural activation. These research-based scores compare relative magnitude between groups rather than established clinical thresholds.	Month 24
Bilateral Amygdala Activation to the Negative Face vs Shape (Threat) Condition of the Emotional Face Matching Task	This measure reflects task-based functional MRI (fMRI) activation within the Amygdala, a central node of the Threat Network, during the Hariri Threat Task. In this task, participants match the emotional expression of angry and fearful faces or match neutral shapes. The reported values represent the Z-score of the BOLD signal contrast for the "faces" condition versus the "shapes" condition, indicating the statistical magnitude of brain activity relative to baseline variability. To produce the single value reported in the data table, Z-scores were extracted from the left and right amygdala and averaged across these two bilateral components. A Z-score of 0 represents no significant difference in activation between viewing faces and shapes, while higher positive Z-scores represent greater neural activation in response to threat-related stimuli. These research-based Z-scores compare relative activation magnitude between groups rather than against established clinical thresholds.	Baseline
Bilateral Amygdala Activation to the Negative Face vs Shape (Threat) Condition of the Emotional Face Matching Task	This measure reflects task-based functional MRI (fMRI) activation within the Amygdala, a central node of the Threat Network, during the Hariri Threat Task. In this task, participants match the emotional expression of angry and fearful faces or match neutral shapes. The reported values represent the Z-score of the BOLD signal contrast for the "faces" condition versus the "shapes" condition, indicating the statistical magnitude of brain activity relative to baseline variability. To produce the single value reported in the data table, Z-scores were extracted from the left and right amygdala and averaged across these two bilateral components. A Z-score of 0 represents no significant difference in activation between viewing faces and shapes, while higher positive Z-scores represent greater neural activation in response to threat-related stimuli. These research-based Z-scores compare relative activation magnitude between groups rather than against established clinical thresholds.	Month 12
Bilateral Amygdala Activation to the Negative Face vs Shape (Threat) Condition of the Emotional Face Matching Task	This measure reflects task-based functional MRI (fMRI) activation within the Amygdala, a central node of the Threat Network, during the Hariri Threat Task. In this task, participants match the emotional expression of angry and fearful faces or match neutral shapes. The reported values represent the Z-score of the BOLD signal contrast for the "faces" condition versus the "shapes" condition, indicating the statistical magnitude of brain activity relative to baseline variability. To produce the single value reported in the data table, Z-scores were extracted from the left and right amygdala and averaged across these two bilateral components. A Z-score of 0 represents no significant difference in activation between viewing faces and shapes, while higher positive Z-scores represent greater neural activation in response to threat-related stimuli. These research-based Z-scores compare relative activation magnitude between groups rather than against established clinical thresholds.	Month 24
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) within the "Self Network," defined as the synchronized spontaneous activity between specific brain regions during a task-free state. BOLD signal time series were extracted from the bilateral anterior nodes (ACC, mPFC, OFC) and posterior nodes (PCC, precuneus). Connectivity was quantified by calculating the Pearson correlation coefficient between the average time series of these regions. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, the resulting Z-scores were averaged across the network nodes. A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Baseline
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) within the "Self Network," defined as the synchronized spontaneous activity between specific brain regions during a task-free state. BOLD signal time series were extracted from the bilateral anterior nodes (ACC, mPFC, OFC) and posterior nodes (PCC, precuneus). Connectivity was quantified by calculating the Pearson correlation coefficient between the average time series of these regions. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, the resulting Z-scores were averaged across the network nodes. A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Month 12
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) within the "Self Network," defined as the synchronized spontaneous activity between specific brain regions during a task-free state. BOLD signal time series were extracted from the bilateral anterior nodes (ACC, mPFC, OFC) and posterior nodes (PCC, precuneus). Connectivity was quantified by calculating the Pearson correlation coefficient between the average time series of these regions. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, the resulting Z-scores were averaged across the network nodes. A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Month 24
Resting-state Functional Connectivity Within the Cognitive Control Network	This measure reflects the average within-network resting-state functional connectivity (RSFC) for the Cognitive Control Network (CCN). Following the processing of CIFTI-space gray-ordinate-wise time series, average time series were extracted for the cortical nodes within the CCN as defined by the Glasser multimodal parcellation. Connectivity was quantified by cross-correlating these ROI time series, and the resulting coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the eight bilateral cortical regions: the dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Baseline
Resting-state Functional Connectivity Within the Cognitive Control Network	This measure reflects the average within-network resting-state functional connectivity (RSFC) for the Cognitive Control Network (CCN). Following the processing of CIFTI-space gray-ordinate-wise time series, average time series were extracted for the cortical nodes within the CCN as defined by the Glasser multimodal parcellation. Connectivity was quantified by cross-correlating these ROI time series, and the resulting coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the eight bilateral cortical regions: the dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Month 12
Resting-state Functional Connectivity Within the Cognitive Control Network	This measure reflects the average within-network resting-state functional connectivity (RSFC) for the Cognitive Control Network (CCN). Following the processing of CIFTI-space gray-ordinate-wise time series, average time series were extracted for the cortical nodes within the CCN as defined by the Glasser multimodal parcellation. Connectivity was quantified by cross-correlating these ROI time series, and the resulting coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the eight bilateral cortical regions: the dorsal anterior cingulate cortex (dACC), rostral anterior cingulate cortex (rACC), medial prefrontal cortex (mPFC), and dorsolateral prefrontal cortex (DLPFC). A Z-score of 0 represents no functional synchronization between regions, while higher positive Z-scores represent stronger positive functional connectivity.	Month 24
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) between the bilateral amygdala and the bilateral medial prefrontal cortex (mPFC), representing a key circuit in the Sustained Threat system. Connectivity was quantified by calculating the Pearson correlation coefficient between the averaged BOLD signal time series extracted from these regions during a task-free state. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the left and right hemispheres (bilateral) for both the amygdala and mPFC. A Z-score of 0 represents no functional synchronization between the amygdala and mPFC, while higher positive Z-scores represent stronger positive functional connectivity.	Baseline
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) between the bilateral amygdala and the bilateral medial prefrontal cortex (mPFC), representing a key circuit in the Sustained Threat system. Connectivity was quantified by calculating the Pearson correlation coefficient between the averaged BOLD signal time series extracted from these regions during a task-free state. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the left and right hemispheres (bilateral) for both the amygdala and mPFC. A Z-score of 0 represents no functional synchronization between the amygdala and mPFC, while higher positive Z-scores represent stronger positive functional connectivity.	Month 12
Resting-state Functional Connectivity Within Anterior and Posterior Nodes of the Self Network	This measure reflects the resting-state functional connectivity (RSFC) between the bilateral amygdala and the bilateral medial prefrontal cortex (mPFC), representing a key circuit in the Sustained Threat system. Connectivity was quantified by calculating the Pearson correlation coefficient between the averaged BOLD signal time series extracted from these regions during a task-free state. To allow for statistical comparison, these correlation coefficients were converted to Z-scores using the Fisher's r-to-z transformation. To produce the single value reported in the data table, these Z-scores were averaged across the left and right hemispheres (bilateral) for both the amygdala and mPFC. A Z-score of 0 represents no functional synchronization between the amygdala and mPFC, while higher positive Z-scores represent stronger positive functional connectivity.	Month 24
Timeline Followback	Timeline Followback Method for NSSI as the total number of days with self-injurious behaviors in the past year	Baseline
Timeline Followback	Timeline Followback Method for NSSI as the total number of days with self-injurious behaviors in the past year	Month 12
Timeline Followback	Timeline Followback Method for NSSI as the total number of days with self-injurious behaviors in the past year	Month 24

Collaborators and Investigators

• Sponsor: University of Minnesota
• Collaborators: National Institute of Mental Health (NIMH)
• Investigators: Principal Investigator: Kathryn Cullen, University of Minnesota

Publications and helpful links

Helpful Links

• Nair et al. 2023, Clinical Trajectories in Adolescents with and without a History of Non-Suicidal Self-Injury: The BRIDGES Longitudinal Study
• Basgoze et al. 2021, Multimodal assessment of sustained threat in adolescents with nonsuicidal self-injury
• Basgoze et al. 2023, A Multilevel Examination of Cognitive Control in Adolescents With Nonsuicidal Self-injury
• Thai et al. 2024, A multi-modal assessment of self-knowledge in adolescents with non-suicidal self-injury: a research domains criteria (RDoC) study

Study record dates

Study Major Dates

• Study Start (Actual): August 1, 2016
• Primary Completion (Actual): July 18, 2022
• Study Completion (Actual): July 11, 2023

Study Registration Dates

• First Submitted: October 17, 2016
• First Submitted That Met QC Criteria: October 26, 2016
• First Posted (Estimated): October 27, 2016

Study Record Updates

• Last Update Posted (Actual): March 17, 2026
• Last Update Submitted That Met QC Criteria: March 16, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Depression; Anxiety
• Additional Relevant MeSH Terms: Mental Disorders; Behavioral Symptoms; Behavior; Anxiety Disorders; Depression; Self-Injurious Behavior
• Other Study ID Numbers: PSYCH-2016-24795; 1R01MH107394-01A1 (U.S. NIH Grant/Contract)