Michelle G. Craske, PhD - US grants

This investigation aims to examine the efficacy of a cognitive-behavioral treatment for nocturnal panic attacks in persons with Panic Disorder/Agoraphobia. Also included are specific investigations of cognitive-attributions as they relate to nocturnal panic, and the relationship between nocturnal panic and measures of physiological arousal during sleep. Furthermore, the study aims to identify factors that may predict the occurrence of nocturnal panic. Nocturnal panic is experienced by almost half of persons with Panic Disorder/Agoraphobia, and yet very little is known about its course or its treatment. After full sleep polysomnogram screening, subjects with recent recurrent nocturnal panics will be randomly assigned to a cognitive-behavioral intervention (tailored specifically to nocturnal panic) versus a wait-list control condition. Pre and post measures include clinician ratings, self- ratings of psychological functioning, self-monitoring- of anxiety and panic, responses to laboratory tasks, and sleep physiology (via ambulatory recording). At completion of the wait-list, subjects in this condition will receive cognitive-behavioral therapy in a type of replication design. in addition, a nine-month follow-up assessment is included to measure the stability of gains. As part of the pre-treatment assessment, subjects will undergo a laboratory assessment designed to examine specific cognitive sets about physiological events that occur during sleep. Subjects will be randomly assigned to "safety" or "no-safety" conditions. The study will measure subjective, behavioral, and physiological responses to an audio signal that ostensibly reflects normal and expected (safety) versus unexpected and strong (no-safety) physiological events. The results of this investigation pertain to the cognitive-behavioral conceptualization of nocturnal panic. Finally, panickers who have never experienced nocturnal panic will be compared to the nocturnal panic sample in terms of cognitive aspects, physiological aspects, sleep history, and general distress, to identify factors that predict nocturnal panic, especially those that are consistent with the cognitive-behavioral conceptualization of nocturnal panic.

With the rise of managed care, the primary care setting is assuming increasing importance as a site for the detection and treatment of all mental health problems. In this setting, panic disorder is prevalent, poorly recognized, and inadequately treated. Because it is both disabling and often masquerades as a variety of other medical conditions, it increases both direct (physician visits and unnecessary testing) and indirect (disability days) costs. This is a multi-institutional collaborative research project designed to implement an intervention to identify and treat panic disorder in the primary care setting and to study its clinical and cost effectiveness over a one-year period. Three sites (UCSD, UCLA, UW) will screen and identify patients in a university primary care clinic who suffer from panic disorder and test an innovative model of service delivery for panic disorder in this setting. Patients will be randomized to receive either care as usual from their primary care physician or collaborative care (CC). CC employs a combination of cognitive-behavioral psychotherapy (delivered in six sessions over eight weeks by a behavioral health specialist [BHS]), expert pharmacotherapy (guided by a psychiatrist's recommendations relayed by the BHS to the prescribing primary care physician), and disease management elements (education and activation of patient and provider and more careful monitoring and sustained follow-up over the next year via phone contact). It is hypothesized that CC will have superior clinical effectiveness (measured in terms of symptom reduction, quality of life, and function). It is also hypothesized that although direct health care costs will be higher for CC than for usual care, indirect costs will be lower and cost-effectiveness analysis will support the adoption of CC as a preferred model of care.

This is a collaborative research effort of Northwestern University and the University of California, Los Angeles to evaluate common and specific risk factors for anxiety disorders and depression. Each site will work a common protocol. We propose a prospective longitudinal study of 700 high school juniors, recruited in two cohorts over consecutive years at two high schools (Evanston and Santa Monica). Using a high-risk design, participants at high risk (according to Neuroticism scores) will be oversampled relative to medium and low risk groups. Their progression will be carefully tracked over the course of 8 to 10 assessments staggered over four to four and a half years of data collection. The participant sample will be geographically, ethnically, and socio-economically diverse. The proposal takes a comprehensive biopsychosocial approach to the conceptualization and measurement of risk factors, which include Neuroticism, depressogenic cognitive style, anxiety sensitivity, introversion and low positive affectivity, sociotropy and autonomy. Measures will include self report, parental report, as well as information processing tasks (modified Stroop, memory tasks), affective modulation of startle reactivity, and ambulatory cortisol assays. In addition, diathesis-stress interactions will be evaluated on the basis of contextual assessment of chronic and episodic life stress. Outcome will be measured in terms of symptoms and diagnosis of anxiety and depression. Various models of commonalities and specificities of risk and their interaction with stress will be tested using hierarchical logistic regression and structural equation modeling. The findings may further our conceptualization of emotional disorders and provide the platform for prevention research.

[unreadable] DESCRIPTION (provided by applicant): The goals of this project are to evaluate risk factors for anxiety disorders. This proposal is a cross-sectional high-risk (HR) design, comparing children with anxiety disorders (AD) to non-disordered children who are at high-risk (HR) based on parental anxiety disorders and non-disordered children at low risk (LR) for anxiety disorders based on the absence of parental psychopathology. Markers of risk for anxiety disorders are hypothesized to include evaluated sympathetic state, lowered vagal tone, selectivity of attention, evaluated reactivity to aversive stimuli, and stronger learning and retention of aversive associations. Thus, the three groups of participants (N=80), aged 7-12 years (pre-pubertal) and matched on gender, will be evaluated in terms of resting sympathetic state and vagal tone, attention to negative facial expressions, startle blink and cardiac reactivity to high decibel tones, with potentiation by "lights off", and rates of acquiring aversive conditioned responses to novel geometric figures paired with airpuffs. In addition, replication of startle and selective attention to facial expressions as well as measurement of recovery of aversive associations will be assessed one week later. Conditioning will be measured in terms of autonomic responses as well as eye gaze movement, and selectivity of attention will be measured using eye gaze. Measurement of eye gaze in youths to gauge ongoing conditioning and attentional processes represents a new methodology.

DESCRIPTION (provided by applicant): Although it is now well established that collaborative, chronic disease management approaches are clinically and cost effective for treating depression in primary care, we know very little about how such models perform for anxiety disorders, which occur more commonly than depression in primary care and are particularly difficult to manage. To address this gap, we propose the first large-scale effectiveness study to test a collaborative care approach to treating primary care anxiety. This work builds directly on our Collaborative Care for Panic study, key informant interviews of clinicians, patients and clinic administrators, and more recent studies in primary care depression. Based on these considerations, we have developed a single, specially designed intervention called CALM (Coordinated Anxiety Learning and Management), that can deliver evidence-based treatment to patients with any of four anxiety disorders prevalent in primary care (Panic Disorder [PD], Generalized Anxiety Disorder [GAD], Social Anxiety Disorder [SAD], and Posttraumatic Stress Disorder [PTSD]), including those patients with comorbid depression and/or moderate substance abuse. This intervention allows patients treatment choice (CBT and/or medication), uses techniques to maximize patient engagement, and includes stepped care algorithms - approaches successfully employed in large-scale studies of primary care depression. We have also included an ethnically diverse study population, as well as Spanish-speaking patients, and propose qualitative studies to better understand how CALM should be tailored to individual clinics and to examine the acceptability of CALM for disadvantaged patients. These qualitative studies will provide valuable information needed for future dissemination of this approach. The primary aims are: (1) to use experimental, quantitative methods to determine the clinical effectiveness of CALM compared to treatment as usual (TAD) for the above four anxiety disorders (as a group and individually) and to compare CALM and TAU in terms of intermediate outcomes such as quality of care, self-efficacy, and social stigma;and (2) to use qualitative methods to assess acceptability and barriers to sustainability of CALM in participating clinics, providers, and patients. The secondary aims are: (1) to use quasi-experimental methods (e.g. instrumental variables) to examine the effects of appropriate treatment, independent of intervention assignment, on functioning and other clinical outcomes;and (2) to estimate health care costs and cost effectiveness of CALM.

DESCRIPTION (provided by applicant): Anxiety disorders are widespread and costly. Social anxiety disorder (SAD) is one of the most common anxiety disorders, and can lead to significant functional impairment. Despite the prevalence of behavior therapy interventions for anxiety disorders, including SAD, the neural mediators of behavioral treatment for anxiety are understudied. Here, we propose research that will elucidate the neurocognitive effects of behavior therapy for SAD. Secondarily, it will allow for the comparison of two distinct types of behavior therapy on a neural level. Participants will be randomized to either cognitive-behavioral therapy (CBT), a wait-list control (WL), or a newer form of behavioral therapy, acceptance and commitment therapy (ACT). Primary comparisons will be between CBT and WL. Secondary comparisons will be between CBT and ACT. Using functional magnetic resonance imaging (fMRI) to scan clients with SAD before and after the 12- week CBT/ACT or a 12-week wait-list control period while they complete a threat processing task that has been used extensively with non-clinical samples (the Linguistic Processing of Affect Task, or L-PAT), we expect to see several changes in neural activations after treatment, relative to the wait list control period. Namely, we expect reductions in the amygdala response to negatively-valenced faces after treatment. We also expect a post-treatment increase in prefrontal activation, specifically in the right ventrolateral prefrontal cortex (RVLPFC), which is implicated in the processing of negative affect. Previous work with non-clinical samples has suggested that the RVLPFC may work to inhibit or dampen the amygdala response to threat;therefore, we predict that therapy will work to strengthen this process. We expect to see a post-treatment difference in the functional connectivity of those two regions (the amygdala and RVLPFC), such that activations in the RVLPFC and amygdala will be more strongly negatively correlated after treatment. We predict that the degree of symptom improvement after treatment will also correlate with the expected changes in neural activation, outlined above. We also present some exploratory hypotheses about differences between CBT and ACT that may emerge at a neural level. The proposed work will be one of the first investigations of the neural mediators of behavior therapy treatment for anxiety, and would also be the first study to compare two behavior therapy treatments on a neural level. This work will shed light on the characteristics of anxiety disorders and improve understanding of the treatments commonly used for anxiety. PUBLIC HEALTH RELEVANCE: The proposed work would be one of the first investigations of the neural mediators of behavior therapy treatment for anxiety, and would also be the first study to compare two behavior therapy treatments on a neural level. This work will shed light on the characteristics of anxiety disorders and improve understanding of the treatments commonly used for anxiety.

DESCRIPTION (provided by applicant): We propose a novel program for post-doctoral training in early detection and prevention of psychotic disorders, mood disorders, and anxiety disorders. Because these disorders strike relatively early in life and tend to be chronic and debilitating, individuals with these conditions utilize a disproportionate share of public funds for medical services and disability. Existing treatments are only partially effective in reducing symptom severity, generally requiring on-going maintenance therapy to do so (with accumulating side effect burden), and do not ameliorate the deficits in occupational and social functioning associated with these syndromes or the vulnerability to future episodes. In addition, for some of these conditions, the earlier treatment is initiated after the onset of illness, the better the long-term outcome, suggesting that progressive aspects of the underlying disease processes make these illnesses increasingly less responsive to interventions. This program will integrate didactic coursework, seminars, and supervised research training in early detection and prevention of psychiatric disorders. We integrate mechanisms to evaluate trainees and the program in relation to four objectives. First, progress toward effective prevention programs for these disorders will be more rapid if the next generation of researchers is armed with the tools needed both to ascertain risk markers and mechanisms and to develop and test preventive interventions. Second, given that these forms of psychopathology are complexly determined phenomena, involving biological and psychosocial processes, the next generation of researchers must be able to integrate effectively across these different levels of analysis in their approach to risk detection and preventive intervention. Third, our focus on transitional age populations will help to overcome the traditional boundaries between research in adolescent and adult forms of psychopathology, facilitating the integration of a developmental perspective. Finally, exposure of trainees to early detection and prevention approaches to multiple domains of psychopathology will provide useful points of comparison and contrast and will stimulate cross-fertilization between areas.

DESCRIPTION (provided by applicant): Fear, whether it occurs in humans suffering from an anxiety disorder or in experimental models with rodents, is reduced by exposing the frightened organism to the fearful stimulus in the absence of any negative consequences (i.e., extinction, or exposure therapy). However, fear often renews when the feared stimulus is encountered in a context different from the exposure context. In rats, we found that interfering with the animal's ability to process contexts during extinction by administering an anticholinergic drug prevented fear renewal. This proposal will determine if the beneficial effect of this drug translates to exposure therapy in socially anxious humans. To this end, 100 individuals with Social Phobia who fear public speaking will undergo repeated sessions of exposure to public speaking, within a virtual reality context. Participants will be randomized to either drug placebo, .2mg/.01 mL Scopolamine, .3mg/.01 mL Scopolamine or .4mg/.01 mL Scopolamine, administered via nasal drops, prior to each session of exposure therapy. One month after completion of exposure therapy, context renewal will be tested by comparing physiological and subjective responses to public speaking in the same virtual context as used during exposure therapy versus a context different than the one used during exposure therapy. The goal is to identify the dose of Scopolamine associated with the greatest reduction in context renewal. In addition, a secondary analysis will attempt to identify those individuals who benefit most from Scopolamine-augmentation of exposure therapy.

DESCRIPTION (provided by applicant): Multi-site systems intervention for unemployed persons with social anxiety The goal of the proposed research is to conduct a multi-site trial of a cognitive-behavioral intervention for enhancing employment success among unemployed persons whose job attainment efforts have been undermined by social anxiety disorder. Social anxiety disorder is a very common and impairing condition, with negative impacts on occupational functioning. Work-related impairments include turning down job offers and promotions, reduced productivity and job performance, lowered educational attainment, increased unemployment, financial dependence, and reduced income. We have shown that social anxiety disorder was the strongest psychiatric predictor of sustained reliance on welfare for support. In response to these documented employment-related impairments, we produced an eight-session work-related group cognitive behavioral therapy (WCBT) that, when coupled with standard vocational rehabilitation services, significantly improved social anxiety, depression, job search behaviors and job search self-confidence compared to a control group of socially anxious jobs seekers who received only vocational services as usual (VAU). Participants in this pilot R34 study were all homeless, primarily African American, urban-based job-seekers with high levels of psychiatric comorbidity and limited education and employment histories. WCBT is designed in a context and style that overcomes accessibility and stigma-related obstacles with special focus on employment-related targets. The current collaborative RO1 will involve two academic, investigative teams and the staff of two community-based employment service agencies to further evaluate WCBT compared to VAU. This two-region study will address whether WCBT enhances job placement and retention assessed over a one-year interval. In addition, this trial will test the implementation of WCBT in a new vocational service agency that was not involved in the development of WCBT, and whether the effects of WCBT generalize to a new population of urban-based, racially diverse job-seekers with educational and educational histories that differ from the original WCBT study site. We will also investigate the system effects of WCBT in the new site that will be informative for broad implementation of WCBT nationwide. Finally, as an innovative extension from the R34, this project will incorporate implementation tools of technology-assisted provider training and delivery of WCBT.

DESCRIPTION (provided by applicant): The goal of this application is to examine the relationship between threat- and reward-related neural circuitries and symptom dimensions of anxiety and depression during the transition from adolescence to adulthood. This project is in line with the Research Domain Criteria's (RDoC) objective of identifying new strategies for psychiatric classification based on observable dimensions and their corresponding neural circuitries. We have identified a tri-level model of symptom dimensions that includes a broad factor (General Distress) common to all anxiety and depressive symptoms, as well as factors of intermediate breadth (Fears, Anhedonia-Apprehension) that are more distinctive to subsets of anxious and depressive symptoms. Existing research highlights dysregulation of a threat-related neural circuit encompassing the amygdala and subgenual portion of the anterior cingulate cortex in both anxiety and depression. Distinctions between anxiety and depression may be present in a reward-related neural circuit encompassing the ventral striatum and orbitofrontal cortex, which appears elevated in anxiety but reduced in depression. By studying both reward- and threat-related brain function, we will address RDoC constructs of Positive Valence (reward sensitivity) as well as Negative Valence (threat sensitivity). We will prospectively examine whether neural profiles predict the course of symptoms, and conversely, whether symptom courses covary with changes in neural activation over 36 months. Taking a vulnerability-stress perspective, we will test whether life-stress moderates the prospective associations between neural profiles and symptom trajectories. In addition, we will evaluate whether neural data possess predictive value above and beyond other indices of threat and reward sensitivity, including self- report, behavioral, and physiological measures. Results are expected to a) enhance our understanding of threat- and reward-related dysregulation in anxiety and depression, b) identify intermediate neural phenotypes that are not limited to existing diagnostic criteria, c) facilitate more targeted pharmacological and neurofeedback based treatments, d) contribute to a classification system for anxiety and depression that is informed by contemporary science, and e) evaluate the precision with which neural data can cross-sectionally and longitudinally classify individuals into empirically established symptom and impairment profiles. To achieve these aims, we will use self-report measures of threat and reward sensitivity to select 18 to 19 year olds who represent the full range on each dimension. Symptoms, impairment, and life stress will be assessed at baseline, 12-months, 24 months, and 36 months, and fMRI scanning of threat- and reward-related brain function will occur at baseline and 36 months. Participants will be recruited from local communities at two sites, UCLA (n=125) and Northwestern University (n=125) when they are 18 to 19 years old.

DESCRIPTION (provided by applicant): The goal of this application is to examine the relationship between threat- and reward-related neural circuitries and symptom dimensions of anxiety and depression during the transition from adolescence to adulthood. This project is in line with the Research Domain Criteria's (RDoC) objective of identifying new strategies for psychiatric classification based on observable dimensions and their corresponding neural circuitries. We have identified a tri-level model of symptom dimensions that includes a broad factor (General Distress) common to all anxiety and depressive symptoms, as well as factors of intermediate breadth (Fears, Anhedonia-Apprehension) that are more distinctive to subsets of anxious and depressive symptoms. Existing research highlights dysregulation of a threat-related neural circuit encompassing the amygdala and subgenual portion of the anterior cingulate cortex in both anxiety and depression. Distinctions between anxiety and depression may be present in a reward-related neural circuit encompassing the ventral striatum and orbitofrontal cortex, which appears elevated in anxiety but reduced in depression. By studying both reward- and threat-related brain function, we will address RDoC constructs of Positive Valence (reward sensitivity) as well as Negative Valence (threat sensitivity). We will prospectively examine whether neural profiles predict the course of symptoms, and conversely, whether symptom courses covary with changes in neural activation over 36 months. Taking a vulnerability-stress perspective, we will test whether life-stress moderates the prospective associations between neural profiles and symptom trajectories. In addition, we will evaluate whether neural data possess predictive value above and beyond other indices of threat and reward sensitivity, including self- report, behavioral, and physiological measures. Results are expected to a) enhance our understanding of threat- and reward-related dysregulation in anxiety and depression, b) identify intermediate neural phenotypes that are not limited to existing diagnostic criteria, c) facilitate more targeted pharmacological and neurofeedback based treatments, d) contribute to a classification system for anxiety and depression that is informed by contemporary science, and e) evaluate the precision with which neural data can cross-sectionally and longitudinally classify individuals into empirically established symptom and impairment profiles. To achieve these aims, we will use self-report measures of threat and reward sensitivity to select 18 to 19 year olds who represent the full range on each dimension. Symptoms, impairment, and life stress will be assessed at baseline, 12-months, 24 months, and 36 months, and fMRI scanning of threat- and reward-related brain function will occur at baseline and 36 months. Participants will be recruited from local communities at two sites, UCLA (n=125) and Northwestern University (n=125) when they are 18 to 19 years old.

PROJECT SUMMARY Anhedonia, or loss of interest or pleasure in usual activities, is characteristic of depression, some types of anxiety, as well as substance abuse and schizophrenia. Anhedonia is a predictor of poor long term outcomes, including suicide, and poor treatment response. Extant psychological and pharmacological treatments are relatively ineffective for anhedonia. Thus, there is an unmet therapeutic need for this high-risk symptom. Recent advances in affective neuroscience have elucidated processes that may underlie anhedonia and should be targeted in therapy. Specifically, anhedonia is associated with deficits in the appetitive reward system, including (1) reward approach-motivation, (2) initial responsiveness to reward attainment, and (3) learning of reward. We have developed a novel transdiagnostic psychosocial treatment for anhedonia, Positive Affect Treatment (PAT), designed to improve deficits in reward sensitivity. In our pilot study of 61 depressed or anxious and functionally impaired individuals we found a strong preliminary efficacy signal and evidence for treatment specificity. Specifically, PAT led to significant improvements in symptoms of anhedonia, depression and anxiety and was more effective for individuals with anhedonia at baseline than a treatment designed to reduce negative affect. The proposed application uses an experimental therapeutics approach to elucidate whether reward approach- motivation, initial responsiveness to reward attainment or reward learning change with PAT (i.e., target engagement) in the R61 phase, and to evaluate whether changes in reward sensitivity mediate outcomes from PAT in the R33 phase. In the R61 phase, 68 individuals with anhedonia who are anxious or depressed and functionally impaired will be randomized to single-dose PAT (15 weekly sessions) or double dose PAT (PAT- DD, 30 twice-weekly sessions). Physiological, behavioral, and self-report measures of reward approach- motivation, initial reward responsiveness, and learning will be assessed repeatedly. Outcomes include clinician and self-report measures of anhedonia and ecologically valid measures of functional impairment (i.e., physical activity and social interaction). Progression to R33 depends on post-treatment average anhedonia scores that are within 1 SD of the norm, and significant (effect size > .8) pre- to post-treatment changes in reward approach- motivation, initial responsiveness to attainment or learning, which covary significantly with anhedonia over treatment. The treatment dose that produces significantly greater pre- to post-treatment change in one or more target measures will be carried forward to R33. In the R33 phase, 100 individuals will be randomized to PAT (or PAT-DD) or to Negative Affect Treatment; the latter designed to reduce threat sensitivity. Indices of reward sensitivity (from R33 phase) and threat sensitivity will be evaluated as mediators of PAT and moderated mediation will be tested by comparing reward sensitivity mediation in PAT vs NAT. If successful, PAT will offer a viable and effective treatment for individuals with anhedonia, and the results will elucidate the mechanisms responsible for the effectiveness of this novel treatment.

PROJECT SUMMARY The objective of this application is to use the novel approach of neuro-reinforcement based on decoded fMRI information to reduce fear responses to fearful stimuli (e.g., spiders, heights) in individuals with phobias, directly and unconsciously in the brain, without repeatedly exposing participants to their feared stimuli. The Specific Aims of the R61 Phase are to: (1) confirm that our method engages the neurobiological target (amygdala reactivity to images of a feared object) in a population of individuals with specific phobia; and (2) to determine dosage-response optimization. The Specific Aims of the R33 Phase are to: (1) quantify how changes in amygdala reactivity with neuro-reinforcement covary with changes in behavioral, subjective, and physiological outcomes in the same context as the treatment; and (2) determine how engagement of the neurobiological target and related effects in subjective fear ratings and physiological outcomes generalize to new contexts and clinically relevant outcomes. Pilot Data demonstrate that the proposed method can effectively reduce amygdala reactivity to feared stimuli (target engagement). The R61 Phase will confirm engagement of the neurobiological target (amygdala reactivity to fearful stimuli) by our intervention method in patients suffering from phobia of everyday objects and animals (e.g., spiders). Varying the number of neuro- reinforcement sessions across different subject groups (3 groups of 10 subjects each) and measuring reductions in the amygdala reactivity will demonstrate the robustness and mechanisms of target engagement. This will also enable assessment of the optimal dosage required to balance between maximal target engagement and imaging costs, in order to inform the studies in the R33 Phase. The R33 Phase will utilize the optimal dosage determined in the R61 Phase to test mediation of clinically relevant outcomes (clinical diagnostic interviews, subjective fear ratings, and skin conductance) by engagement of the neurobiological target, both immediately after treatment and after a 4-month period. The R33 Phase will further assess whether the effects generalize to a clinically relevant measure of phobic individuals' readiness to approach the feared objects in a virtual environment (Behavioral Approach Test).

PROJECT SUMMARY Rates of anxiety and depression in youth are substantial, causing a major unmet need for effective interventions. Although some progress has been made in preventing these internalizing problems in adolescents, further research is needed that specifically targets theoretically and empirically supported risk processes. An important and salient risk factor found to increase the likelihood of anxiety and depression is negative affectivity ? a partially heritable trait propensity to experience and express more frequent, intense, and enduring aversive emotional states. The proposed randomized controlled prevention trial builds on our finding from our longitudinal study that elevated levels of negative affectivity during adolescence prospectively predicted internalizing disorders in early adulthood (Zinbarg et al., 2016); moreover, this relation was mediated by changes in momentary negative affect (mNA) measured with ecological momentary assessment (EMA) (Adam et al., 2018). The first phase (R61) of the proposed selective prevention trial will test whether an app-based, coach-supported mindfulness intervention as compared to an assessment-only control reduces momentary negative affect, measured with ecological momentary assessment (EMA), in 120 adolescents (age 12-16) at- risk based on their having high levels of trait negative affectivity. EMA will be used to measure average daily mood, (the ?Target?) collected six times a day across three days at pre-, mid-, and post- intervention. ?Target? engagement will be defined as a medium effect size (>.40) in the comparison of youth randomized to MBI versus control on the target ? momentary negative affect ? at post-test, adjusting for pre-test levels. We also will assess the dose-response relation by testing the association between number of sessions and exercises completed with changes in momentary negative affect and weekly mood ratings. In the second phase (R33), we will conduct a replication trial with a new sample of 360 at-risk (i.e., high trait negative affectivity) youths (ages 12-16) randomized to one of three conditions ? MBI, a nonspecific control, or an assessment-only control. Youth will be evaluated with regard to the target (i.e., mNA), internalizing symptoms and disorders, and functioning (e.g., social, academic) at baseline and post-intervention (R61 and R33), and at a 6-month follow-up (R33). Finally, in the R33 we will test if significant reductions in momentary negative affect are associated with improvements (or less worsening) in internalizing symptoms and fewer onsets of internalizing disorders.