Here we Test Out an Online Compassion Focused Intervention for Adolescents With Social Anxiety Disorder and Analyze Its Effect on (Self)Compassion, Emotion Regulation and Social Anxiety Symptoms, as Well as Heart Rate Variability.

Emotion Regulation (ER) is defined as the process through which the experience and expression of emotions are shaped to serve adaptive behavior. It is a key socio-emotional skill, substantially developed during adolescence, as reliance on parental support decreases in favor of internal ER. Emotion dysregulation has been linked to the development of psychopathology in adolescents, including Social Anxiety Disorder (SAD), and Conduct Disorder (CD). Both disorders are highly prevalent, presenting high rates of comorbidity and putting adolescents at risk for psychopathology in adulthood.

According to Gilbert's evolutionary theory, there are three ER systems. The threat system works towards protection/safety when facing danger; the drive system motivates individuals to seek resources/achieve goals; the soothing system allows for social affiliation/safeness and is stimulated through compassion. ER reflects the balance between these systems; their unbalanced/disproportional use is stimulated throughout childhood care experiences, creating a diathesis to emotion dysregulation.

Socially anxious and disruptive adolescents often come from critical/abusive environments, lacking social acceptance/safeness experiences and presenting overactivated threat and drive systems and an underdeveloped soothing system. In these cases, soothing situations may trigger the perception of interpersonal unsafeness and be considered threatening. Adolescents with CD are traditionally less responsive to reward and adolescents with SAD have fear of negative evaluation; both present withdrawal from reward-seeking, as motivational drive is processed as threat.

Emotional experiences associate with variations in physiological correlates. Stress responses and motivations to approach reward, related to threat/drive activity, are associated with decreased Heart Rate Variability (HRV), while soothing activity/social safeness are associated with increased HRV.

HRV is a suitable biomarker to be used within the framework of compassion and Gilbert's evolutionary theory on ER. Lower HRV relates to difficulties in ER; higher HRV relates to ER and compassion. Subjects with SAD and CD present lower HRV than healthy controls. Healthy adolescent females have lower HRV than healthy adolescent males; literature on sex-differences in adolescents with CD and SAD is sparse and reports inconsistent findings.

Compassion Focused Therapy (CFT) has been proven effective in helping adults with SAD. Compassion training, linked to higher HRV, has been used to downregulate stress responses and achieve better ER. Despite the growing interest on CFT and the evidence of its effectiveness in reducing mental health problems, evidence concerning its efficacy in adolescents with SAD is lacking.

ER has been widely studied recently; nevertheless, research has captured neither the full range of positive and negative emotional experiences, nor the complexity of ER in real life scenarios. This project aims to overcome these limitations by investigating HRV via an experimental procedure encompassing baseline, reactivity and recovery phases in the triggering of the three ER systems, in healthy adolescents and adolescents with SAD and CD. A clinical trial will test if CFT is suitable to reduce SAD symptoms and increase ER in adolescents, and if the therapeutic change is reflected at a psychophysiological level.

Objectives The main goal of this project is to explore differences in HRV as a biomarker of ER. Baseline/reactivity/recovery periods will be compared after the triggering of each ER system, contrasting patterns of healthy adolescents and adolescents with SAD/CD. This project also aims to test if a compassion-based intervention is suitable to promote changes in ER/HRV in adolescents with SAD.

Specific Goals:

• To test for differences in HRV patterns between community adolescents and adolescents with SAD and for gender-based differences in both samples. It is hypothesized that adolescents with SAD will present lower HRV that community adolescents and that girls will present lower HRV than boys in both samples. It is also expected that adolescents with SAD will present distress markers for the three systems' triggering due to soothing-as-threat and drive-as-threat processing, deviating from the HRV patterns of community adolescents (decreased HRV in threat and drive triggering and increased HRV in soothing triggering).
• To test for differences in HRV patterns between adolescents with SAD and adolescents with CD. Due to soothing-as-threat and drive-as-threat processing, adolescents with SAD are expected to present distress markers (decreased HRV) in all activation and recovery periods. Due to diminished reactivity to reward and soothing-as-threat processing, adolescents with CD are expected to present distress markers during activation and recovery of the threat and soothing systems, but not in the triggering of the drive system.
• To investigate changes over time in SAD symptoms, self-reported difficulties in ER, and HRV patterns in adolescents with SAD, after a compassion-focused intervention. It is expected that both HRV and self-report measures will improve after intervention and will keep stable over time.
• To explore the relationship between self-reported improvements in SAD symptoms/ER and psychophysiological changes. It is expected that changes in HRV will predict changes in self-reported variables.

PROJECT RELEVANCE This project adds to the knowledge on Heart Rate Variability (HRV) as a biomarker of Emotion Regulation (ER) in adolescence, in community and clinical samples with Social Anxiety Disorder/Conduct Disorder (SAD/CD), contributing to empirically test the ER systems theory. Testing the efficacy of Compassion Focused Therapy (CFT) and its clinical impact on difficulties in ER and HRV will also offer support to the use of CFT for adolescents with SAD. CFT for adolescents with CD is part of another ongoing research project and will not be tested at this time.

PROJECT DESIGN In the framework of an existing R&D project, a standardized procedure for the measurement of HRV when triggering the three ER systems (PTDC/PSI-ESP/29294/2017) was developed. This project extends on and analyzes data collected during the R&D project.

The standardized procedure lasts 55 minutes and comprises 5-minute baseline, reactivity and recovery periods for the threat, drive and soothing systems' triggering. The triggering is achieved using audio scenarios describing daily situations in which adolescents typically activate each one of the systems. To assess subjective perception of emotional response, a State Affect Scale (SAS) will be filled out by the participant for each period. Control questions will be used to ascertain whether each scenario represented a daily situation for participants (cf. attachment 2).

STUDY 1: CROSS-SECTIONAL STUDY OF DIFFERENCES IN ER/HRV BETWEEN COMMUNITY AND SAD SAMPLE (cf. attachment 3) HRV and self-report ratings of emotions (SAS) will be collected, analyzed and compared between community and SAD adolescents across baseline, reactivity and recovery periods of the standardized procedure.

STUDY 2: CROSS-SECTIONAL STUDY OF DIFFERENCES IN ER/HRV BETWEEN SAD AND CD SAMPLE (cf. attachment 3) The procedure is identical to Study 1.

STUDY 3: LONGITUDINAL STUDY OF THERAPEUTIC CHANGE IN ER/HRV AFTER CFT IN ADOLESCENTS WITH SAD (cf. attachment 3) A compassion-focused intervention for adolescents with SAD was developed during an ongoing R&D project (PTDC/PSI-ESP/29445/2017). According to the CONSORT (2010) guidelines for RCTs, eligibility criteria (inclusion/exclusion) will be assessed first. Then, eligible participants will be assigned to treatment and control groups. Treatment group will receive a 10-session compassion-focused intervention. Control group will be waitlisted for future intervention. Both groups will be assessed at pre-, post-treatment, and at a 3-month follow-up. HRV and self-report measures focusing on SAD and ER will also be collected and analyzed.

Staff partaking in the randomization process will neither be therapists, nor assessors. The candidate will act as the assessor. Therapists will be psychologists trained and working on the ongoing R&D project. This intervention will be carried out with the support of the University of Coimbra Cognitive Behavioral Clinical Psychology Unit, composed by experts in contextual behavioral therapy. Treatment integrity will be assured through the use of a manualized intervention, the qualification of therapists, therapists' supervision and the rating of treatment integrity.

SAMPLES

All samples will be convenience samples, composed by adolescents, aged 14-18 YO, from both sexes. The community sample and the sample with SAD will be recruited in schools. The sample with CD will be recruited in detention centers for young offenders. Power analysis for medium effects (f=.25) with a significance level of .05 and power of .95 required the following sample size:

Community sample: N=60; SAD sample: N=60; CD sample: N=60. Sample size for studies 3&4 is also N=60, (treatment group=30; control group=30).

Inclusion criteria

(according to the MINI-KID):

• Community sample: No diagnosis
• SAD sample: Main diagnosis of Generalized SAD
• CD sample: Main diagnosis of CD

Exclusion criteria

• Cognitive impairment
• Presence of psychotic symptoms according to the MINI-KID
• Presence of diseases and/or use of substances/medications that alter cardiac functioning
• Body mass index over 30 kg/m2, due to its impact on cardiac functioning
• Illiteracy

DATA ANALYSIS Data analysis will be conducted using IBM SPSS Statistics 23. Analysis of Variance will be carried out with gender, sample and assigned group as between-group factors and type of scenario and therapeutic moment as within-group factors. Dependent variables will be psychophysiological HRV measures for all studies and also self-report measures for studies 3&4. For studies 1&2, a repeated measures MANOVA will be conducted, in which gender (boys vs. girls) and sample (community vs. SAD; SAD vs. CD) will be considered as between-group factors and type of scenario as within-group factor. In study 3, a Repeated Measures ANOVA will be used to analyze treatment effects. Assigned group (treatment vs. control group) will be the between-group factor and therapeutic moment will be the within-group factor. All analyses will be carried out according with the intention-to-treat principle. In study 4, mechanisms of change will be tested using two-wave latent change score models.

INSTRUMENTS/VARIABLES

Screening and inclusion/exclusion criteria:

• Informed Consent
• Individual Description Sheet
• Structured Clinical Interview (MINI-KID)
• Cognitive assessment (MoCA)

Self-report protocol:

• Difficulties in Emotion Regulation Scale (DERS-AV)
• Social Anxiety Scale (SAS-A)
• Compassion Scale - Adolescents (CS-A)
• Self-Compassion Scale - Adolescents (SCS-A)
• Fears of Compassion Scale - Adolescents (FCS-A)
• Social Safeness and Pleasure Scale (SSPS)
• Social Anxiety and Avoidance Scale for Adolescents (SAASA)

Phase 1: Literature review Phase 2: Sample selection and assessment for studies 1&2 Phase 3: Data analysis and dissemination of findings I Phase 4: Sample selection and pre-intervention assessment for studies 3&4 Phase 5: Clinical studies Phase 6: Data analysis and dissemination of findings II

Major risks pertain to experimental mortality across all studies. Participant's recruitment in schools will reduce risk for the collection of the community and SAD samples, as students remain attending school throughout the assessment process. The CD sample collection is already in progress. In study 3, the intervention team is composed by experts with considerable experience in conducting clinical trials with adolescents, therefore reducing risk of experimental mortality. In all studies, the researcher vouches to collect data from N+10 participants, to ensure that dropout doesn't affect the completion of the project.

Ethics Ethical concerns regarding these populations encompass the collection of sensitive information, informed consent by vulnerable populations (minors) and the risk of triggering negative affect states. The participation of adolescents is required to study the ER processes and impact of a CFT in these samples. The collection of sensitive data such as clinical and academic information is demanded to verify the inclusion and exclusion criteria and participants' description. This information will be kept private by the researcher and all participants will be assured that this information will only be used for research purposes in scientific publications and specialty seminars. A code will be attributed to each participant and the informed consent forms will be kept separate from the self-report protocols. No identifying elements will be inserted in datasets. As for consent, an information sheet will be delivered to each participant and it will be required for both the participant and a parent/guardian to provide written consent after an explanation of the research project. Participants will also be informed that they can decline to participate at any time. This project will be submitted to approval by the Ethics Committee of the Faculty of Psychology and Educational Sciences of the University of Coimbra, which has already approved the R&D projects that share these similarities. As it requires the collaboration of schools and detention centers for young offenders, the project will also be submitted for approval of the Directorate-General for Education and the Directorate-General of Reintegration and Prison Services. Participants will not be deceived or offered any incentives for their participation. Participants will be informed that their participation is voluntary, and their decision to (not) participate will not be followed by any consequences regarding their grades or justice measures (in the case of the young offenders). The risk of triggering negative affect states during the standardized procedure is slight and cannot be fully eliminated. To minimize risk, the scenarios were developed to trigger mild, not intense emotional responses. Nevertheless, the researcher will be watchful over each participant during the assessments and will monitor any signs or symptoms that may arise. In case of severe psychopathology or intense emotional response detected during any of the assessments, the researcher will flag the case or forward the participant to the school/detention center staff, with the knowledge and agreement of the participant.

Measures' Description

• Screening and inclusion/exclusion criteria

• Informed Consent The informed consent form will contain a 2-page explanation of every procedure to be performed with the participant (interview, self-report assessment and psychophysiological assessment with the standardized procedure), as well as two consent sheets, one for the parent/legal guardian and one for the participant.
• Individual Description Sheet The individual description sheet will be filled with information regarding the socioeconomic background and clinical history of each participant, including: age, sex, date of birth, parent's professions, years of education and grade retentions, height, weight, body mass index, previous health problems, current medications, family mental health history and participant's mental health history.
• Structured Cognitive Interview The Mini International Neuropsychiatric Interview for Children and Adolescents (MINI-KID; Sheehan et al., 2010) was developed in the early 90s but has since been updated to the latest version of each classification (DSM-5 and ICD-10). The Portuguese translation of this instrument will be used in this research project as a screening tool, allowing the assessment of inclusion and exclusion criteria (exclusion criterion: participants presenting psychotic disorders; inclusion criteria: main diagnosis of Social Anxiety Disorder or Conduct Disorder for each of the samples).
• Cognitive assessment The Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) was developed to detect mild cognitive impairment in older adults (Nasreddine et al., 2005) but has also been used in other populations, including adolescents (Pike et al., 2017; Noorani et al., 2020). In this project, the MoCA will serve as a screening tool, as cognitive impairment constitutes exclusion criteria and, therefore, participants who present it will not be included in the studies' samples.

• Self-report protocol

• Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) DERS is a self-report instrument, which assesses multiple aspects of difficulties in emotion regulation in adolescents. It consists of 36 items, to be rated in a 5-point Likert scale according to how often the items apply to them from 1 (almost never) to 5 (almost always). It's composed by 6 dimensions: Non-Acceptance of Emotional Responses (non-acceptance); Difficulties Engaging in Goal-Directed Behavior (goals); Impulse Control Difficulties (impulse); Lack of Emotional Awareness (awareness); Limited Access to Emotion Regulation Strategies (strategies); and Lack of Emotional Clarity (clarity). Higher scores, after converting all reverse-coded items, both for the subscales and the overall score, indicate greater difficulties in emotion regulation.
• Social Anxiety Scale (SAS-A; La Greca & Lopez, 1998) SAS-A is a self-report instrument, which assesses adolescents' social anxiety experiences in the context of peer relations. It consists of 22 items, 4 of which are neutral/filler items, to be rated on a 5-point Likert scale according to how often the items apply to the participants, from 1 (not at all) to 5 (all the time). It's composed by three factors: fear of negative evaluation (FNE), social avoidance and distress in new situations (SAD-New), and social avoidance and distress in general (SAD-General). Higher scores indicate a more severe subjective experience of social anxiety.
• Self-Compassion Scale - Adolescents (SCS-A; Neff, 2003; Portuguese version for adolescents by Pinto-Gouveia, Cunha, Xavier & Castilho, 2011) SCS-A is a self-report instrument which assesses self-compassion dimensions, according to six subscales: Self-Kindness; Self-Judgement; Common Humanity; Isolation; Mindfulness; Over-Identification. It consists of 26 items, to be rated in a 5-point Likert scale according to how often the items apply to the participant in difficult times, from 1 (almost never) to 5 (almost always). The subscale scores are computed by calculating the mean of the subscale's item responses. The total score is calculated by summing up the scores from the Self-Kindness, Common Humanity, and Mindfulness subscales with the reverse scores from the Self-judgement, Isolation and Over-identification subscales. Higher scores indicate greater self-compassion.
• Compassion Scale - Adolescents (CS-A; Pommier, Neff & Tóth-Király, 2020; Portuguese version for adolescents by Sousa, Paulo, Brazão, Castilho & Rijo, 2022) CS-A is a self-report instrument which assesses compassion, containing a global factor of Compassion towards others, and four subscales: Kindness, Common Humanity, Mindfulness and Indiference. It consists of 16 items, to be rated in a 5-point Likert scale according to how often the items apply to the participant from 1 (almost never) to 5 (almost always). Scores for the subscales are computed by summing the item score for each subscale. A higher global score reflects higher compassion towards others.
• Fears of Compassion Scales - Adolescents version (FCS-A; Gilbert, McEwan, Matos & Rivis, 2011; Portuguese version for adolescents by Duarte, Pinto-Gouveia & Cunha, 2014) FCS-A is a self-report instrument assessing fear of compassion for others, for the self, and from others, which allows one to identify barriers to giving compassion to others, to oneself and receiving compassion. It contains three subscales according to the flows of compassion. Higher scores indicate more severe fears of compassion in each flow.
• Social Safeness and Pleasure Scale - Adolescents (SSPS-A; Gilbert, McEwan, Mitra, Richter, Franks, Mills, Bellew & Gale, 2009; Portuguese version for adolescents by Miguel, Sousa, Santos, Brazão, Rijo, Castilho & Gilbert, 2022) SSPS-A is a unidimensional self-report instrument, assessing how participants experience positive and pleasurable feelings in social situations. It consists of 11 items, to be rated from 1 (almost never) to 5 (almost always), according to how the answerer subjectively interprets the world as safe and warm. The overall score ranges from 11 to 55, and the higher the score, the higher social safeness is perceived by the participant.
• Social Anxiety and Avoidance Scale for Adolescents (Cunha, Pinto-Gouveia & Salvador, 2002) SAASA is a 34-item self-report 5-point Likert-type instrument assessing the degree of anxiety (1=none to 5=very much) and frequency of avoidance (1=never to 5=almost always) in a wide range of social situations, representing the most frequent social fears in adolescence. Both subscales load in 6 factors: interactions in new situations, interaction with the opposite sex, performance in social situations, assertive interaction, observation by others, eating and drinking in public.

• Experimental Procedure

• Psychophysiological measures A Firstbeat Bodyguard 2 heartbeat recording device will be used to collect several statistical parameters of HRV. For each moment of the procedure (baseline, reactivity and recovery for the three systems), both time-domain measures (Mean HR and Root Mean Square Successive Difference (RMSSD)) and frequency-domain measures (High-Frequency HRV power components (absolute value (ms2) and normalized units) will be collected.

RMSSD, root-mean square differences of successive R-R intervals, reflects the integrity of vagus nerve-mediated autonomic control of the heart (Task Force of the European Society of Cardiology & the North American Society of Pacing & Electrophysiology, 1996) and is an indicator of Heart Rate Variability and, therefore, Emotion Regulation.

HF power components refer to the power in the High Frequency range in normalized units. The main contributor for the HF component of HRV is efferent vagal activity, meaning these measurements are usually interpreted as markers of parasympathetic influence on vagal tone (Task Force of the European Society of Cardiology & the North American Society of Pacing & Electrophysiology, 1996), as they are highly correlated with RMSSD (Shaffer & Ginsberg, 2017), being an indicator of Heart Rate Variability and Emotion Regulation.

Mean HR is a measure of speed of the heartbeat, representing the average number of contractions per minute (bpm). It has been used in studies pertaining to emotion regulation as a means of increasing predictive value of HRV measurements (Petrocchi et al, 2017; Task Force of the European Society of Cardiology & the North American Society of Pacing & Electrophysiology, 1996). Mean HR is inversely correlated with Heart Rate Variability and Emotion Regulation.

Kubios HRV Standart will be used to assess and correct HRV measures. Correction for noise/artifacts will not exceed 5% of the collected heartbeats, so a very low threshold will be used for artifact correction and correction percentage will be included in the dataset.

• Self-report measures: State Affect Scale (SAS) and Control Questions SAS is a self-report instrument, which assesses the intensity of specific emotional responses activated by the audio scenarios and the neutral images during the standardized procedure. It consists of 10 items, to be rated in a 6-point Likert scale, according to the participant's perception of intensity for each emotion. The items are organized into 3 dimensions: Threat emotions (Anger, Shame, Sadness and Fear), Soothing emotions (Calm, Safeness and Tranquility) and Drive emotions (Enthusiasm, Excitement and Energy).

Two Control Questions are presented at the end of each triggering period of the standardized procedure (baseline, reactivity, recovery), assessing how easily the participants could imagine themselves in the described situation and how similar the scenario was to a situation they experienced themselves in their daily life. Participants are also asked to rate their answers in a 6-point Likert scale.

Inclusion criteria

• Main Diagnosis of Social Anxiety Disorder

Exclusion criteria

• Presence of an externalizing disorder according to the MINI-KID
• Cognitive impairment
• Presence of psychotic symptoms according to the MINI-KID
• Presence of diseases and/or use of substances/medications that alter cardiac functioning
• Body mass index over 30 kg/m2, due to its impact on cardiac functioning
• Illiteracy