Using measures of intrinsic homeostasis and extrinsic modulation to evaluate mental health in adolescents: Preliminary results from the longitudinal adolescent brain study (LABS)

A short-interval longitudinal study of associations between psychological distress and hippocampal grey matter in early adolescence

This study of Australian adolescents (N = 88, 12-13-years-old) investigated the relationship between hippocampal grey matter volume (GMV) and self-reported psychological distress (K10) at four timepoints, across 12 months. Participants were divided into two groups; those who had K10 scores between 10 and 15 for all four timepoints were categorised as "low distress" (i.e., control group; n = 38), while participants who had K10 scores of 16 or higher at least once over the year were categorised as "moderate-high distress" (n = 50). Associations were tested by GEE fitting of GMV and K10 measures at the same time point, and in the preceding and subsequent timepoints. Analyses revealed smaller preceding left GMV and larger preceding right GMV were associated with higher subsequent K10 scores in the "moderate-high distress" group. This was not observed in the control group. In contrast, the control group showed significant co-occurring associations (i.e., at the same TP) between GMV and K10 scores. The "moderate-high distress" group experienced greater variability in distress. These results suggest that GMV development in early adolescence is differently associated with psychological distress for those who experience "moderate-high distress" at some point over the year, compared to controls. These findings offer a novel way to utilise short-interval, multiple time-point longitudinal data to explore changes in volume and experience of psychological distress in early adolescents. The results suggest hippocampal volume in early adolescence may be linked to fluctuations in psychological distress.

clusterBMA: Bayesian model averaging for clustering

Various methods have been developed to combine inference across multiple sets of results for unsupervised clustering, within the ensemble clustering literature. The approach of reporting results from one 'best' model out of several candidate clustering models generally ignores the uncertainty that arises from model selection, and results in inferences that are sensitive to the particular model and parameters chosen. Bayesian model averaging (BMA) is a popular approach for combining results across multiple models that offers some attractive benefits in this setting, including probabilistic interpretation of the combined cluster structure and quantification of model-based uncertainty. In this work we introduce clusterBMA, a method that enables weighted model averaging across results from multiple unsupervised clustering algorithms. We use clustering internal validation criteria to develop an approximation of the posterior model probability, used for weighting the results from each model. From a combined posterior similarity matrix representing a weighted average of the clustering solutions across models, we apply symmetric simplex matrix factorisation to calculate final probabilistic cluster allocations. In addition to outperforming other ensemble clustering methods on simulated data, clusterBMA offers unique features including probabilistic allocation to averaged clusters, combining allocation probabilities from 'hard' and 'soft' clustering algorithms, and measuring model-based uncertainty in averaged cluster allocation. This method is implemented in an accompanying R package of the same name. We use simulated datasets to explore the ability of the proposed technique to identify robust integrated clusters with varying levels of separation between subgroups, and with varying numbers of clusters between models. Benchmarking accuracy against four other ensemble methods previously demonstrated to be highly effective in the literature, clusterBMA matches or exceeds the performance of competing approaches under various conditions of dimensionality and cluster separation. clusterBMA substantially outperformed other ensemble methods for high dimensional simulated data with low cluster separation, with 1.16 to 7.12 times better performance as measured by the Adjusted Rand Index. We also explore the performance of this approach through a case study that aims to identify probabilistic clusters of individuals based on electroencephalography (EEG) data. In applied settings for clustering individuals based on health data, the features of probabilistic allocation and measurement of model-based uncertainty in averaged clusters are useful for clinical relevance and statistical communication.

Early adolescent psychological distress and cognition, correlates of resting-state EEG, interregional phase-amplitude coupling

Delineating neurobiological markers of youth mental health is crucial for early identification and treatment. One promising marker is phase-amplitude coupling (PAC), cross-frequency coupling between the phase of slower oscillatory activity and the amplitude of faster oscillatory activity in the brain. Prior research has demonstrated that PAC is associated with both cognition and mental health and can be modulated using neurostimulation. However, to date research investigating PAC has focused primarily on adults, and only within-region theta-gamma coupling in the context of mental health. We investigated associations between interregional resting-state PAC (posterior-anterior cortex), and cognition and psychological distress in N = 77 (M_age = 12.58 years, SD = 0.31; 51 % female) 12-year-olds. Firstly, while left theta-beta PAC showed a moderate positive correlation (r = 0.529, p < .01), right theta-gamma PAC showed a weak positive correlation, with psychological distress (r = 0.283, p < .05). In terms of cognition, moderate correlations were observed between: (i) increased left theta-beta PAC and increased psychomotor speed (r = -0.367, p < .05); (ii) increased left alpha-beta PAC and decreased attention (r = 0.355, p ≤0.01); and (iii) increased left alpha-beta PAC and decreased verbal learning and memory (r = -0.352, p < .01). Whereas weak associations were observed for: (i) increased left alpha-beta PAC and decreased executive functioning scores (r = 0.284, p < .05); and (ii) increased left alpha-gamma PAC and increased attention (r = -0.272, p < .05). The overall findings of this exploratory study are encouraging, although all the correlations were in the weak-to-moderate range and require replication. Further research may confirm interregional resting-state PAC as a biomarker that can help us better understand the link between mental health and cognition in adolescents and improve treatment of cognitive related deficits in mental illness.

Social Connectedness, Cyberbullying, and Well-Being: Preliminary Findings from the Longitudinal Adolescent Brain Study

Social connectedness is well established as an important aspect of adolescence, with higher levels typically resulting in positive mental health and well-being. Cyberbullying is a prevalent concern during adolescence and is a significant contributor to poor mental health outcomes during this important phase of life. Research shows that social connectedness may act as a protective factor for mental health and well-being when young people experience cyberbullying. However, further research is required to elucidate associations between social connectedness, well-being, and cyberbullying over time. This article outlines preliminary findings from the Longitudinal Adolescent Brain Study (LABS). Data are reported from N = 64 LABS participants recruited at age 12 and assessed at four timepoints over a 12-month period, with a total of 204 completed assessments. Structural equation modeling revealed a mediating effect of social connectedness on the relationship between cyberbullying and well-being. In other words, the negative influences of cyberbullying and cybervictimization on well-being scores over time are influenced by levels of social connectedness. The present findings highlight that increased social connectedness in young people is vital to promoting positive well-being over time and can protect well-being in those experiencing cyberbullying and/or cybervictimization. Findings can inform cyberbullying education programs, health care practitioners, parents, and educators on the importance of young people remaining socially connected when experiencing cyberbullying and/or cybervictimization.

Basal ganglia correlates of wellbeing in early adolescence

It has been suggested that biological markers that define mental health are different to those that define mental illness. The basal ganglia changes dramatically over adolescence and has been linked to wellbeing and mental health disorders in young people. However, there remains a paucity of research on wellbeing and brain structure in early adolescence. This cross-sectional study examined relationships between grey matter volume (GMV) of basal ganglia regions (caudate, putamen, pallidum and nucleus accumbens) and self-reported wellbeing (COMPAS-W), in a sample of Australian adolescents aged 12 years (N = 49, M = 12.6, 46.9% female). Significant negative associations were found between left hemisphere caudate GMV and scores on 'total wellbeing', 'composure' and 'positivity'. The results of this study indicate that smaller caudate GMV at age 12 is linked to increased subjective wellbeing. While seemingly counter-intuitive, our finding is consistent with previous research of decreased GMV in the pons and increased COMPAS-W scores in adults. Our results suggest that protective neurobiological factors may be identifiable early in adolescence and be linked to specific types of wellbeing (such as positive affect and optimism). This has implications for interventions targeted at building resilience against mental health disorders in young people.

A novel, complex systems approach to modelling risk of psychological distress in young adolescents

Adolescence is a period of significant anatomical and functional brain changes, and complex interactions occur between mental health risk factors. The Longitudinal Adolescent Brain Study commenced in 2018, to monitor environmental and psychosocial factors influencing mental health in 500 adolescents, for 5 years. Participants are recruited at age 12 from the community in Australia's Sunshine Coast region. In this baseline, cross-sectional study of N = 64 participants, we draw on the network perspective, conceptualising mental disorders as causal systems of interacting entities, to propose a Bayesian network (BN) model of lifestyle and psychosocial variables influencing chances of individuals being psychologically well or experiencing psychological distress. Sensitivity analysis of network priors revealed that psychological distress (Kessler-10) was most affected by eating behaviour. Unhealthy eating increased the chance of moderate psychological distress by 600%. Low social connectedness increased the chance of severe psychological disorder by 200%. Certainty for psychological wellness required 33% decrease in unhealthy eating behaviours, 11% decrease in low social connectedness, and 9% reduction in less physical activity. BN can augment clinician judgement in mental disorders as probabilistic decision support systems. The full potential of BN methodology in a complex systems approach to psychopathology has yet to be realised.

Short strides to important findings: A short interval longitudinal study of sleep quality, psychological distress and microstructure changes to the uncinate fasciculus in early adolescents

BACKGROUND

Developmental studies have shown adolescence is a period of ongoing white matter (WM) development, reduced sleep quality and the onset of many mental disorders. Findings indicate the WM development of the uncinate fasciculus (UF), a WM tract suggested to play a key role in mental disorders, continues throughout adolescence. While these studies provide valuable information, they are limited by long intervals between scans (1 to 4 years) leaving researchers and clinicians to infer what may be occurring between time-points. To allow inferences to be made regarding the impact that sleep quality may be having on WM development, longitudinal studies with much shorter between-scan intervals are required.

METHODS

The current study reports longitudinal data of self-reported sleep quality (PSQI), diffusion tensor imaging (DTI) measures of WM development and psychological distress (K10) for n = 64 early adolescents spanning the first twelve months (four time-points; Baseline, 4, 8, & 12 months) of the Longitudinal Adolescent Brain Study (LABS) study currently underway at the Thompson Institute.

RESULTS

Generalised Estimating Equation analysis showed a significant relationship between sleep quality and psychological distress over the four time-points. Reduced radial diffusivity and increased fractional anisotropy of the UF is also reported with increasing age suggesting that ongoing myelination is occurring. Adding sleep quality to the model, however, negatively impacted this myelination process.

CONCLUSION

These findings represent an important step towards elucidating how sleep, psychological distress and maturation of the UF may co-develop during early adolescence.