Neurocognitive predictors of self-reported reward responsivity and approach motivation in depression: A data-driven approach

A machine-learning approach to model risk and protective factors of vulnerability to depression

There are multiple risk and protective factors for depression. The association between these factors with vulnerability to depression is unclear. Such knowledge is an important insight into assessing risk for developing depression for precision interventions. Based on the behavioral data of 496 participants (all unmarried and not cohabiting, with a college education level or above), we applied machine-learning approaches to model risk and protective factors in estimating depression and its symptoms. Then, we employed Random Forest to identify important factors which were then used to differentiate participants who had high risk of depression from those who had low risk. Results revealed that risk and protective factors could significantly estimate depression and depressive symptoms. Feature selection revealed four key factors including three risk factors (brooding, perceived loneliness, and perceived stress) and one protective factor (resilience). The classification model built by the four factors achieved an ROC-AUC score of 75.50% to classify the high- and low-risk groups, which was comparable to the classification performance based on all risk and protective factors (ROC-AUC = 77.83%). Based on the selected four factors, we generated a mood vulnerability index useful for identifying people's risk for depression. Our findings provide potential clinical insights for developing quick screening tools for mood disorders and potential targets for intervention programs designed to improve depressive symptoms.

Catch the drift: Depressive symptoms track neural response during more efficient decision-making for negative self-referents

Background

Adolescence is a time of heightened risk for developing depression and also a critical period for the development and integration of self-identity. Despite this, the relation between the neurophysiological correlates of self-referential processing and major depressive symptoms in youth is not well understood. Here, we leverage computational modeling of the self-referential encoding task (SRET) to identify behavioral moderators of the association between the posterior late positive potential (LPP), an event-related potential associated with emotion regulation, and youth self-reported symptoms of depression. Specifically, within a drift-diffusion framework, we evaluated whether the association between the posterior LPP and youth symptoms of major depression was moderated by drift rate, a parameter reflecting processing efficiency during self-evaluative decisions.

Methods

A sample of 106 adolescents, aged 12 to 17 (53% male; Mage = 14.49, SD = 1.70), completed the SRET with concurrent high-density electroencephalography and self-report measures of depression and anxiety.

Results

Findings indicated a significant moderation: for youth showing greater processing efficiency (drift rate) when responding to negative compared to positive words, larger posterior LPPs predicted greater depressive symptom severity.

Limitations

We relied on a community sample and our study was cross-sectional in nature. Future longitudinal work with clinically depressed youth would be beneficial.

Conclusions

Our results suggest a neurobehavioral model of adolescent depression wherein efficient processing of negative information co-occurs with increased demands on affective self-regulation. Our findings also have clinical relevance; youth's neurophysiological response (posterior LPP) and performance during the SRET may serve as a novel target for tracking treatment-related changes in one's self-identity.

Comparing resting state and task-based EEG using machine learning to predict vulnerability to depression in a non-clinical population

Major Depressive Disorder (MDD) affects a large portion of the population and levies a huge societal burden. It has serious consequences like decreased productivity and reduced quality of life, hence there is considerable interest in understanding and predicting it. As it is a mental disorder, neural measures like EEG are used to study and understand its underlying mechanisms. However most of these studies have either explored resting state EEG (rs-EEG) data or task-based EEG data but not both, we seek to compare their respective efficacy. We work with data from non-clinically depressed individuals who score higher and lower on the depression scale and hence are more and less vulnerable to depression, respectively. Forty participants volunteered for the study. Questionnaires and EEG data were collected from participants. We found that people who are more vulnerable to depression had on average increased EEG amplitude in the left frontal channel, and decreased amplitude in the right frontal and occipital channels for raw data (rs-EEG). Task-based EEG data from a sustained attention to response task used to measure spontaneous thinking, an increased EEG amplitude in the central part of the brain for individuals with low vulnerability and an increased EEG amplitude in right temporal, occipital and parietal regions in individuals more vulnerable to depression were found. In an attempt to predict vulnerability (high/low) to depression, we found that a Long Short Term Memory model gave the maximum accuracy of 91.42% in delta wave for task-based data whereas 1D-Convolution neural network gave the maximum accuracy of 98.06% corresponding to raw rs-EEG data. Hence if one has to look at the primary question of which data will be good for predicting vulnerability to depression, rs-EEG seems to be better than task-based EEG data. However, if mechanisms driving depression like rumination or stickiness are to be understood, task-based data may be more effective. Furthermore, as there is no consensus as to which biomarker of rs-EEG is more effective in the detection of MDD, we also experimented with evolutionary algorithms to find the most informative subset of these biomarkers. Higuchi fractal dimension, phase lag index, correlation and coherence features were also found to be the most important features for predicting vulnerability to depression using rs-EEG. These findings bring up new possibilities for EEG-based machine/deep learning diagnostics in the future.

Altered electroencephalography resting state network coherence in remitted MDD

Individuals with remitted depression are at greater risk for subsequent depression and therefore may provide a unique opportunity to understand the neurophysiological correlates underlying the risk of depression. Research has identified abnormal resting-state electroencephalography (EEG) power metrics and functional connectivity patterns associated with major depression, however little is known about these neural signatures in individuals with remitted depression. We investigate the spectral dynamics of 64-channel EEG surface power and source-estimated network connectivity during resting states in 37 individuals with depression, 56 with remitted depression, and 49 healthy adults that did not differ on age, education, and cognitive ability across theta, alpha, and beta frequencies. Average reference spectral EEG surface power analyses identified greater left and midfrontal theta in remitted depression compared to healthy adults. Using Network Based Statistics, we also demonstrate within and between network alterations in LORETA transformed EEG source-space coherence across the default mode, fronto-parietal, and salience networks where individuals with remitted depression exhibited enhanced coherence compared to those with depression, and healthy adults. This work builds upon our currently limited understanding of resting EEG connectivity in depression, and helps bridge the gap between aberrant EEG power and brain network connectivity dynamics in this disorder. Further, our unique examination of remitted depression relative to both healthy and depressed adults may be key to identifying brain-based biomarkers for those at high risk for future, or subsequent depression.

Associating broad and clinically defined polygenic scores for depression with depression-related phenotypes

Twin studies indicate that 30-40% of the disease liability for depression can be attributed to genetic differences. Here, we assess the explanatory ability of polygenic scores (PGS) based on broad- (PGS_BD) and clinical- (PGS_MDD) depression summary statistics from the UK Biobank in an independent sample of adults (N = 210; 100% European Ancestry) who were extensively phenotyped for depression and related neurocognitive traits (e.g., rumination, emotion regulation, anhedonia, and resting frontal alpha asymmetry). The UK Biobank-derived PGS_BD had small associations with MDD, depression severity, anhedonia, cognitive reappraisal, brooding, and suicidal ideation but only the association with suicidal ideation remained statistically significant after correcting for multiple comparisons. Similarly small associations were observed for the PGS_MDD but none remained significant after correcting for multiple comparisons. These findings provide important initial guidance about the expected effect sizes between current UKB PGSs for depression and depression-related neurocognitive phenotypes.

Computational Modeling of Self-Referential Processing Reveals Domain General Associations with Adolescent Anxiety Symptoms

What an adolescent thinks about themselves, commonly termed self-referential processing, has significant implications for youth long-term psychological well-being. Self-referential processing plays an important role in anticipatory and reactive processing in social contexts and contributes to symptoms of social anxiety. Previous work examining self-referential processing largely focuses on child and adolescent depression, relying on endorsement and reaction time for positive and negative self-describing adjectives in a self-referential encoding task (SRET). Here, we employ computational methods to interrogate the latent processes underlying choice reaction times to evaluate the fit of several drift-diffusion models of youth SRET performance. A sample of 106 adolescent, aged 12-17 (53% male; M_age = 14.49, SD = 1.70) completed the SRET and self-report measures of anxiety and depression. Our results support the utility of modeling the SRET, where the rate of evidence accumulation (i.e., drift rate) during negative self-referential processing was related to social anxiety above-and-beyond mean task performance. Our regression analyses indicated that youth efficiency in processing of self-referential views was domain general to anxiety, highlighting the importance of assessing both social and physiological anxiety symptoms when predicting SRET performance. The computational modeling results revealed that self-referential views are not uniquely related to depression-related constructs but also facets of anxiety.

Self-processing in relation to emotion and reward processing in depression

BACKGROUND

Depression is characterised by a heightened self-focus, which is believed to be associated with differences in emotion and reward processing. However, the precise relationship between these cognitive domains is not well understood. We examined the role of self-reference in emotion and reward processing, separately and in combination, in relation to depression.

METHODS

Adults experiencing varying levels of depression (n = 144) completed self-report depression measures (PHQ-9, BDI-II). We measured self, emotion and reward processing, separately and in combination, using three cognitive tasks.

RESULTS

When self-processing was measured independently of emotion and reward, in a simple associative learning task, there was little association with depression. However, when self and emotion processing occurred in combination in a self-esteem go/no-go task, depression was associated with an increased positive other bias [b = 3.51, 95% confidence interval (CI) 1.24-5.79]. When the self was processed in relation to emotion and reward, in a social evaluation learning task, depression was associated with reduced positive self-biases (b = 0.11, 95% CI 0.05-0.17).

CONCLUSIONS

Depression was associated with enhanced positive implicit associations with others, and reduced positive learning about the self, culminating in reduced self-favouring biases. However, when self, emotion and reward processing occurred independently there was little evidence of an association with depression. Treatments targeting reduced positive self-biases may provide more sensitive targets for therapeutic intervention and potential biomarkers of treatment responses, allowing the development of more effective interventions.

Self-referential processing in individuals with nonsuicidal self-injury: An fMRI study

•

Individuals with NSSI rated negative adjectives as more relevant.

•

Altered self-referential processing in NSSI related to temporoparietal and subcortical areas.

•

Brain activity in inferior parietal lobe related to ‘nonsuicidality’ in people with NSSI.

Nonsuicidal self-injury (NSSI) is associated with considerable deficits in managing negative self-directed internal experiences. The present study explores the neurophysiological correlates of self-referential processing in individuals with NSSI. A total of 26 individuals with NSSI (≥5 episodes of NSSI behavior in the past year, without suicide attempts) and 35 age-, sex-, education-, and intelligence quotient (IQ)-matched controls participated in this study. Participants underwent fMRI scanning as they performed a personal relevance rating task, which required them to evaluate the personal relevance of emotional words. As predicted, we found that individuals engaging in NSSI tended to rate negative adjectives as more relevant and positive adjectives as less relevant. An analysis of functional neuroimaging data showed that the NSSI group had increased activity relative to the control group in the inferior parietal lobe, inferior temporal gyrus, calcarine, insula, and thalamus in response to positive adjectives. The NSSI group also demonstrated greater activation in the calcarine and reduced activation in the inferior frontal gyrus in response to negative self-referential stimuli compared with the control group. In addition, increased right inferior parietal lobe activity during positive self-referential processing was correlated with reduced suicidal ideation in the NSSI group. Our study provides neural evidence for self-referential processing bias in individuals with NSSI and highlights the need for further research to clarify the pathophysiological features that are specific to NSSI.

Beyond vernacular: Measurement solutions to the lexical fallacy in disgust research

Disgust may play an important role in several mental disorders, in part because disgust seems impervious to corrective information, a feature noted long before it was studied by clinical psychologists. A deeper understanding of disgust could improve not only the treatment of mental disorders, but also other societal problems involving this peculiar emotion. In this paper, we review the measurement of disgust and identify issues that hold back progress in understanding how to treat this emotion. First, self-report measures of disgust, although optimized in terms of reliability, are compromised in terms of validity due to the "lexical fallacy," that is, the assumption that vernacular usage of emotion terms reveals natural kinds. Improved self-report measures that parse disgust from neighboring states of discomfort and disapproval can address this limitation, but these approaches are absent in clinical psychology. Second, "objective" measures of disgust, although free of vernacular limitations, require greater psychometric scrutiny. In a critical review, we find that most instrument-based measures fail to demonstrate adequate reliability, rendering them unsuitable for the individual differences research crucial to clinical psychology. In light of this assessment, we provide several recommendations for improving the reliability and validity of disgust measurement, including renewed attention to theory.

Supervised Machine Learning: A Brief Primer

Machine learning is increasingly used in mental health research and has the potential to advance our understanding of how to characterize, predict, and treat mental disorders and associated adverse health outcomes (e.g., suicidal behavior). Machine learning offers new tools to overcome challenges for which traditional statistical methods are not well-suited. This paper provides an overview of machine learning with a specific focus on supervised learning (i.e., methods that are designed to predict or classify an outcome of interest). Several common supervised learning methods are described, along with applied examples from the published literature. We also provide an overview of supervised learning model building, validation, and performance evaluation. Finally, challenges in creating robust and generalizable machine learning algorithms are discussed.