Using automated computer vision and machine learning to code facial expressions of affect and arousal: Implications for emotion dysregulation research

As early as infancy, caregivers' facial expressions shape children's behaviors, help them regulate their emotions, and encourage or dissuade their interpersonal agency. In childhood and adolescence, proficiencies in producing and decoding facial expressions promote social competence, whereas deficiencies characterize several forms of psychopathology. To date, however, studying facial expressions has been hampered by the labor-intensive, time-consuming nature of human coding. We describe a partial solution: automated facial expression coding (AFEC), which combines computer vision and machine learning to code facial expressions in real time. Although AFEC cannot capture the full complexity of human emotion, it codes positive affect, negative affect, and arousal-core Research Domain Criteria constructs-as accurately as humans, and it characterizes emotion dysregulation with greater specificity than other objective measures such as autonomic responding. We provide an example in which we use AFEC to evaluate emotion dynamics in mother-daughter dyads engaged in conflict. Among other findings, AFEC (a) shows convergent validity with a validated human coding scheme, (b) distinguishes among risk groups, and (c) detects developmental increases in positive dyadic affect correspondence as teen daughters age. Although more research is needed to realize the full potential of AFEC, findings demonstrate its current utility in research on emotion dysregulation.

Respiratory sinus arrhythmia reactivity across empirically based structural dimensions of psychopathology: A meta-analysis

Low resting respiratory sinus arrhythmia (RSA) is observed in many mental health conditions, including anxiety disorders, mood disorders, schizophrenia spectrum disorders, disruptive behavior disorders, and nonsuicidal self-injury, among others. Findings for RSA reactivity are more mixed. We evaluate associations between RSA reactivity and empirically derived structural categories of psychopathology-including internalizing, externalizing, and thought problems-among physically healthy adults. We searched multiple electronic databases for studies of RSA among participants who were assessed either dimensionally using well-validated measures or diagnostically using structured interviews. Strict inclusion criteria were used to screen 3,605 published reports, which yielded 37 studies including 2,347 participants and 76 effect sizes. We performed a meta-analysis, with meta-analytic regressions of potential moderators, including psychopathology subtypes. The sample-wide meta-analytic association between RSA reactivity and psychopathology was quite small, but heterogeneity was considerable. Moderation analyses revealed significant RSA reactivity (withdrawal) specifically in externalizing samples. Additional moderators included (a) stimulus conditions used to elicit RSA reactivity (only negative emotion inductions were effective), (b) sex (women showed greater RSA reactivity than men), and (c) adherence to established methodological guidelines (e.g., higher electrocardiographic sampling rates yielded greater RSA reactivity). These findings indicate that associations between RSA reactivity and psychopathology are complex and suggest that future studies should include more standardized RSA assessments to increase external validity and decrease measurement error.

Comorbidity and continuity of depression and conduct problems from elementary school to adolescence

Despite nonoverlapping criterion sets, conduct disorder and depression co-occur at much higher rates than expected by chance. Contemporary model-based approaches to explaining heterotypic comorbidity use factor analysis and its variants to evaluate interrelations among symptoms in large population-based and twin samples. These analyses invariably yield broadband internalizing and externalizing factors, which load on a higher-order general liability factor-findings that are robust across age and informant. Although model-based approaches elucidate structural aspects of comorbidity, they are variable-centered, and usually cross-sectional. Most therefore do not assess developmental continuity of comorbidity, or whether noncomorbid individuals are prospectively vulnerable to heterotypic comorbidity. We use an accelerated longitudinal design to evaluate growth in parent-reported conduct problems (CPs) and depression among children, ages 8-15 years, who were recruited at study entry into depressed only (n = 27), CPs only (n = 28), comorbid (n = 81), and control (n = 70) groups based on levels of symptoms. Consistent with normative developmental trends across this age range, steep growth in depression was exhibited by all groups, including those who reported only CPs at study entry. In contrast, growth in CPs was restricted to those who reported high symptoms at intake (with or without comorbid depression), compared with low and stable among depressed only and control participants. To our knowledge, this is the first study to demonstrate, using carefully ascertained "pure" versus comorbid groups who were followed naturalistically, that comorbid depression is likely to develop among those with pure CPs, but comorbid CPs are not likely to develop among those with pure depression. (PsycINFO Database Record