The INTERSPEECH 2018 Computational Paralinguistics Challenge addresses four different problems for the first time in a research competition under well-defined conditions: In the Atypical Affect Sub-Challenge, four basic emotions annotated in the speech of handicapped subjects have to be classified; in the Self-Assessed Affect Sub-Challenge, valence scores given by the speakers themselves are used for a three-class classification problem; in the Crying Sub-Challenge, three types of infant vocalisations have to be told apart; and in the Heart Beats Sub-Challenge, three different types of heart beats have to be determined. We describe the Sub-Challenges, their conditions, and baseline feature extraction and classifiers, which include data-learnt (supervised) feature representations by end-to-end learning, the 'usual' ComParE and BoAW features, and deep unsupervised representation learning using the AUDEEP toolkit for the first time in the challenge series.

Hintergrund: In Deutschland sind etwa 4,9 Millionen Menschen an Depressionen erkrankt. Depressionen sind für die Betroffenen und die Gesellschaft mit enormen Belastungen verbunden. Gesundheits-Apps haben hier das Potenzial, die Versorgungslage zu verbessern. Das Ziel dieser systematischen Übersichtsarbeit ist es, die Qualität, Inhalte und Praxisrelevanz von deutschsprachigen Apps für die Anwendung bei Depressionen zu untersuchen. Methode: Die deutschen Google-Play- und iTunes-Stores wurden systematisch nach Apps durchsucht, die explizit mit der Thematik «Depression/Depressivität» warben. Die so ermittelten Apps wurden mithilfe einer Skala zur Einschätzung der Qualität (Mobile Application Rating Scale) von 2 unabhängigen Gutachtern bewertet. Apps mit überdurchschnittlichem Rating wurden von 2 praktisch tätigen Verhaltenstherapeuten im Hinblick auf ihren Nutzen für die klinische Praxis beurteilt. Ergebnisse: Von 1156 identifizierten Apps wurden 38 eingeschlossen. Inhaltlich reichten diese von Informations- bis zu Interventions-Apps. Die Apps wiesen eine mittlere Gesamtqualität auf (M = 3,01; Standardabweichung = 0,56). Vier Apps zeigten überdurchschnittliche Werte. Sie wurden durch 2 Psychotherapeuten getestet und als bedingt empfehlenswert für die klinische Praxis beurteilt. Zu keiner der eingeschlossenen Apps konnte eine Wirksamkeitsstudie gefunden werden. Schlussfolgerungen: Deutschsprachige Depressions-Apps weisen qualitative Mängel auf. Zusätzlich fehlt es an klinischen Studien zum Nutzen und zu Risiken, weshalb der Einsatz in der klinischen Praxis nur bedingt empfohlen werden kann. Ein Gütesiegel für qualitativ hochwertige und praxisrelevante Gesundheits-Apps könnte Nutzer vor Fehlinformationen und Missbrauch schützen und Leistungserbringern den Einsatz digitaler Medien substanziell erleichtern.

Human state-of-mind (SOM; e.g.: perception, cognition, attention) constantly shifts due to internal and external demands. Mental health is influenced by the habitual use of either adaptive or maladaptive SOM. Therefore, the training of conscious regulation of SOM could be promising in self-help (e-and m-health), blended care and psychotherapy. The presented study indicates that SOM can be influenced by telling personal narratives. Furthermore, SOM and narrative sentiment (positive vs. negative) can be predicted through word use. Such results lay the groundwork for the development of applications that analyse text and speech for: i) the early detection of mental health; ii) the early detection of maladaptive changes in emotion dynamics; (iii) the use of personal narratives to improve emotion regulation skills; iv) the distribution of tailored interventions; and finally, v) the evaluation of therapy outcome.

The study of individual differences in human social behavior has a long tradition in (personality) psychology focusing on traits such as extraversion linked to vividness and assertiveness. The study of molecular genetic underpinnings of individual differences in social behavior produced many genetic association studies with only few genetic variants, robustly associated with individual differences in personality. One possible reason for non-replication of findings might be the different inventories used to assess human social traits. Moreover, self-report methods to assess personality and social behavior might be problematic due to their susceptibility to different biases such as social desirability or poor abilities in self-reflection. We stress the importance of including recorded behavior to understand the molecular genetic basis of individual differences in personality and linked social traits. We present preliminary data linking oxytocin genetics to individual differences in social network size derived from smartphones. Here, the genetic variation rs2268498, located in the adjacent area of the promoter of the gene coding for the oxytocin receptor (OXTR), was linked to the number of active contacts and incoming calls, tracked on the smartphone for 12 days (note that these results became a bit weaker when age was controlled for). Although the present empirical findings should only be seen as a proof of concept study, this work demonstrates the feasibility to combine molecular genetic variables with real world behavior. If this approach keeps its promises, the field of personality research might experience a boost in psychometric quality in the near future.

Language analyses reveals crucial information about an individual's current state of mind. Maladaptive psychological functioning appears in cognition, emotional experience and behaviour. In the time of the internet of things, a vast number of text and speech is available; subsequently, the interest in the automated detection of psychological functioning via language is rising. The current study indicates that depression and narcissism can be predicted through word use in personal narratives. Both conditions are characterised by an altered word count regarding anxiety and we (LIWC-based). While depressive individuals use less social words and more anxietyrelated words, narcissists do the opposite. This might reflect the verbal correlate of the cognitive triad in depression. In contrast, narcissists' word use mirrors their excommunicated anxiety of being an undesired self and their inability to reach long-term goals due to a lack of impulse control. The automated recognition of mental state through word use could improve early detection of mental disease, monitoring of disease course, delivery of tailored interventions and evaluation of therapy outcome.