Search

Both affiliation motive theory and the social affiliation model postulate that social interactions follow a homeostatic regulation principle: People try to regulate their daily social interactions towards their “ideal range of social contact”, which is influenced by their personality characteristics, such as affiliative needs. “Less-than-ideal” social interactions result in social deprivation and a desire to increase social contact, whereas more (especially unpleasant) social contact than needed leads to social oversatiation and a desire to be alone. How quickly and how strongly people experience social deprivation or oversatiation varies with individual differences in social personality characteristics (e.g., affiliative needs) as well as situational constraints (e.g., availability of contact partners, possibility to be alone).

Music listening is a pervasive behavior in everyday life that varies both inter- and intraindividually. While personality science has repeatedly related interindividual differences in overall music preferences (i.e., musical taste) to various stable personality traits (e.g., Anderson et al., 2021; Rentfrow & Gosling, 2003), intraindividual fluctuations in daily music-listening behaviors remain largely unaccounted for. Because music is commonly used to modify or maintain the occurrence, duration, and intensity of emotional experiences (e.g., DeNora, 1999; Juslin & Laukka, 2004; Schäfer et al., 2013), mood states may be an important predictor of these music-listening fluctuations. As mood regulation and music preferences are, in turn, differentially related to personality, the Big Five traits may also play a role in these mood-music associations. Hence, the present study will investigate the relationship between music-listening behaviors, fluctuating mood states, and stable personality traits. For this purpose, we will explore natural music listening behavior on the level of listening events collected via smartphone sensing. Thereby, we will represent music listening via listeners’ momentary choice of music in terms of technical audio features.

This project is the central platform for the Network of (German-speaking) Open Science Initiatives (Netzwerk der Open-Science-Initiativen, NOSI). We're providing protocols, links, and resources that might be of interest to others interested in Open Science.

With increasing digitization, every individual produces large amounts of digital footprints every day, which in turn allows researchers to get closer to people's everyday lives than ever before. Accordingly, sensor technologies (especially those integrated into smartphones, our daily companions) have great potential as they provide researchers with useful information about everyday behaviors in different situations on a large scale. However, this data collection method, called mobile sensing, is just in its infancy in social sciences. Therefore, as with the evolvement of any other assessment tool, we first need to understand potential pitfalls of the method to be in turn able to develop it further and to consequently fully benefit from its advantages. Insufficiently researched aspects of mobile sensing include, for example, selection biases associated with sampling. Who is actually willing to participate in these kinds of data-intensive studies? Often, this question is not so easy to answer because little to nothing is known about the non-participants. As we conduct a 14-days mobile sensing study within the framework of a large-scale representative panel study, we have the opportunity to get information on both participants and non-participants. Accordingly, the overarching goal of this preregistered study is to get a better understanding of individuals’ willingness to participate and of resulting selection biases in mobile sensing studies. Specifically, we focus on individual characteristics (socio-demographics and personality traits) associated with willingness to participate.

This preregistration is part of a multi-study project which addresses the application of smartphones as disembodied sensors for the detection of sleep-wake behaviors in daily life. The overall project focuses on the development of different alternative algorithms for the detection of sleep-wake parameters and their comparison by inspecting their validity. The present preregistration addresses the latter, i.e., the validation study in which the sleep-wake behaviors derived by the different smartphone-based methods will be validated against the gold standard of sleep measurement in the field (actigraphy). In addition, a nomological network perspective will be applied for validation purposes.

This study investigates the expression of the traditional concept of sensation seeking in the everyday-usage of smartphones. Data used in this study is part of the PhoneStudy Project.

This project contains materials, data and code for the study "Predicting Personality from Patterns of Behavior Collected with Smartphones", published in PNAS in 2020. This repository contains all the necessary data, syntax and other materials to reproduce the results from the variable-extraction point onwards. The Markdown script "paper.Rmd" knits the corresponding manuscript into a pdf file. Despite our strong commitment to open science practices, the raw log data cannot be made available openly, due to unsolved privacy implications of geolocation data and issues arising from the combination of other behavioral indicators (e.g., communication, app-usage, music consumption etc.). Below you also find a link to an interactive website that allows for the exploration of additional results for and provides resources to enable a better understanding of the used machine learning and data analytical tools.

This project serves as the collection of all research, teaching, and development activities, conducted with and around the PhoneStudy Android application. The aim of the PhoneStudy project is the development of smartphone applications, a server infrastructure, and adjacent data-tools for both the efficient collection and analyses of behavioral data in psychology and other sciences.

To better understand the association between sleep hygiene patterns, chronotype and sleep quality

OSF repository of an introductory one-day workshop presented by Clemens Stachl (clemens.stachl@unisg.ch), Ramona Schödel (Ramona.Schoedel@psy.lmu.de), and Florian Pargent (Florian.Pargent@psy.lmu.de) to audiences from Psychology and other Social Sciences. The materials in this repository will be periodically updated to the latest version of the workshop.

interdisciplinary project about automatic personalization in automotive context

The purpose of this study is to answer the question whether smartphone sensing data can be used to predict subjective sleep quality on a day-to-day level. Subjective Sleep Quality was gathered via consecutive questionnaires in a experience sampling timeframe of a larger-scaled study. Smartphone Sensing data includes information on smartphone usage in general, usage of certain app categories (Social Media/Messaging, Gaming), data on physical activity generated via API, and data on ambient noise and light. These broad categories of relevant smartphone sensing data are derived from literature.

The question how (often) people interact with other people lies at the heart of research on social relationships. However, research on short-term processes of social relationships in daily life has for a long time been hindered by measurement constraints: measuring social relationships in daily life with a high time-resolution is burdensome for participants and researchers alike. Multiple methods with differing degrees of participant burden and intrusiveness have been suggested to assess social relationships in daily life, such as Experience Sampling (ESM), the Day Reconstruction Method (DRM), and more recently Mobile Sensing (MS). In this study, we aim to compare the measurement properties of these three methods with a focus on the measurement of the frequency, duration, and quality of personal contact, as well as mediated contact such as calling and text messaging. We aim to compare the three methods regarding their within-person agreement, e.g., at time points where all three methods provide data, how much do the methods agree? Additionally, we will conduct analyses to gain more information on the potential of each method to capture social interactions that were “missed” by the other methods (e.g., “Can DRM provide data on missed ESM assessments?” or “Does ESM miss text messages sent at night that are captured by MS?”).

Using smartphone data for the predictive modelling of coping strategies - benchmarking of machine learning algorithms based on smartphone sensing, experience sampling and survey data

Project repository for the article of the same name accepted for the special issue on "Behavioral Personality Science in the Age of Big Data" at the European Journal of Personality

This repository offers materials collected for the eCLOC project.

Individuals differ in their circadian timing. These differences are referred to as the trait chronotype and can be observed in between-person varying diurnal rhythms of physiological processes (e.g., body temperature), psychological processes (e.g., positive affect), and behaviors (e.g., sleep-wake timing). Influential theories such as the Social Zeitgeber theory assume that socially relevant behaviors also follow repetitive daily patterns. But as previous research methodology in this context has been dominated by self-reports in recent decades studies have rarely examined whether social behavior actually follows diurnal rhythms. Another question that has remained unanswered so far is whether inter-individual differences are evident in the hypothesized social rhythms and which role chronotype plays in this context. In our study, we aim to contribute to these two gaps in the literature by investigating data collected in an intensive 2-day ambulatory assessment study. First, we aim to replicate previous research on diurnal rhythms in positive affect depending on chronotype (RQ1). Second, we investigate whether smartphone-sensed social behavior also follows a diurnal pattern and whether inter-individual differences can be explained by chronotype (RQ2). Finally, we connect chronotype, socially relevant behaviors, and affect by investigating interaction effects of chronotype, time of day, and perceived valence of social interaction on positive affect (RQ3).

Supervised machine learning (ML) is becoming an influential analytical method in psychology and other social sciences. However, theoretical ML concepts and predictive modeling techniques are not yet widely taught in psychology programs. This tutorial is intended to provide an intuitive but thorough primer and introduction to supervised ML for psychologists in four consecutive modules. After introducing the basic terminology and mindset of supervised ML, Module I covers how to use resampling methods to evaluate the performance of ML models (bias-variance tradeoff, performance measures, k-fold cross-validation). Module II introduces the nonlinear random forest, a type of ML model that is particularly user-friendly and well-suited to predicting psychological outcomes. Module III is about performing empirical benchmark experiments (comparing the performance of several ML models on multiple datasets). Finally, Module IV discusses the interpretation of ML models, including permutation variable importance measures, effect plots (partial dependence plots, individual conditional expectation profiles), and the concept of model fairness. Throughout the tutorial, intuitive descriptions of theoretical concepts are provided, with as few mathematical formulas as possible, and followed by code examples, using the mlr3 and companion packages in R. Key practical analysis steps are demonstrated on the publicly available PhoneStudy dataset (N = 624), which includes over 1800 variables from smartphone sensing to predict Big Five personality trait scores. The manuscript contains a checklist to be used as a reminder of important elements when performing, reporting, or reviewing ML analyses in psychology. Additional examples and more advanced concepts are demonstrated in online materials (https://osf.io/9273g/).

It is a long-held belief in psychology and beyond that individuals’ music preferences reveal information about their personality traits. While initial evidence relates self-reported preferences for broad musical styles to the Big Five dimensions, little is known about day-to-day music listening behavior and the intrinsic attributes of melodies and lyrics that reflect these individual differences. The present study (N = 330) proposes a personality computing approach to fill these gaps with new insights from ecologically valid music listening records from smartphones. We quantified participants’ music preferences via audio and lyrics characteristics of their played songs through technical audio features from Spotify and textual attributes obtained via natural language processing. Using linear elastic net and non-linear random forest models, these behavioral variables served to predict Big Five personality on domain and facet levels. Out-of-sample prediction performances revealed that – on the domain level – Openness was most strongly related to music listening (r = .25), followed by Conscientiousness (r = .13), while several facets of the Big Five also showed small to medium effects. Hinting at the incremental value of audio and lyrics characteristics, both musical components were differentially informative for models predicting Openness and its facets, whereas lyrics preferences played the more important role for predictions of Conscientiousness dimensions. In doing so, the models’ most predictive variables displayed generally trait-congruent relationships between personality and music preferences. These findings contribute to the development of a cumulative theory on music listening in personality science and may be extended in numerous ways by future work leveraging the computational framework proposed here.