Predicting personality from patterns of behavior collected with smartphones.

Smartphones enjoy high adoption rates around the globe. Rarely more than an arm's length away, these sensor-rich devices can easily be repurposed to collect rich and extensive records of their users' behaviors (e.g., location, communication, media consumption), posing serious threats to individual privacy. Here we examine the extent to which individuals' Big Five personality dimensions can be predicted on the basis of six different classes of behavioral information collected via sensor and log data harvested from smartphones. Taking a machine-learning approach, we predict personality at broad domain ([Formula: see text] = 0.37) and narrow facet levels ([Formula: see text] = 0.40) based on behavioral data collected from 624 volunteers over 30 consecutive days (25,347,089 logging events). Our cross-validated results reveal that specific patterns in behaviors in the domains of 1) communication and social behavior, 2) music consumption, 3) app usage, 4) mobility, 5) overall phone activity, and 6) day- and night-time activity are distinctively predictive of the Big Five personality traits. The accuracy of these predictions is similar to that found for predictions based on digital footprints from social media platforms and demonstrates the possibility of obtaining information about individuals' private traits from behavioral patterns passively collected from their smartphones. Overall, our results point to both the benefits (e.g., in research settings) and dangers (e.g., privacy implications, psychological targeting) presented by the widespread collection and modeling of behavioral data obtained from smartphones.

Variation in social media sensitivity across people and contexts.

Social media impacts people's wellbeing in different ways, but relatively little is known about why this is the case. Here we introduce the construct of "social media sensitivity" to understand how social media and wellbeing associations differ across people and the contexts in which these platforms are used. In a month-long large-scale intensive longitudinal study (total n = 1632; total number of observations = 120,599), we examined for whom and under which circumstances social media was associated with positive and negative changes in social and affective wellbeing. Applying a combination of frequentist and Bayesian multilevel models, we found a small negative average association between social media use AND subsequent wellbeing, but the associations were heterogenous across people. People with psychologically vulnerable dispositions (e.g., those who were depressed, lonely, not satisfied with life) tended to experience heightened negative social media sensitivity in comparison to people who were not psychologically vulnerable. People also experienced heightened negative social media sensitivity when in certain types of places (e.g., in social places, in nature) and while around certain types of people (e.g., around family members, close ties), as compared to using social media in other contexts. Our results suggest that an understanding of the effects of social media on wellbeing should account for the psychological dispositions of social media users, and the physical and social contexts surrounding their use. We discuss theoretical and practical implications of social media sensitivity for scholars, policymakers, and those in the technology industry.

Situating smartphones in daily life: Big Five traits and contexts associated with young adults' smartphone use.

We examine individual differences in smartphone behavior to understand the independent effects of Big Five traits and four different contextual factors (places, people, co-occurring activities, and psychological situations) on the frequency and duration of smartphone use in daily life. Using survey, experience sampling, and mobile sensing data collected over the span of 2 weeks from two samples of college students (Sample 1, N = 634; Sample 2, N = 211), we conducted a series of multilevel Bayesian gamma hurdle and negative binomial hurdle models to explain smartphone use (vs. nonuse) and the degree of use. Our pooled findings suggest that extraversion was associated with more frequent use, while conscientiousness was associated with smartphone nonuse and shorter durations of use. In terms of context, our findings show that smartphones were used more frequently when people were out and about in public places (e.g., cafes, stores) and less frequently in particularly social places (e.g., bars, friends' houses). Smartphones were also used more frequently with weak ties (e.g., classmates, coworkers) and less frequently with close ties (e.g., roommates, family, significant others). Smartphones were also used less and for shorter durations when people were engaged in certain activities (e.g., studying, commuting, chores, exercising), and when in situations perceived to be romantic or involving work. We discuss the findings with regard to past work on smartphone use and describe the next steps for research on smartphone behavior. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Well-being in social interactions: Examining personality-situation dynamics in face-to-face and computer-mediated communication.

Decades of research show that people's social lives are linked to their well-being. Yet, research on the relationship between social interactions and well-being has been largely inconclusive with regard to the effects of person-situation interactions, such as the interplay between contextual factors (e.g., interactions occurring in physical vs. digital contexts, different interaction partners) and dispositional tendencies (e.g., Big Five personality traits). Here, we report on exploratory and confirmatory findings from three large studies of college students (Study 1: N = 1,360; Study 2: N = 851; Study 3: N = 864) who completed a total of 139,363 experience sampling surveys (reporting on 87,976 social interactions). We focus on the effects of different modes of communication (face-to-face [FtF] interactions, computer-mediated communication [CMC], and mixed episodes [FtF + CMC]), and types of interaction partners (close peers, family members, and weak ties). Using multilevel structural equation modeling, we found that FtF interactions and mixed episodes were associated with highest well-being on the within-person level, and that these effects were particularly pronounced for individuals with high levels of neuroticism. CMC was related to lower well-being than FtF interactions, but higher well-being than not socializing at all. Regarding the type of interaction partner, individuals reported higher well-being after interactions with close peers than after interactions with family members and weak ties, and the difference between close peers and weak ties was larger for FtF interactions than for CMC. We discuss these findings with regard to theories of person-situation interactions and research on well-being and social interactions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Who benefits from which activity? On the relations between personality traits, leisure activities, and well-being.

Leisure activities have been emphasized as an important predictor of well-being. However, little research has examined the effects of leisure activity enactment on well-being over time. Moreover, it is unknown which activities are most beneficial for whom. We integrate diverse theoretical accounts of person-environment relations and propose a generic Personality-Activity-Well-Being (PAW) framework, which highlights different relations between personality traits, activities, and well-being. To investigate these relations, we used 11 annual waves from the Longitudinal Internet Studies for the Social Sciences (LISS) panel (total N = 12,703 participants, N = 59,108 assessments), which included measures of the Big Five personality traits, 15 different leisure activities, and affective well-being and life satisfaction. Our preregistered multilevel models revealed three sets of findings. First, we observed on average small expected between-person associations between leisure activities and well-being (e.g., higher average levels of holidays, evening socializing, talking to close others, exercise, and cultural activities were associated with higher well-being). Annual within-person fluctuations in several leisure activities also predicted well-being in expected ways, but effect sizes were very small and varied strongly across participants. Second, personality traits were related to leisure activities in hypothesized ways, yielding on average small but also some moderate and large correlations. Third, Personality Trait × Leisure Activity interactions were only evident on the between-person level, very small in size, and in the opposite direction of our expectations. Personality traits did not moderate well-being benefits from leisure within persons. We discuss the implications of our findings and sketch an agenda for future work. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Evaluating voice samples as a potential source of information about personality.

Speech is a powerful medium through which a variety of psychologically relevant phenomena are expressed. Here we take a first step in evaluating the potential of using voice samples as non-self-report measures of personality. In particular, we examine the extent to which linguistic and vocal information extracted from semi-structured vocal samples can be used to predict conventional measures of personality. We extracted 94 linguistic features (using Linquistic Inquiry Word Count, 2015) and 272 vocal features (using pyAudioAnalysis) from 614 voice samples of at least 50 words. Using a two-stage, fully automatable machine learning pipeline we evaluated the extent to which these features predicted self-report personality scales (Big Five Inventory). For comparison purposes, we also examined the predictive performance of these voice features with respect to depression, age, and gender. Results showed that voice samples accounted for 10.67 % of the variance in personality traits on average and that the same samples could also predict depression, age, and gender. Moreover, the results reported here provide a conservative estimate of the degree to which features derived from voice samples could be used to predict personality traits and suggest a number of opportunities to optimize personality prediction and better understand how voice samples carry information about personality.

Personality-place transactions: Mapping the relationships between Big Five personality traits, states, and daily places.

People actively select their environments, and the environments they select can alter their psychological characteristics in the moment and over time. Such dynamic person-environment transactions are likely to play out in the context of daily life via the places people spend time in (e.g., home, work, or public places like cafes and restaurants). This article investigates personality-place transactions at 3 conceptual levels: stable personality traits, momentary personality states, and short-term personality trait expressions. Three 2-week experience sampling studies (2 exploratory and 1 confirmatory with a total N = 2,350 and more than 63,000 momentary assessments) were used to provide the first large-scale evidence showing that people's stable Big Five traits are associated with the frequency with which they visit different places on a daily basis. For example, extraverted people reported spending less time at home and more time at cafés, bars, and friends' houses. The findings also show that spending time in a particular place predicts people's momentary personality states and their short-term trait expression over time. For example, people reported feeling more extraverted in the moment when spending time at bars/parties, cafés/restaurants, or friends' houses, compared with when at home. People who showed preferences for spending more time in these places also showed higher levels of short-term trait extraversion over the course of 2 weeks. The findings make theoretical contributions to environmental psychology, personality dynamics, as well as the person-environment transactions literature, and highlight practical implications for a world in which the places people visit can be easily captured via GPS sensors. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Multi-modal data collection for measuring health, behavior, and living environment of large-scale participant cohorts.

BACKGROUND: As mobile technologies become ever more sensor-rich, portable, and ubiquitous, data captured by smart devices are lending rich insights into users' daily lives with unprecedented comprehensiveness and ecological validity. A number of human-subject studies have been conducted to examine the use of mobile sensing to uncover individual behavioral patterns and health outcomes, yet minimal attention has been placed on measuring living environments together with other human-centered sensing data. Moreover, the participant sample size in most existing studies falls well below a few hundred, leaving questions open about the reliability of findings on the relations between mobile sensing signals and human outcomes. RESULTS: To address these limitations, we developed a home environment sensor kit for continuous indoor air quality tracking and deployed it in conjunction with smartphones, Fitbits, and ecological momentary assessments in a cohort study of up to 1,584 college student participants per data type for 3 weeks. We propose a conceptual framework that systematically organizes human-centric data modalities by their temporal coverage and spatial freedom. Then we report our study procedure, technologies and methods deployed, and descriptive statistics of the collected data that reflect the participants' mood, sleep, behavior, and living environment. CONCLUSIONS: We were able to collect from a large participant cohort satisfactorily complete multi-modal sensing and survey data in terms of both data continuity and participant adherence. Our novel data and conceptual development provide important guidance for data collection and hypothesis generation in future human-centered sensing studies.

A process-oriented approach to respecting privacy in the context of mobile phone tracking.

Mobile phone tracking poses challenges to individual privacy because a phone's sensor data and metadata logs can reveal behavioral, contextual, and psychological information about the individual who uses the phone. Here, I argue for a process-oriented approach to respecting individual privacy in the context of mobile phone tracking by treating informed consent as a process, not a mouse click. This process-oriented approach allows individuals to exercise their privacy preferences and requires the design of self-tracking systems that facilitate transparency, opt-in default settings, and individual control over personal data, especially with regard to: (1) what kinds of personal data are being collected and (2) how the data are being used and shared. In sum, I argue for the development of self-tracking systems that put individual user privacy and control at their core, while enabling people to harness their personal data for self-insight and behavior change. This approach to mobile phone privacy is a radical departure from current standard data practices and has implications for a wide range of stakeholders, including individual users, researchers, and corporations.

Sensing sociability: Individual differences in young adults' conversation, calling, texting, and app use behaviors in daily life.

Sociability as a disposition describes a tendency to affiliate with others (vs. be alone). Yet, we know relatively little about how much social behavior people engage in during a typical day. One challenge to documenting social behavior tendencies is the broad number of channels over which socializing can occur, both in-person and through digital media. To examine individual differences in everyday social behavior patterns, here we used smartphone-based mobile sensing methods (MSMs) in four studies (total N = 926) to collect real-world data about young adults' social behaviors across four communication channels: conversations, phone calls, text messages, and use of messaging and social media applications. To examine individual differences, we first focused on establishing between-person variability in daily social behavior, examining stability of and relationships among daily sensed social behavior tendencies. To explore factors that may explain the observed individual differences in sensed social behavior, we then expanded our focus to include other time estimates (e.g., times of the day, days of the week) and personality traits. In doing so, we present the first large-scale descriptive portrait of behavioral sociability patterns, characterizing the degree to which young adults engaged in social behaviors and mapping these behaviors onto self-reported personality dispositions. Our discussion focuses on how the observed sociability patterns compare to previous research on young adults' social behavior. We conclude by pointing to areas for future research aimed at understanding sociability using mobile sensing and other naturalistic observation methods for the assessment of social behavior. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Personality trait predictors and mental well-being correlates of exercise frequency across the academic semester.

RATIONALE: Regular exercise is frequently recommended as a means of combating the negative effects of stress on mental health. But, among college students, exercise frequency remains below recommended levels. OBJECTIVE: To better understand exercising behaviors in college students, we examined how exercise patterns change across an academic semester and how these changes relate to personality traits and mental well-being. METHOD: We conducted two longitudinal experience sampling studies, using data from four cohorts of students, spanning four semesters (Fall 2015 - Spring 2017). In Study 1, a large sample of United States college students (cohort 1; N = 1126) reported the number of days they exercised and their levels of happiness, stress, sadness, and anxiety each week over the course of one academic semester (13 weeks). Study 2 (cohorts 2-4; N = 1973) was conducted to replicate our exploratory results from Study 1. RESULTS: Using latent growth curve modeling, we observed the same normative pattern of change across both studies: The average student exercised twice during the first week of the semester and showed consistent decreases in exercise frequency in following weeks. Across both studies, higher initial levels of exercise frequency at the start of the semester were consistently related to higher extraversion, higher conscientiousness, and lower neuroticism. Furthermore, exercise frequency and mental well-being fluctuated together after controlling for time trends in the data: In weeks during which students exercised more than predicted, they also reported being happier and less anxious. CONCLUSIONS: We contextualize the findings with regard to past research and discuss how they can be applied in behavior change interventions to promote students' well-being.

Using Smartphones to Collect Behavioral Data in Psychological Science: Opportunities, Practical Considerations, and Challenges.

Smartphones now offer the promise of collecting behavioral data unobtrusively, in situ, as it unfolds in the course of daily life. Data can be collected from the onboard sensors and other phone logs embedded in today's off-the-shelf smartphone devices. These data permit fine-grained, continuous collection of people's social interactions (e.g., speaking rates in conversation, size of social groups, calls, and text messages), daily activities (e.g., physical activity and sleep), and mobility patterns (e.g., frequency and duration of time spent at various locations). In this article, we have drawn on the lessons from the first wave of smartphone-sensing research to highlight areas of opportunity for psychological research, present practical considerations for designing smartphone studies, and discuss the ongoing methodological and ethical challenges associated with research in this domain. It is our hope that these practical guidelines will facilitate the use of smartphones as a behavioral observation tool in psychological science.