ACTWELL: Activities, Contexts, and Traits in Well-Being in Everyday Life Longitudinally

Our subjective well-being has a profound influence on peoples’ individual flourishing, but also for society as a whole. The ACTWELL project is a research initiative dedicated to understanding well-being in our daily lives. We are harnessing the power of modern smartphone technology, psychological research, and machine learning to unravel the complexities of our well-being and inform effective strategies for enhancing it.

Background

The pressing discrepancy between objective and subjective well-being in the world.

Well-being is an important life goal for many people. While objective indicators of quality of life (e.g., income) have risen in many regions over the past years, subjective well-being (SWB), encompassing one’s life satisfaction and everyday emotional experience, has been steadily declining around the world. Exacerbated by the COVID-19 pandemic, rising levels of loneliness, depression, and anxiety have placed substantial strains on individuals causing significant societal and economic costs.

Mission

Better understand well-being in daily life to inform effective and scalable interventions.

Our mission is to decode the complex dynamics of subjective well-being in everyday life and develop data-driven strategies for enhancing it. By leveraging extensive smartphone datasets and cutting-edge machine learning techniques, we aim to identify the behavioral, contextual, and dispositional factors influencing our well-being and provide actionable insights for monitoring, support, and intervention.

The ACTWELL Approach

Harnessing artificial intelligence to decipher everyday well-being from big smartphone data.

The ACTWELL project leverages multiple extensive (“big data”) smartphone datasets, containing behavioral (e.g., use of social media apps) and contextual data (e.g., time spent in nature) as well as trait self-reports (e.g., on personality), from project partners in Europe and the United States. Based on that vast data, we use machine learning algorithms (“artificial intelligence”) to predict well-being from individuals’ activities, contexts, and traits.

Research Agenda

As a first step, we identify those behaviors, contexts, and traits that are expected to be linked to subjective well-being from the existing literature. Next, we pre-process the raw smartphone logs from the collected data sets so that they can be used to train machine learning algorithms to predict well-being from individuals’ activities, contexts, and traits. Here, we also investigate how the behavioral and contextual factors contributing to one’s well-being vary across people, for example, due to individual differences (e.g., age, personality traits). Lastly, we examine how changes in sequences of activities and contexts influence well-being over time.

Funding and Timeline

The ACTWELL project is a four-year initiative funded by the Swiss National Science Foundation (SNSF) from 2023 until 2027 with CHF 801’602,-.

Research Team

The ACTWELL project is based at the Institute of Behavioral Science and Technology (IBT) at the University of St. Gallen and supported by our team and external collaborators.