A (More) Behavioural Science of Personality in the Age of Multi–Modal Sensing, Big Data, Machine Learning, and Artificial Intelligence

Abstract

Over more than a decade ago, Baumeister, Vohs, and Funder (2007) asked: whatever happened to actual behaviour in psychology? Shortly after, Furr (2009a) published a target article here in EJP on personality psychology as a truly behavioural science and discussed the (low) prevalence, meaning, importance, and measurement of behavioural assessment (see also Furr, 2009b). Now, it is time to take stock: have we, as personality psychologists, moved towards using more behavioural assessment and understanding what people are actually doing in their personal, social, and occupational lives? Has new technology helped us achieve these goals?From the available literature, it would seem that behavioural measurement has been slowly on the rise but also that we still only rarely use actual behaviour as a data source (Nave, Feeney, & Furr, 2018). However, rapid technological advancements, the digitization of daily life, the ubiquity of data-gathering tools (e.g. smartphones), and the introduction of multi-modal sensing methodology, 'big data' applications, machine learning, and artificial intelligence approaches to psychological science promise more and perhaps even better behavioural assessment than was possible decades or even just some years ago. For example, what people speak (e.g. from audio snippets), write (e.g. on online social networks, text-massaging apps, and emails), or do (e.g. captured in videos via minicameras, geospatial movements via GPS, app usage on the smartphone and internet, gameplay, or pictures of activities posted on online social media platforms) in their daily lives yields intensive and massive amounts of data that need to be mined and modelled appropriately (Blake,

Personality computing: New frontiers in personality assessment

“…Personality scientists should especially promote psychometric evaluation and nomological validity of ML data. Accordingly, it is important that personality scientists assist in the interpretation of results in PC research to clarify what aspects of personality are being measured (Rauthmann, 2020), but they should also participate in designing studies and data collection. Moreover, state‐of‐the art knowledge in personality science is essential to move beyond a mere prediction of personality trait scores and to steer PC towards descriptive and explanatory personality research as well as relevant research questions revolving around personality dynamics.…”

Let the Algorithm Speak: How to Use Neural Networks for Automatic Item Generation in Psychological Scale Development

“…In psychology, by far the most common form of assessment is the self-report method (Simms, 2008). Despite criticisms and calls for a greater emphasis on behavioural measures (e.g., Baumeister et al, 2007;Rauthmann, 2020), the persistent popularity of self-report measures is easy to understand: they are flexible and efficient, can be administered anywhere, to a large number of people at once and at little cost. Moreover, given the nature of psychology and its subject of study, self-reports may often be the most direct method to gather information about a construct of interest, such as people's attitudes, thoughts, emotions, actions, or personality traits (Simms, 2008).…”

Descriptive, Predictive and Explanatory Personality Research: Different Goals, Different Approaches, but a Shared Need to Move beyond the Big Few Traits

“…For example, technological progress has provided new sources of information (Rauthmann, 2020). Several recent articles describe how personality and its associations with other variables can be assessed through objectively measured behaviour or digital traces of behaviour (e.g.…”

“…However, direct observations of behaviour have remained comparatively rare in personality research, likely because they are harder to obtain for sufficiently large samples and broad domains of behaviour. We hope that recent technological advances, such as those described in a recent special issue of European Journal of Personality (Rauthmann, 2020), will improve the situation.…”

“…However, by defocusing from broader variability patterns and meticulously studying specific and contextualized behaviours, thoughts, feelings and goals, individual causes of variance may ultimately be identifiable in useful and potentially even controllable ways. Still, such causal explanations may be more complex, phenomenon‐specific and person‐specific than anticipated thus far. Progress in all three areas—description, prediction and explanation—will likely require availability of far higher dimensional models based on traits much more specific than the Big Few, as well as supplementing typical subjective trait‐ratings with alternative sources of information such as informant‐reports and behavioural measurements (Rauthmann, 2020). Therefore, an area with immediate and immense opportunities is developing a new generation of psychometric tools that allow sampling ‘persome’ — the universe of variables capturing personality variability—more broadly than currently available measures do.…”