Introductions
The combined EJP and EJPA special issue on New approaches to Personality Conception and Assessment is out now and contains amazing and innovative work on an older question! We had the opportunity to talk with authors of several papers and one of the editors of the special issue about their interest in the research area and their own work.
Specifically, we talked with authors Bell Cooper about her paper titled, “Personality assessment through the situational and behavioral features of Instagram photos“, Gaby Lunansky about her work, “Personality, resilience, and psychopathology: A model for the interaction between slow and fast network processes in the context of mental health“, and Gerard Saucier about his paper, titled “Comparing predictive validity in a community sample: High‐dimensionality and traditional domain‐and‐facet structures of personality variation“. We also chatted with editor David Condon about the special issue he helped put together.
Read more about their research interests and research below.
Q: Hi all, can you tell me a little about yourself and your research?
Bell: I am an Assistant Professor of Quantitative Methods and Business Analytics at Lynn University in Boca Raton, Florida. My research explores all aspects of the personality triad – persons, situations, and behaviors. I examine how personality and situations combine to influence our behaviors and emotions in work, school, online, and everyday life.
Gaby: My research focuses on the concept of resilience from a complex systems perspective. I am a PhD student at the Psychological Methods group from the University of Amsterdam. Currently, I’m working in the PsychoSystems lab of Prof. dr. Denny Borsboom, who is my supervisor together with dr. Claudia van Borkulo. This labgroup introduced and developed the network theory of psychopathology, which states that mental disorders emerge from a network of causally interconnected symptoms. Symptoms can activate each other, leading to a vicious cycle of persistent symptom activity and development of mental disorders. These exciting theoretical proposals are accompanied by novel research methods, allowing researchers from all different backgrounds to work together, analyzing data and building on the theoretical framework.
In my research, I try to deconstruct the concept of resilience to better understand which factors account for its improvement or deterioration. Resilience is a deeply dynamical concept, and static indicators of resilience might be useful to understand differences between individuals, but cannot answer the question of how these differences occur. For this, I study how the symptom network’s parameters might relate to its resilience, by simulating longitudinal data and developing resilience metrics. Currently, I’m also working on expanding existing network models with risk and protective factors. This paints a more dynamic picture of various but specific pathways via which resilience could develop.
Gerard: I’m a professor of psychology at the University of Oregon, USA. My research focuses on the structure of individual differences with respect to personality and various kinds of attitudes, with an emphasis on the impact of moral and cultural considerations.
David: My work is focused on the challenge of figuring out how to capture the many important sources of psychological variability without relying on very broad dimensions. I stumbled on this topic earlier in my career – before becoming a psychologist – while searching for an evidence-based measurement framework that was more informative than the most popular omnibus measures. This line of research has become increasingly central to my identity as I get further into it, in part because I'm passionate about it but also because it has put me in contact with so many other researchers doing interesting science. These include the folks I work with in my current post as an Assistant Professor at the University of Oregon and prior positions at Northwestern University as well as many others around the world.
Q: David, can you tell us a bit about what the Special Issue is about and what readers may expect to learn?
David: The call for papers invited authors to contribute novel approaches to conceptualizing and assessing personality, and we were pleased to receive an enthusiastic response – well over a hundred proposals were submitted! From these, proposals were selected to reflect a diverse mix of promising innovations, and I think this diversity of topics is one of the highlights of the Special Issues.
The innovations range from the use of mobile sensors (Wiernik et al.), social media (Cooper et al.), and video games (Ammannato) as assessment tools to new directions in item-level analyses (Elleman et al.), network psychometrics (Christensen et al., Castro et al.) and dynamic systems (Danvers et al., Sosnowska et al.), advanced modeling (Holtmann et al., Rast et al., Park et al.), evolutionary reasoning (Lukaszewski et al.), the relation between normal and pathological variation (Lunansky et al.)... this is just a sampling! And, nearly all of these articles point towards multiple intriguing topics for further study.
As a collection, the articles in this EJP/EJPA joint special issue as well as the recent EJP special issue pointing "towards a more behavioral science of personality" suggest to me that personality research is entering a really exciting period. I think readers who explore the special issues will get this feeling too.
Q: What makes you especially excited about new approaches to personality conception and assessment?
Bell: More and more evidence is emerging that personality is an important predictor of life outcomes. However, there is still more to be explored in terms of how to comprehensively measure personality. Historically, researchers have relied on survey methods reported by individuals themselves or their acquaintances. With recent technological advancements and the capability for more complex analytical methodologies, we are now able to collect and analyze more personality and behavioral data than ever before, beyond survey methods. Through smartphones, internet browsers, social media, smart devices, and more, people leave behind millions of small digital cues to their own personalities, emotions, behaviors, and everyday situations. These cues are easily accessible and ripe for analyses, which allows us to chip away at the overall puzzle of personality science and ultimately understand more about why people behave the way they do.
Gaby: I am very excited about novel approaches trying to assess personality, using novel theoretical frameworks about what personality is. How we measure and assess personality greatly influences the types of data we can analyze and the models we can estimate. For example, many questionnaires are developed with the goal to design items assessing unidimensional personality traits. This means that all these items are designed to measure the same underlying construct. Contrary, taking a complexity approach means developing items which all measure distinct elements of one entity or emergent property.
Gerard: Older paradigms for conceiving and assessing personality have tended to lead to stagnant and limited approaches to the topic, because they freeze in place ways of thinking generated in (or before) the 1980s, and this has constrained the range and magnitude of prediction afforded by personality assessment. A fresh look at newer paradigms is needed to make understandings of personality more comprehensive and efficiently predictive.
David: This is a tough question to answer briefly – there is so much! That said, the editorial articles (in EJP and EJPA) address this topic very thoroughly. Both of these articles grew out of a meeting in Edinburgh in 2018 and a lot of effort was required to identify areas of agreement among the many authors. The EJP article highlights the tremendous utility of personality assessment for many different purposes, and across different approaches to personality research (descriptive, predictive, and explanatory). It also attempts to explain the benefits of developing models of personality with higher dimensionality than the familiar “Big Few” models.
The EJPA article complements this by suggesting procedures for collaborative development of a more detailed taxonomy of personality traits, one that is constructed from the bottom-up (unlike existing personality hierarchies). In short, we argue for the identification of a comprehensive pool of nuances for psychometric evaluation and, eventually, for use as taxonomic building blocks in personality assessment development. I think this has the potential to extend the utility of personality assessment even further, and it's where my own interests are most focused. I'm excited to see this work conducted over the next several years.
Q: What is your study about?
Bell: Social networking sites such as Instagram have become a popular method in the scientific community for studying personality, behavior, and emotions in recent years. This study demonstrates that we can determine a lot about a person just by viewing a handful of photos that the person posts on Instagram. In this study, we measured the different levels of situational and behavioral content of Instagram photos (e.g., picture of self or landscape, situation displaying positivity or negativity). We examined Instagram photos from a variety of perspectives using human- and machine- detected features of the photos. This means that not only can people glean a lot about someone by just looking at their Instagram posts, but also that computers can easily and instantaneously deduce a wide variety of characteristics about the person who posted the photos.
We also explored both “bright-” and “dark-side” features of people’s personality in their social media posts. People’s bright side characteristics are those that appear normally in everyday life, such as how emotionally-stable, ambitious, agreeable, conscientious, social, and intellectual you are. Dark side characteristics are those that emerge in times of distress. Dark side characteristics include being moody, stubborn, risk-taking, dramatic, eccentric, too meticulous, too eager to please, or narcissistic. This study should really expand our understanding of the relationships between mental health, personality attributes, and social media behaviors. If situational and behavioral features of photos on Instagram can produce a comprehensive personality profile for someone, that profile is likely related to key outcomes across the lifespan, such as relationships, school, job performance, mental health, and more.
Gaby: Our article is about the relationship between personality and psychopathology from a complex systems perspective. We propose personality is a process of slow development, shaping the landscape in which the faster process of psychopathology develops. This is to say that someone’s personality shapes them to encounter certain situations and develop specific behaviors and cognitions, making them more prone to or more resilient against developing a particular set of symptoms. Consequently, one’s history of mental health also feeds back into one’s personality development.
Instead of conceptualizing, for example, neuroticism and depression as two distinct entities, we propose they might be entangled. The points where they touch can be viewed in items which share content overlap, such as the neuroticism item “I get irritated easily” and the Generalized Anxiety Disorder symptom “irritability”. We propose that psychopathology and personality are interconnected in a network architecture, where slow-evolving personality items affect dispositions of fast-evolving symptoms with which they share content overlap. In terms of network models, this means that related personality items strengthen or weaken the symptom’s threshold parameters. In our example, someone who scores high on the “I get irritated easily” item, also has a weaker threshold of the “irritability” symptom, increasing their likelihood to suffer from that symptom when triggered by some event.
In order to outline this idea, we simulate data from empirically estimated neuroticism and depression networks. We establish which items and symptoms are connected to each other by looking at their content overlap. Then, we let participant’s scores on the neuroticism items affect (i.e., increase or decrease) the threshold parameters of related depression symptoms. In this way, each participant has their own set of symptom thresholds. The resilience of these individual symptom networks feeds back into the neuroticism items, increasing or decreasing their neuroticism scores. We iterate this process and show how this model simulates empirical phenomena, such as the strong correlation between neuroticism and depression.
Gerard: To investigate how differing formats and lengths of personality questionnaires vary in their predictiveness, we compared 21 inventories (sets of scales) with respect to prediction of variables related to health and psychopathology. We found that, although the differences in predictive capacity were not huge, there was advantage in questionnaires that have more relatively mutually-independent variables, even if that means having somewhat more items, but not those with more words (i.e., longer items). This finding goes against the norm of focusing on only five or six broad variables and having long and finely specified items in the questionnaire.