An interview
Recently, we talked to René Mõttus, whose paper, "Differences in Personality Characteristics Tends to Increase from Early Childhood to Early Adolescence" is featured in EJP's July/August issue. René currently works at the University of Edinburgh as a lecturer in the psychology of individual differences.
Q: Hi René, can you tell me a little about yourself and your research interests?
I am an individual differences researcher through and through, from when I completed my Msc (2005) to my PhD (2009), to now. I studied at the University of Tartu, did my postgraduate work there with Jüri Allik, then moved to Edinburgh to work with Ian Deary as a post doc. I currently have a permanent position in Edinburgh.
My interests are a little bit all over the place. Currently, the most re-occurring themes are the nature and modelling of psychological constructs, their associations with life outcomes, developmental regularities, and genetic influences. I am also interested in cross-cultural differences and psychopathology.
Q: How did you become interested in personality psychology?
By chance, really. After finishing my Bachelor’s, I spoke with Jüri and he said, “Come and do a MSc on personality”. I was actually embarking on a career as a clinical psychologist at the time, but I went anyway, and ended up staying. I later dropped my clinical ambitions in favour of pursuing research.
Q: What big changes have you seen in the field?
I think that there has been a general move away from simplistic theories based on strong but unquestioned assumptions, towards complexity. At least, there seems to be an increasing willingness to acknowledge complexity. For example, people are openly questioning the underlying assumptions of decades of factor-analytic work, and the value of carrying on doing this work. For instance, there has been a tacit assumption that personality consists of a limited set of invariant “true scores” that we need to locate in a contrived measurement space and then find their causes and consequences. But, are these true scores there at all? Or are we looking for actors from our TV set, like the Karlsson-on-the-Roof?
Another complexity-related change that I have noticed is in the area of behavior genetics and development. People are starting to acknowledge that the nature-nurture opposition is extremely misleading. Genetic influences are entangled with the environment. What we see as, say, heritability at any given time is a developmental outcome rather than some invariant genetic stream of influences that has been there all along. Gene-environment correlations and interactions are no longer fringe concepts, but are making their way into mainstream theories. The idea that we will soon discover the personality genes and that these genetic variants will help us to delineate the neurological systems exclusively underlying particular personality constructs seems to, well, have been quite naïve.
In sum, the representation of personality seems to have become more dynamic: from static snap-shots (like the Big Five scores) towards the unfolding of processes, with individuals interacting with their environment. The increasing use of time-series data such as those coming from experience sampling designs, and the development of methodological concepts and statistical tools that can handle such data attest to this change. Additional evidence of this is the increasing emphasis on the systematic assessment of situations and using these measurements in personality models.
However, I do hope that the pendulum does not really swing to the extreme end of the other direction. We don’t want our ideas about personality to become so specific and complex that we are essentially back at square one, bewildered and overwhelmed. Science is about coming up with general, invariant principles. So I hope that the acknowledgment of personality as a dynamic system in how it develops and plays out in real life eventually leads to an understanding of some general organizational principles rather than an endless catalog of constructs, processes and what have you. Basically, while it may have seemed that the beauty and simplicity of personality is that there are a small number of stable traits – like the Big Five – and everything else revolves around them, I hope that the new generation of simplicity and beauty will yield some simple organizational principles that govern the processes of complex systems such as the development of personality. I actually do think that this is where things are moving.
Q: Could you tell us a little about your paper?
The background is a symposium at the European Conference on Personality in Lausanne, where Brent Roberts was talking about the corresponsive principle -- the idea that individuals select or create environments that match their characteristics (or evoke reactions from the environment that match their characteristics) so that environmental influences reinforce their already existing characteristics. I hadn’t really thought carefully about this idea, but the way Brent described it made me think that if the idea holds, individuals should become less alike as they develop. This is because self-selected, or self-created objective and subjective environments should keep reinforcing pre-existing individual differences. Back in the early 1990s, Jeff McCrae had interpreted an earlier version of the principle similarly in a commentary. He argued, sort of in passing, that empirical data did not support this. I looked for more recent relevant empirical literature and found that there really was none. A few studies had presented standard deviations of personality trait scores for different age groups as descriptive statistics, but as far as I could tell, no study had focused on this in a more substantive way. This was really surprising, because the corresponsive principle appeared to be a quite well established idea, but something that almost inevitably seemed to follow from it had not been tested. Also, most of personality research is actually about variance (individual differences), so quantifying developmental trends in its magnitude seems like quite an obvious thing that one may want to do.
So, we decided to compare variances of trait scores across late adolescence and adulthood in a paper that was published in EJP in 2016. In this study, variance did not seem to differ across these periods, which was not consistent with the corresponsive principle. Then we thought that perhaps the kinds of developmental processes that pertain to the corresponsive principle take place earlier in life. Christopher Soto had collected an excellent set of data for testing this idea: 16,000 children aged between 3 and 18 years, all rated by their parents using the same questionnaire. Therefore, the method was constant. He kindly came on board.
There appeared to be a very consistent pattern whereby trait variance increased from early childhood to about mid-adolescence and then plateaued. It generalized across pretty much every trait and even across items of the questionnaire, and the pattern did not seem to be an artifact of mean-level changes.
I then happened to talk to Helena Slobodskaya about these exciting findings at a conference and she kindly offered a dataset that she had collected, so that we could replicate the findings -- and the findings did replicate quite well -- in a different culture and using a different measure. So it really seems like kids are becoming less alike when they develop.
We are currently following this up and testing whether the increases in variance reflect genetic influences on personality playing out more strongly over time, or increases in environmental influences. Again the findings are surprisingly consistent in that variance tends to increase in childhood in all traits, but I am not going to give you any spoilers on whether the increases in variance reflect genetic or environmental influences. Stay tuned!
Q: What are the most important implications of your study’s findings?
Well, first, there is now a solid empirical phenomenon that needs to be explained, so there is more input for whatever theories there are on personality development.
Second, I think the findings show that a lot of systematic changes in personality – and the processes that bring about these changes – happen much earlier in life than they are typically studied. Increases in the magnitude of individual differences are almost linear from early childhood but seem to have ended by somewhere in puberty, so whatever drives these changes is doing its job quite early, and may even lose its relevance later. If I had to speculate, I see three explanations for this. One is consistent with the corresponsive principle and states that self-selected and self-created objective and subjective environments may amplify characteristics that are already there. A second option is that the development and diversification of affective, cognitive and behavioral repertoires allow individual differences to become more clearly manifested over time. Thirdly, it is also possible that individuals follow an intrinsic developmental trajectory and it may simply be that it takes time to realize it. This would be consistent with the Five-Factor Theory position and would probably also be consistent with increases in genetic variance, but also decreases in environmental variance.
As a third implication, I think studies on mean-level changes may need to factor in our findings. Mean-level changes are interpreted as pertaining to an average individual, but they may actually describe no one or only subsets of people. For example, if there is no mean-level change and even the stability of individuals’ rankings is extremely high, it could be that everyone is still moving – towards their tail of the distribution. This is evidenced by increases in variance.
Q: If you could go back in time, what advice would you give to yourself as a graduate student?
Do not waste your time on irrelevant stuff when it comes to doing research. I got this advice from my supervisor Jüri Allik but it did not help really, as I still did waste a lot of time. It is easy to get lost in details and waste time on things that do not matter in the grander scheme of things. For example, trying out every new statistical or otherwise methodological tool is like an addiction that one may get swamped in entirely. Wasting time still happens to me, all the time.
Q: Some other advice or tips that you would have for young researchers in general?
When you have an exciting idea that seems important, go with it. Immediately, forget everything else. It may go nowhere but then again, unexciting ideas almost certainly go nowhere.
I suspect many quite obvious ideas are still waiting to be expressed and discovered.
Q: Where do you see yourself in the (near) future?
Probably doing the same things, hopefully a little bit better.