An Interview with 2019 Tanaka Dissertation Award Winner Aaron Weidman
1. In a nutshell, what was your dissertation about?
I am fascinated with how our methods shape our theory: What we conclude about people from any study is contingent on how we conduct that study. The best way to arrive at sound theoretical conclusions is to first vet our measures and manipulations through rigorous construct validation. In my dissertation, I highlight this often-overlooked link between methods and theory in the context of emotion research, an area in which little attention has historically been paid to issues of construct validation. For example, I show that the field's cumulative knowledge about the emotion humility (i.e., that it is a virtue) is based on a narrow conceptualization of humility, which in fact has two distinct dimensions, involving both appreciation and self-abasement. Similarly, I show that our knowledge about the link between happiness and spending (i.e., you should buy experiences, not material things) is based on a narrow conceptualization of retrospective, "afterglow" happiness; when we assess happiness in the moment, spending money on things comes out looking much better. Finally, I apply some of the first construct validation techniques in the domain of positive emotions, using bottom-up methods to derive lay-person definitions of each regularly studied emotion (e.g., awe, gratitude, sympathy, pride) and to develop self-report scales to measure each of these states. I hope that this work can provide a solid methodological foundation on which to build knowledge about positive emotions.
2. What are you working on right now, and what do you want to do in the future?
Building on my fascination with methods, I have recently been thinking about means of assessing emotion without relying on self-report surveys. Self-report surveys provide a great window into people's subjective experience (what better way to know how someone feels than to ask them?). Yet, they are cumbersome to complete and completing them can interfere with the subjective experience we wish to capture. Along with several colleagues, I have recently worked on a project in which we try to harness emergent technologies (e.g., smartphone sensing, machine learning) to predict people's mood based on the sound of their voice. We have thus far found this issue tough to tackle: It turns out that mundane, everyday conversation doesn't provide a ton of signal into how someone is feeling. Furthermore, most people feel relatively neutral — as opposed to very happy or very sad — most of the time, which further reduces our ability to pick up on fluctuations in mood as people go about their daily lives. We have only scratched the surface, however, in terms of the types of data we can use to gain insight into mood (e.g., physiological data, speech content, activity/movement data) and the predictive methods for tackling these kinds of problems are constantly evolving and improving. As a result, I am extremely excited to continue work in this area. Maybe someday we will be able to partially supplant our reliance on self-report surveys of emotion by using technology-based assessment tools!
3. What research or statistical methods are you most excited to see pursued in our field in the coming years?
In line with what I wrote above, I am most excited about methods that capture people's in vivo experiences (e.g., experience-sampling, smartphone sensing) and statistical techniques that can handle the vast amounts of data that these in vivo research methods yield (e.g., multilevel modeling, machine learning). I would argue that the two major methodological engines of social-personality psychology in the 20th century were the self-report survey and the laboratory experiment (think of a groundbreaking finding or literature from your intro textbook and I bet it was built on one of these two methods). These methods are still invaluable and foundational, but as a field we are now much more aware of their limitations. Self-report surveys are subject to many biases (e.g., acquiescence, self-presentation) and we know that people have blind spots in their self-knowledge. Laboratory experiments of course typically lack ecological validity and, furthermore, it is logistically very difficult to recruit the kinds of sample sizes that we now know are required to adequately power our studies when you are bringing participants into the lab one-by-one. Moving forward I am excited to see our field increasingly rely on methods that capture people's experiences repeatedly as they go about their daily lives, which has the twin benefits of increasing ecological validity and statistical power. The upshot of all this might be a larger proportion of descriptive research (vs. causal inference) but I imagine personality psychologists will be receptive to this shift!
4. Do you have any advice for grad students? What was the best advice you got and helped you?
I like stepping back and asking myself what I am accomplishing and what my research goals are. On a day-to-day level, I really like the concept from industry of the "stand-up", where a team will meet at the end of the day and each person will stand up and share what tasks they accomplished that day. I think graduate students can benefit from having a private stand-up with themselves each day, asking what they accomplished to "move the needle forward" on their research. Maybe you drafted an Introduction section, wrote an IRB proposal, or learned a new statistical technique for your data. Productivity can come in many forms (but checking Twitter probably doesn't count)! A daily "stand-up" can help you keep track of what you are getting done and what you might need to better prioritize. On a higher level, my doctoral adviser Jess Tracy always stressed the need to think about the "theoretical contribution" of any research project. I think there is tremendous wisdom in this advice, even if we take it more broadly to mean that we have to stop and ask ourselves what exactly is the point of a project: How are we advancing our knowledge about personality, or how are we providing a methodological innovation that itself will advance our ability to understand personality? My sense is that if graduate students ask themselves these questions before diving into a research project, it will help focus their efforts and make them more efficient in pumping out impactful work.