An interview
We talked with Zoë Francis about her recently accepted paper titled, “Morning Resolutions, Evening Disillusions: Theories of Willpower Affect How Health Behaviors Change Across the Day”, which will be published in an upcoming issue of EJP. Zoë is an assistant professor of psychology at the University of the Fraser Valley, Canada.
Read on to learn more about Zoë’s study on willpower beliefs and health behavior across the day!
Q: Hi there Zoë! Can you tell us a bit about who you are and what made you become interested in studying personality psychology?
I’m from Canada and I first started doing personality psychology in my undergrad. I was interested in the field of psychology, because at my undergraduate university it was the only discipline that was both in the faculty of arts and the faculty of science. In high school, I had always loved both the humanities courses and the science courses, and I saw psychology as being in the middle. And also, people are interesting!
In my third year of my psychology degree, I started the Honors program and chose a lab to work in. I had two options for a lab. I ended up choosing the lab that I did because it had all sorts of different colors on the whiteboard, which made the whole lab seem so fun! The other lab had only one whiteboard color and all of their research seemed the same, so I was like, “Ah, that seems boring”. I started my first research in personality psychology in this lab, which was focused on individual differences in both fish and humans. (I did the human personality side).
When I applied to grad school, I originally wanted to study religion and religiosity and I applied to labs that had some research lines involving these topics. But, since I actually started grad school, I’ve been primarily studying self-control and self-regulation. I’ve only done one study that involved religion – but what I am studying now is interesting too, so it’s not a complaint! It's just funny that my decision points didn’t necessarily have much to do with what followed, though it all turned out really well. But, I think that through it all, I’ve always been interested in why some people are different than other people – that’s something you can see in the world.
Even though the lab I did my PhD in doesn’t have individual differences as a primary focus, it became something I was more and more interested in. I received a Canadian scholarship that allowed me to travel for a semester to work with another researcher. I went to work with Veronika Job, who was in Zurich at the time, and who is also the senior author on this paper. Together we’ve conducted a number of studies and published a few papers that look at individual differences in people’s beliefs about self-control and willpower, and how some people believe and experience self-control as really limited. That’s also where this project came from.
Three weeks ago, I started an assistant professorship at a Canadian university on the West Coast, at the University of the Fraser Valley!
Q: What do you like to do outside of work?
I like a lot of sports. I play volleyball and I’ve been rock climbing and bouldering recently. I also like a sport called ringette that I grew up playing and that I will probably be able to play again now that I’ve moved from Toronto. Most people don’t know about ringette, but it’s originally a Canadian sport that is played on ice. It’s kind of like hockey except the stick is just a straight stick without the blade on it, and there is a blue rubber ring that you pull around and try to score with.
I also like going to restaurants and trying new foods. And everyone likes Netflix and movies and YouTube, so that’s not interesting for me to say…
Q: Can you tell us about your study on willpower beliefs and health behaviors?
On my first or second day with Veronika Job in Zurich, she told me about an archival dataset she had. It was an experience sampling dataset that included measures of people’s snacking behaviours and their physical activity levels (how many minutes of low, moderate, and high intensity physical activity people had engaged in the last couple of hours). It also had a measure of willpower theories, as more limited or non-limited. These willpower theories describe whether people think their self-control runs out, or whether exerting self-control actually gives them more energy to do things that require self-control afterwards.
Initially, she was just like, “This is a great dataset no one has done anything with”. Lavinia Flueckiger, who is also an author on the paper, had collected all of the data during her master’s but had since finished and left her program. Someone else in my lab in Toronto had previously done a study looking at time-of-day and performance on an online learning system, and so I had the idea to use this experience sampling data to look at time-of-day changes in health goal behaviours. Not only was I interested in looking at how people’s level of physical activity changes across the course of the day, but also at whether physical activity changes differently across the day depending on people’s beliefs about willpower. The idea was that people may get more tired throughout the day, but that this might be more true for people who think that their self-control runs out.
I spent a couple of weeks trying to format the data, both because it was experience sampling data (and I had never worked with this type of data before), and also because all of the variable names were in German and I don’t speak German! I eventually sorted out what variable was what with Google translate and help from other people in the lab. The original files and codebook were beautifully structured though; Lavinia had done an amazing job at organizing everything. Otherwise, I would have had no hope!
When I eventually got to the point where I was able to test my research questions, I did find that physical activity changed across the course of the day, where people were becoming more physically active. This was especially true for people with more non-limited willpower theories, who thought that they got energized. People with more limited willpower theories also became relatively more active during the day, but with a more gradual slope. So I found moderation by willpower theory. I also found the same thing to be true for how many snacks people had eaten in the previous two hours, where someone with a more limited willpower theory ate more and more and more snacks across the day. Someone with a relatively less-limited theory also ate more snacks later than earlier, but less so. I also looked at the same interaction for how many servings of fruits and vegetables people ate, but I didn’t find a moderation for that.
Then, we tried to submit this article to three journals and received three rejections. One problem was that this dataset had some real limitations in terms of missing data, maybe since – as an experience sampling study – the questionnaires were a little bit too long and people hadn’t completed the surveys as often as we would have hoped. We decided to add a new dataset and re-write the paper. Veronika actually had a second experience sampling dataset available that had many of the same variables (i.e., willpower theory, snacks, and physical activity). So then I had a new German dataset to sort out, and we did it all again. The physical activity finding replicated in the second dataset, but the snacking moderation did not. In this second dataset, we found that limited willpower theorists snacked more all the time across the whole day.
It should be mentioned that the second dataset was based on quite a different sample. The first sample was a community sample, almost entirely women, of people who had responded to advertisements saying that they were unhappy with their exercise or eating behaviors. So basically, they had some health-related goals. The sample in which we were trying to replicate our findings was a student sample of undergraduates, most of whom didn’t necessarily have any health goals. For them, snacking might not have necessarily been a bad thing that they were trying to avoid.
So, we replicated the physical activity finding but not the snacking finding. We saw this as there being better evidence that at least changes to physical activity across the course of the day are moderated by willpower theories. This was relevant because a previous study or two had shown that people with more limited willpower theories tended to have higher Body Mass Indices (BMI). Our finding may be one potential reason for why a limited willpower theorist might have a higher BMI; they may be more sedentary, and especially more sedentary later in the day.
An alternative explanation for our findings might be that people change their intentions across the day. However, it seemed that people with more limited willpower theories set more modest goals for themselves across the course of the day: they didn’t expect to walk as much and then they didn’t. So it didn’t seem so much like a self-control failure (setting the goal and then failing to meet the goal); they actually weren’t even intending to be active.
I think one important implication of our research is that time-of-day effects can matter. Time-of-day isn’t looked at very often, but it’s an interesting factor to think about. So much of what we do is tied to specific times of day – mood and sleepiness fluctuate across the day, work and socializing tend to happen between particular times of day, and so do food and exercise. So, one take-away is that time-of-day effects are interesting and relevant for social and personality psychologists, and it’s relatively straightforward to examine these effects when you have experience sampling data. That’s something I would like to see a little bit more of.
And then the second one, more directly related to the main findings, is that willpower theories do seem to affect how physically active people are across different times of the day, and that willpower theories matter more at some times of day than at other times. In the morning, willpower theory might not predict physical activity, but willpower theory does matter later in the day.
Q: What’s next for you?
Well, my new job! I’ll be starting a lab that studies willpower theories and other individual differences, as well as self-regulation. I’m still interested in the fluctuations that people perceive and experience in mental energy and self-control, and also how people structure their environments to help them succeed and to accommodate themselves. I will also be teaching more, starting with Introduction to Psychology, and then also teaching statistics and social psychology classes.
In terms of research, I would say a lot of my future work overlaps with constructs used in this present research. I have other research looking at time-of-day effects, and I certainly expect that I will circle back around to looking at snacking and physical activity and these kinds of health-related behaviors again in the future. But I personally don’t have short-term plans to directly follow up on the present studies.
I have specific studies that I’m planning but they are a little bit further removed from this particular line of work, so I won’t go too much into them.
Q: Do you have any tips or advice for young researchers?
I am a big proponent of automating almost anything that you can possibly automate. I conduct all of my analyses in R, which is a really great program to use. When you are doing tasks more than three or five times, it’s helpful to learn how to write a simple program to do that for you. Research is about avoiding mistakes and – when you do make mistakes – about finding them, fixing them, and then avoiding making more. When you have written code, you can see what you’ve done and go back and look. So there’s a lot more accountability and you can have other people look over your work. And it’s such a time saver! Even if you’re in a discipline that doesn’t seem like you need any programming, learning just the basics is hugely beneficial.
Similarly, whenever someone is starting grad school, I tell them to use a reference manager. Whenever I hear someone spending hours changing the formatting of their citation style, I’m like “No! You can do that with the click of a button in reference software!”. So, along the lines of automating, I’m a big fan of reference managers.
More psychologically, knowing that sometimes the path to a publication can be really difficult and non-linear is also important. For example, we had the current paper rejected a few times, added a whole new study, and replicated some but not all of the findings. And then we ended up, of course, having the article accepted at a really great journal where people will be aware of our work. The quick papers aren’t always the most rewarding. And after you’re working on a paper for four years, you’re always happy to see it leave.
Finally, if I could go back to the start of my PhD, I would maybe let my future-self know that – just like how a paper can be non-linear – it’s very hard to know how long grad school is going to take. Luck is involved in academia and publishing to some extent, but I think with sufficient time, effort, and flexibility things should turn out well in the end. Originally when I started grad school, I thought that I was going to get my PhD done in four years. I ended up in grad school for an extra two years – which isn’t uncommon in North America, but I didn’t see it coming – and it ended up being great! I didn’t do a postdoc but managed to start an assistant professorship immediately out of grad school. I had time to incubate papers that took a long time, some of which didn’t have linear roads. As a case in point, my master thesis took about three to four years to publish, because we kept on running studies and kept on not really replicating and then replicating and then not replicating and then not replicating. And in the end, my first empirical paper is twelve studies that I meta-analyzed, and the title includes the phrase, “inconclusive results”! Extra time and perseverance when things are not working out is needed, because you can’t predict what’s going to happen.