Mirror on the wall, who's the best at mirroring?
The better you mirror each other's behavior, the better you appear to work together. In her PhD research at the unit Cognitive Psychology in Leiden, Friederike Behrens has developed a measure to capture the dynamic process of mirroring in numbers. PhD defense on 28 October.
Laughing when the other laughs, we all recognize that. Or adopting someone's posture or gestures; mirroring each other's behavior usually happens unconsciously. And if your body responds, for example with palpitations and sweating, you have little control over it. But what does that say about the people having the conversation? Because Friederike Behrens was curious about how people mirror each other and what that means for their collaboration, she researched measures of heart rate and skin conductance.
Measuring heart rate and skin conductance
For the study, Behrens and her team measured the heart rate and skin conductance of participants during a two player game in the lab. The participants earned points in this game, whereby the choice to collaborate yielded the most points for both together. However, if a participant chose for himself, he had the chance to obtain more points individually. How do the participants behave in this prisoner's dilemma? Behrens: 'We measured the participants during several rounds, in which they saw or did not see each other and where they were or were not given information about the social behavior of the other. Each time the participants could choose whether they wanted to collaborate or not. It was a successful collaboration if both chose the option to work together. Those participants also mirrored each other the most in the measured values for skin conductance.
International collaboration project
How can you best quantify this mirroring behavior? Behrens brought that question to the US, in a joint research project with Professor Steven Boker, University of Virginia in Charlottesville. She reached out to him on the basis of his earlier research. 'Nobody here knew Boker, but I thought: I need that man. He received me very well and I look back on the collaboration with pleasure. ' In the project with Boker she was able to develop her own direction. 'Sometimes I mirror you, sometimes you mirror me. Sometimes we mirror each other well, sometimes less. You have to include these dynamic components of the interaction into a statistical analysis.’
Comparable research results
Behrens has developed a statistical standard measurement method for mirror behavior, using the Windowed Cross-Correlation method. 'You specify certain parameters in order to tailor the method to the measurement you are viewing. I have looked at which values are the best for the parameters heart rate, skin conductance, facial expressions and pupil changes. Based on those values, I have drawn up guidelines, after biological considerations of how a certain measurement behaves. For example, how quickly do you see changes in your heart rate and how long does a 'typical' heart rate response last. The guidelines can be used in other mirror studies so that the results are mutually comparable.’
Quantifying mirror behaviour
Behrens sees the statistical measurement method as the most important contribution of her doctoral research to science. Studies of mirror behavior and small changes in physiological measurements are not new in themselves. The link to collaboration has also been established before. But she thought new research was desperately needed, given the differences in outcomes and methods used so far. 'We put two people facing each other, instead of in front of a computer. We have made a lot of progress, especially in the quantification of mirror behavior. Scientists can build on this in new research into mirror behavior. ' The new PhD candidates in the research group of cognitive psychologist Mariska Kret, of which Behrens was a PhD candidate, are now doing the same.
'As a PhD candidate you start in a dark environment and you get a flashlight in your hand to explore it. In the process you get to know the environment better and better, so that you can show other people the way ', Behrens states. She brings her knowledge of statistical analysis with her to her new job at Sogeti, a secondment company in ICT and part of Capgemini. She hopes to employ her psychological background in this field. In times of corona she is also working from home in front of her PC screen. 'That affects collaboration, although it is not the case that all collaboration collapses when you look at each other digitally. You do miss a lot of subtle information though. And of course, mirroring becomes a lot more difficult. '
The Leiden Universitair Fonds (LUF) was willing to sponsor Behren's international collaboration, with as a result the pre-print. The phase before peer review.
Quantifying Physiological Synchrony through Windowed Cross-Correlation Analysis: Statistical and Theoretical Considerations