Pupillary Contagion in a Developmental Context
How infants relate to others in their social environment is a fundamental question in psychology. Infants pay particular attention to people from very early in life, but what information are they sensitive to and how do they learn to respond to others' emotional states? One possible mechanism that could link states of arousal across individuals is pupillary contagion. This phenomenon - in which the pupil of an observer increases with the pupil size of someone they observe - has been demonstrated in adults and recently in infants as well in our own research (in press, Psychological Science). In adults, pupillary contagion is related to empathy, trust, and group membership, yet the underlying mechanism could still be a transfer of arousal between two people. That is, seeing that someone else is aroused based on the dilation of their pupils leads to arousal in the self, which leads to pupil dilation. The current literature makes clear that pupillary contagion is a robust phenomenon in adults and now has been observed in infants as well. What is not yet known are the mechanisms that drive pupillary contagion. In the current project, we will use a developmental perspective to test the hypothesis that pupillary contagion is initially driven by underlying arousal, and later becomes modulated by social factors and integrated into a network of social-cognitive skills.
Final report
In this project, we used a developmental perspective to examine the phenomenon of pupillary contagion: when the pupil of an observer dilates with the pupil size of someone they observe. The initial aims of the project were: 1) to show that early pupillary contagion is driven by arousal stemming from perception of eyes with dilated pupils, 2) to examine the role of social modulation in pupillary contagion, and 3) to uncover the developmental precursors and outcomes of pupillary contagion.
During the first year of the project, it was determined that based on recently published work from other research groups that Aim 1 was no longer a critical open question. Several studies were run to explore the social modulation of pupillary contagion (Aim 2). That is, what factors might enhance or suppress humans’ automatic tendency to engage in pupillary contagion with another individual? Aim 3 was addressed with a longitudinal study, the exact design of which was changed slightly based on the need to accommodate new timelines and recruitment issues given the covid-19 pandemic. That is, the sample size was increased from 60 to 75, but the number of time points in the study was decreased from 5 to 3 (at 5, 9, and 15 months instead of 3, 6, 9, 12, and 15 months).
A new topic that was investigated during the project is the pupil’s role in social learning, a theme in line with the overall aim of the project to examine the development of pupillary contagion and its role in social-cognitive development. Specifically, the study used pupil dilation as an outcome measure for infants’ social learning of music preferences. Here, we examine whether infants’ pupil dilation can be used to assess whether they share another person’s negative reaction to certain pieces of music (Fawcett & Kreutz, 2021).
Another aspect of the project that was not originally planned was investigations into adults’ pupillary contagion, particularly in relation to individual differences and generalization of the phenomenon across species. These topics were determined to contribute to the broader aims of the study concerning under what circumstances pupillary contagion is likely to occur and complement the findings that we have from the developmental studies.
Implementation
During the project, in my role as primary investigator, I designed and lead the implementation of six research studies in collaboration with other researchers and with the help of project assistants. Studies were carried out primarily at the Uppsala Child and Baby Lab using eye tracking technology, as well as questionnaires, with infants between 5 and 15 months of age. One study with adults was run in Uppsala and one was run in Finland, in collaboration with the FinnBrain project team.
Results and contributions
One of the most important findings of the project concerns social modulation of pupillary contagion. That is, we uncovered characteristics of the observed individuals that do and do not impact pupillary contagion. In one study, we showed that adults experience pupillary contagion when viewing images of cats and dogs, not only humans (Axelsson & Fawcett, 2021). Specifically, adults viewed images of cats, dogs, and humans with large, medium, and small pupils. For all three species, participants showed a pupillary contagion effect and it was not moderated by individual differences in participants’ liking of or experience with cats or dogs. The results suggest that pupillary contagion is a robust phenomenon that can occur across familiar species. For infants, however, we found that social similarity impacts their pupillary contagion. That is, for 10-month-old infants, pupillary contagion does not occur when viewing someone who speaks a language that they are not familiar with. This finding suggests that infants are more likely to share arousal with those they are similar to and may resist doing so with those who belong to a different social group.
Our findings also showed that the tendency to experience pupillary contagion can also depend on individual differences in observers. For adults, pupillary contagion can vary based on one’s socio-economic status (SES). That is, for individuals with high SES, pupillary contagion is less likely than for those with lower SES, regardless of whether the observed person had a neutral or emotional facial expression. We reasoned that SES is linked to social power and status, making it less likely that one would need to be in tune with the arousal states of others if they are of high status, in line with other work on social status and emotional reactivity. In the same study, we found no differences in pupillary contagion across different levels of anxiety, depression, and sleep problems in participants, suggesting that the phenomenon is largely robust to individual differences. In the same sample of adults, we also examined alexithymia and found that while it did not relate to pupillary contagion, there was an intriguing interaction between gender and alexithymia for overall pupillary dilation to the images of eyes. That is, women with higher levels of alexithymia tended to show greater pupil dilation to the images, while for men, higher levels of alexithymia predicted less pupil dilation to the images. We speculate that this could be related to societal pressure on women than men to be more sensitive to others’ emotions such that seeing the eyes led to greater arousal and cognitive load as they struggled to interpret the social signals presented.
Finally, the project contributed to our understanding of the development of pupillary contagion. First, we found that, in contrast to what was observed in adults (Fawcett et al., 2021), 9-month-olds do not engage in pupillary contagion when viewing emotionally-expressive eyes (Fawcett, in preparation). Specifically, while there were differences in overall pupil dilation to different emotional expressions (e.g., greater dilation to anger), the pupil size of the observed individual did not impact infants’ own pupil size. We propose that for infants, the emotional expressions are already a strong signal of arousal levels in the observed individual and the salience of that expression is enough to keep them from reacting to the pupil size differences. In the longitudinal study, we did not find stability in the degree of infants’ pupillary contagion from 5 to 9 months of age. We also did not find a relationship between infants’ pupillary contagion and their later prosocial behavior. We did however find a relationship between parents’ socialization of prosocial behavior and infants’ tendency to engage in prosocial behavior at 15 months of age, suggesting that parents’ socialization may be a more important factor for early prosocial behavior than infants’ tendencies for arousal sharing, as measured by pupillary contagion.
New research questions
First, regarding individual differences, future research should examine what other factors might be impactful for pupillary contagion, especially among infants and children, since most of this work so far concerned adults. Second, what social factors are enough to create boundaries on pupillary contagion. Why is a foreign language enough to restrict infants’ pupillary contagion, while adults show the phenomenon for cats and dogs? Finally, the findings from the longitudinal study open up further questions regarding early contributors to the development of prosocial behavior.
International aspects
During the project, new collaborations were established with researchers in Australia, Germany, and Finland. When Emma Axelsson was working in the Uppsala Child and Baby Lab as a researcher on another project, she became interested in the pupillary contagion work. We began a collaboration to examine adults’ pupillary contagion across species (Axelsson & Fawcett, 2021). Emma is now a Lecturer at the University of Newcastle in Australia. In the future, we hope to extend this work to infants and children.
A second international collaboration is with Gunter Kreutz, a professor of music psychology at the University of Oldenburg. Together we designed and carried out the project on social learning of music preferences.
A third international collaboration is with the FinnBrain project in Turku, Finland. This project is an ongoing, longitudinal project examining environmental and genetic factors in child development. The researchers were interested in including social perception measures in their tasks for children and parents and approached me about including pupillary contagion. We decided to include both neutral and emotional faces with varying pupil sizes and are currently collecting data from the participating children’s parents.
Dissemination
Results from the project were presented at a conference in Austria (Interpersonal Synchrony) and two remote conferences (International Conference on Infant Studies and Deutsche Gesellschaft für Musikpsychologie). Invited talks were given at Stockholm University’s Department of Linguistics, Karolinska Institute’s Department of Psychology and Children’s Hospital, and Tobii eye tracking.
Four studies from the project have already been published in top international journals and three additional manuscripts are in preparation. In addition to publishing the work in academic journals, the results have been disseminated on social media (Twitter, Facebook) and on the Uppsala Child and Baby lab webpage for interested parents, academics, and other members of the community.
During the first year of the project, it was determined that based on recently published work from other research groups that Aim 1 was no longer a critical open question. Several studies were run to explore the social modulation of pupillary contagion (Aim 2). That is, what factors might enhance or suppress humans’ automatic tendency to engage in pupillary contagion with another individual? Aim 3 was addressed with a longitudinal study, the exact design of which was changed slightly based on the need to accommodate new timelines and recruitment issues given the covid-19 pandemic. That is, the sample size was increased from 60 to 75, but the number of time points in the study was decreased from 5 to 3 (at 5, 9, and 15 months instead of 3, 6, 9, 12, and 15 months).
A new topic that was investigated during the project is the pupil’s role in social learning, a theme in line with the overall aim of the project to examine the development of pupillary contagion and its role in social-cognitive development. Specifically, the study used pupil dilation as an outcome measure for infants’ social learning of music preferences. Here, we examine whether infants’ pupil dilation can be used to assess whether they share another person’s negative reaction to certain pieces of music (Fawcett & Kreutz, 2021).
Another aspect of the project that was not originally planned was investigations into adults’ pupillary contagion, particularly in relation to individual differences and generalization of the phenomenon across species. These topics were determined to contribute to the broader aims of the study concerning under what circumstances pupillary contagion is likely to occur and complement the findings that we have from the developmental studies.
Implementation
During the project, in my role as primary investigator, I designed and lead the implementation of six research studies in collaboration with other researchers and with the help of project assistants. Studies were carried out primarily at the Uppsala Child and Baby Lab using eye tracking technology, as well as questionnaires, with infants between 5 and 15 months of age. One study with adults was run in Uppsala and one was run in Finland, in collaboration with the FinnBrain project team.
Results and contributions
One of the most important findings of the project concerns social modulation of pupillary contagion. That is, we uncovered characteristics of the observed individuals that do and do not impact pupillary contagion. In one study, we showed that adults experience pupillary contagion when viewing images of cats and dogs, not only humans (Axelsson & Fawcett, 2021). Specifically, adults viewed images of cats, dogs, and humans with large, medium, and small pupils. For all three species, participants showed a pupillary contagion effect and it was not moderated by individual differences in participants’ liking of or experience with cats or dogs. The results suggest that pupillary contagion is a robust phenomenon that can occur across familiar species. For infants, however, we found that social similarity impacts their pupillary contagion. That is, for 10-month-old infants, pupillary contagion does not occur when viewing someone who speaks a language that they are not familiar with. This finding suggests that infants are more likely to share arousal with those they are similar to and may resist doing so with those who belong to a different social group.
Our findings also showed that the tendency to experience pupillary contagion can also depend on individual differences in observers. For adults, pupillary contagion can vary based on one’s socio-economic status (SES). That is, for individuals with high SES, pupillary contagion is less likely than for those with lower SES, regardless of whether the observed person had a neutral or emotional facial expression. We reasoned that SES is linked to social power and status, making it less likely that one would need to be in tune with the arousal states of others if they are of high status, in line with other work on social status and emotional reactivity. In the same study, we found no differences in pupillary contagion across different levels of anxiety, depression, and sleep problems in participants, suggesting that the phenomenon is largely robust to individual differences. In the same sample of adults, we also examined alexithymia and found that while it did not relate to pupillary contagion, there was an intriguing interaction between gender and alexithymia for overall pupillary dilation to the images of eyes. That is, women with higher levels of alexithymia tended to show greater pupil dilation to the images, while for men, higher levels of alexithymia predicted less pupil dilation to the images. We speculate that this could be related to societal pressure on women than men to be more sensitive to others’ emotions such that seeing the eyes led to greater arousal and cognitive load as they struggled to interpret the social signals presented.
Finally, the project contributed to our understanding of the development of pupillary contagion. First, we found that, in contrast to what was observed in adults (Fawcett et al., 2021), 9-month-olds do not engage in pupillary contagion when viewing emotionally-expressive eyes (Fawcett, in preparation). Specifically, while there were differences in overall pupil dilation to different emotional expressions (e.g., greater dilation to anger), the pupil size of the observed individual did not impact infants’ own pupil size. We propose that for infants, the emotional expressions are already a strong signal of arousal levels in the observed individual and the salience of that expression is enough to keep them from reacting to the pupil size differences. In the longitudinal study, we did not find stability in the degree of infants’ pupillary contagion from 5 to 9 months of age. We also did not find a relationship between infants’ pupillary contagion and their later prosocial behavior. We did however find a relationship between parents’ socialization of prosocial behavior and infants’ tendency to engage in prosocial behavior at 15 months of age, suggesting that parents’ socialization may be a more important factor for early prosocial behavior than infants’ tendencies for arousal sharing, as measured by pupillary contagion.
New research questions
First, regarding individual differences, future research should examine what other factors might be impactful for pupillary contagion, especially among infants and children, since most of this work so far concerned adults. Second, what social factors are enough to create boundaries on pupillary contagion. Why is a foreign language enough to restrict infants’ pupillary contagion, while adults show the phenomenon for cats and dogs? Finally, the findings from the longitudinal study open up further questions regarding early contributors to the development of prosocial behavior.
International aspects
During the project, new collaborations were established with researchers in Australia, Germany, and Finland. When Emma Axelsson was working in the Uppsala Child and Baby Lab as a researcher on another project, she became interested in the pupillary contagion work. We began a collaboration to examine adults’ pupillary contagion across species (Axelsson & Fawcett, 2021). Emma is now a Lecturer at the University of Newcastle in Australia. In the future, we hope to extend this work to infants and children.
A second international collaboration is with Gunter Kreutz, a professor of music psychology at the University of Oldenburg. Together we designed and carried out the project on social learning of music preferences.
A third international collaboration is with the FinnBrain project in Turku, Finland. This project is an ongoing, longitudinal project examining environmental and genetic factors in child development. The researchers were interested in including social perception measures in their tasks for children and parents and approached me about including pupillary contagion. We decided to include both neutral and emotional faces with varying pupil sizes and are currently collecting data from the participating children’s parents.
Dissemination
Results from the project were presented at a conference in Austria (Interpersonal Synchrony) and two remote conferences (International Conference on Infant Studies and Deutsche Gesellschaft für Musikpsychologie). Invited talks were given at Stockholm University’s Department of Linguistics, Karolinska Institute’s Department of Psychology and Children’s Hospital, and Tobii eye tracking.
Four studies from the project have already been published in top international journals and three additional manuscripts are in preparation. In addition to publishing the work in academic journals, the results have been disseminated on social media (Twitter, Facebook) and on the Uppsala Child and Baby lab webpage for interested parents, academics, and other members of the community.