Can wise empathy training improve well-being during Instagram use?
Social media can lift people’s mood or leave them feeling worse, and much of the difference may lie in how they respond empathically to the emotions they see in other people’s posts. Sharing a friend’s joy is one response. Comparing yourself with them, or becoming overwhelmed by their distress, is another.
The Wise Empathy study tested whether a short training could change these responses during people’s everyday Instagram use. Over 8 days, 260 adult Instagram users completed five daily check-ins through the Avicenna app. Halfway through the study, one group received training in sharing positive emotions, responding to negative emotions with compassion rather than personal distress, and avoiding social comparison. The other group completed a control activity about the history of social media.
By collecting repeated reports of Instagram use, empathic responses, and well-being, the researchers could examine whether the training changed how participants experienced social media in their everyday lives.
The study was run by researchers in the Department of Psychology at the University of Toronto, with collaborators at the Rady School of Management, University of California San Diego.
An intervention’s effect can be hidden if the surveys do not record the moment it targets.
Study at a Glance
| Study feature | Detail |
|---|---|
| Study design | Randomized ecological momentary intervention with experience sampling |
| Population | Adults with active Instagram accounts, mostly in the United States (92%) |
| Participants | 300 recruited; 260 included in the analysis (134 intervention, 126 control) |
| Recruitment | Cloud Connect, an online participant recruitment platform |
| Study period | 8 days per participant |
| Experience sampling | Five check-ins a day at random times within five windows between 10:00 a.m. and 9:45 p.m. |
| Intervention | Midpoint survey with two short videos, then three reflection surveys a day for the last 3 days |
| Observations | 5,902 check-ins, about 23 per participant |
| Compensation | Gift cards of $8, $15, or $20, or $20 plus entry to win $250, depending on how many surveys were completed |
| Tools | Avicenna app, R |
| Preregistration | AsPredicted, for both the empathy and the effort analyses |
| Ethics approval | University of Toronto Research Ethics Board, protocol 00036941 |
How the Study Worked
Enrollment
Participants read the consent form, confirmed their eligibility, and downloaded the Avicenna app on their own phones. On the first day, they completed a welcome survey with demographic and individual difference measures. They were randomly assigned to the intervention or control condition.
Five Check-Ins a Day
For the next 6 days, a check-in arrived at a random time within each of five windows: 10:00 to 11:45 a.m., 12:15 to 2:00 p.m., 2:30 to 4:15 p.m., 4:45 to 6:30 p.m., and 7:00 to 9:45 p.m.
Each check-in opened with two questions:
- Instagram use: Had they used Instagram in the last 30 minutes, and if so, for how long?
- Exposure to emotions: How much had they been exposed to other people’s emotions in the last 30 minutes?
The remaining questions depended on their answers. People who had used Instagram rated the emotions they had seen there, and their empathy questions were worded around browsing (“While browsing Instagram in the last 30 minutes…”). People who had not used Instagram but had encountered others’ emotions answered the same questions about offline experiences. People who had done neither rated their own emotions and then answered filler questions of similar length.
Everyone then rated their well-being: how joyful, positive, and content they felt, their life satisfaction, and their sadness and stress. The researchers combined these into one well-being score.
The check-ins also asked participants to pick their main activity in the last 30 minutes from a checklist (such as working, eating, exercising, or social media) and to rate how effortful, meaningful, and enjoyable it was on a scale from 0 to 7.
The Intervention
After 3 days of check-ins, all participants completed a midpoint survey. The intervention group watched two short videos on responding with wise empathy on Instagram, then answered comprehension questions. Participants who answered incorrectly were asked to watch the videos again. The control group watched a video about the history and current state of social media and answered questions about it.
For the last 3 days, both groups received three reflection surveys a day, at 9:45 a.m., 2:15 p.m., and 4:30 p.m., alongside their check-ins. The intervention group practiced wise empathy on an example post or on a post they had recently seen. The control group reflected on which other platforms they used and why.
Participants were asked to apply what they had learned during their normal Instagram use, not at set times.
What the Study Found
Participants had used Instagram in the 30 minutes before 38% of check-ins, for an average of about 14 minutes.
- Empathic responses on Instagram changed: After the training, the intervention group reported more appreciative joy, positive emotion sharing, and compassion on Instagram, and less negative emotion sharing, personal distress, and upward social comparison. These changes appeared on Instagram but not offline.
- Well-being improved in moments following longer Instagram use: The intervention had no clear overall effect on well-being across all check-ins. However, among check-ins following approximately 12 minutes or more of Instagram use, participants in the intervention group reported higher well-being than those in the control group. The difference increased with longer periods of Instagram use.
- Instagram use remained unchanged: The training did not change how much or how participants used Instagram. It changed how they responded to what they saw.
The authors note that the training was delivered through the Avicenna app rather than inside Instagram, and the post-intervention period was only 3 days. Whether these effects persist beyond that period, or work when built into a social media platform, is still open.
An Additional Research Question: Effort and Meaning
The same activity ratings collected during the check-ins also contributed to a separate five-study paper on effort and meaning in everyday life. In one of its studies, the researchers used data from the Wise Empathy study to examine whether putting more effort into everyday activities was associated with experiencing them as more meaningful.
Across activities, participants rated moments as more meaningful when they put in more effort than they typically did. People who generally put in more effort than others also tended to rate their activities as more meaningful.
The relationship with enjoyment depended on the type of activity. Effortful non-leisure activities, mostly work and housework, felt less enjoyable. Effortful leisure felt more meaningful without reducing enjoyment. Taking a trip, going for a walk, socializing, and cooking were among the most meaningful leisure activities, while watching TV and online videos were low in both effort and meaning.
What the Design Made Possible
The well-being effect only appeared because each check-in recorded whether, and for how long, the person had recently used Instagram. If the researchers had averaged well-being across all check-ins without accounting for recent Instagram use, the effect could have been diluted by the 62% of check-ins that did not follow Instagram use.
Two design choices made this possible:
- Ask about the targeted context first. Opening each check-in with the Instagram questions let the researchers compare moments after Instagram use with moments without it, and test the effect where the training was meant to work.
- Branch on the answers. Tailoring the follow-up questions to what each person had just done kept the empathy items relevant, while filler questions kept every check-in about the same length.
Adding a short activity checklist to the same check-ins also gave the team data for a separate research question, without a second study.
Where Avicenna Fits
Avicenna supported the study’s repeated data collection through five daily check-ins, a midpoint survey, and reflection surveys.
Using branching logic, the check-ins adapted empathy questions to participants’ recent experiences, distinguishing between Instagram use and offline exposure to others’ emotions. The researchers also included activity ratings in the same check-ins, allowing them to investigate a separate question about effort and meaning without additional data collection.
For similar studies, researchers could complement self-reported social media use with passively collected App Usage data. Timestamped app activity records could help validate reported usage patterns and examine their relationship with participants’ momentary experiences and well-being.
Research Team
Gregory John Depow
First author (Wise Empathy) and co-author (Effortful Leisure)
Department of Psychology, Lakehead University
Aidan Vern Campbell
First author (Effortful Leisure)
Department of Psychology, University of Toronto
Michael Inzlicht
Senior author
Department of Psychology and Rotman School of Management, University of Toronto
Publications
- Wise Empathy Improves Well-Being on Instagram. Depow, G. J., Oveis, C., Zohar, E., & Inzlicht, M. (2026). Social Media + Society.
- Effortful leisure is a source of meaning in everyday life. Campbell, A. V., Depow, G. J., Agarwal, S., & Inzlicht, M. (2025). Communications Psychology.
Planning a Similar Study?
Avicenna supports experience sampling at random times within fixed windows, surveys that branch on earlier answers, and different survey content for each study arm. Explore the documentation to learn how to set up these components:
- Time Triggering Logic
- Flow Control
- Randomize Participants for Multi-Arm Studies
- Criteria
- Participation Period
- Raw Data
For questions about study design, data collection, or the right setup for your research, explore Avicenna’s features and integrations or talk to our team.
