Table 4.

Key findings of all cycles.

Cycles	Key communication goals achieved	Key communication goals not achieved	Summary of psychophysiological data (where applicable)	Summary of how issues were addressed in the next cycle
First cycle	• Nearly all participants reported that the color yellow represents vulnerable people. • Most participants reported that the blue band around vulnerable people meant protection. • All participants reported that the avatars in the cluster of hexagons represent members of the community.	• Most participants did not understand that the central avatar represents them. • Some participants did not understand that the avatars around them could include nonfamily contacts, for example, coworkers. • All participants understood the purpose of visualization as promoting immunization rather than explaining the concept of community immunity.	• Most participants visually attended to the appearance of vulnerable people. • Most participants had peaks in arousal and showed high engagement when avatars got infected. • All participants showed high engagement when red lines showed the spread of infection. • Most participants visually attended, and all participants had peaks in arousal and showed high engagement when vulnerable people got infected. • Most participants visually attended and had peaks in arousal when community immunity was explained. • All participants showed high engagement when blue lines around people spreading out from vaccinated people showed community immunity.	• We presented the center avatar, immediate family members, colleagues, and communities in the same visual frame by zooming in and out. • We removed the term immunization in the narration script and focused more on community immunity. • We decided to test the terms herd immunity, community protection, and community immunity by asking participants which term they relate to and prefer. • We added a question to be asked in the next cycle about the shape of hexagons presented in the visualization. • We added a new design element in the next cycle explaining how different viruses (eg, measles, pertussis, influenza) spread at different rates and require different vaccine schedules.
Second cycle	• All participants reported that yellow signaled vulnerability. • All participants reported that red signaled infection. • All participants reported that hexagons represent people. • Nearly all participants reported community immunity safeguards vulnerable people when sufficient people around them are vaccinated. • All participants preferred the term community immunity over herd immunity or community protection.	• Few participants reported that blue meant vaccinated or immune. • No participants reported that gray meant susceptible. • Most participants did not report that different diseases spread at different rates. • Most participants did not report the role of vaccine effectiveness in community immunity. • Few participants reported vaccine-induced immunity. • None of the participants reported accurate understanding of natural immunity. • All participants suggested that the visualization should be shorter.	• N/Aa	• We removed images of viruses but retained in the narration explanation of how different infections spread at different rates. • We added images about different levels of vaccine coverage to achieve community immunity for different diseases. • We removed images about different vaccine doses and natural and vaccine-induced immunity. • We shortened the visualization to about 2 min and used the term community immunity.
Third cycle	• Most participants reported that the older woman and baby avatars represent vulnerable people or those with fragile immune systems. • Most participants reported that vaccines prevent the spread of infection. • Most participants reported that community immunity safeguards everyone. •Some participants reported that the thick blue band around an older woman and the baby shows community immunity protecting vulnerable populations. •Some participants reported that their decision to get vaccinated or not has an impact on other people in their community.	• Nearly all participants found it difficult to identify with the generic avatars.	• Some participants had peaks in arousal when the avatar first appeared. • Most participants had peaks in arousal when the vulnerable population was explained in the visualization. • Some participants had peaks in arousal when the infection first entered the community in the visualization. • Most participants visually attended when the community immunity was shown, along with how it safeguards everyone. • Overall, participants showed high engagement and an optimal workload throughout the visualization.	• We added a functionality for people to build their own avatars and their own communities.
Fourth cycle	• All participants reported an accurate understanding of what community immunity is and how it works. • All participants reported that community immunity safeguards vulnerable people and everyone in the community. • All participants reported that some infections spread faster and need enough people to get vaccinated to prevent the spread of infections. • All participants found it easy to create avatars by following instructions without a tutorial. • All participants liked the palettes for skin and hair colors.	• For all participants, the avatar creation tutorial was confusing. They preferred to make avatars just by reading the step-by-step instructions. • A participant suggested adding additional accessories such as caps, hats, hijab, and other head and hair coverings.	• Nearly all participants visually attended to the avatar creation elements, including written instructions.	• Head and hair covering options (caps, hats, hijab, turban) were added.

^aN/A: not applicable.