Thanks to everybody who chimed in! I arrived at the conclusion that (1) there's a lot of interesting stuff about interactions and (2) the figure I was looking for does not exist. So, I made it myself! Here's a simple illustration of how to control for confounding in interactions:>

When do epigenetics matter most for child health? At birth, predicting later development? Or are effects stronger, when measured at the same time as the child health outcome? Our new findings suggest that the answer may depend on how you define relevance. doi.org/10.1186/s13073-025-01451-7