The other night over dinner, hours after my mother-in-law had returned home to New York, I casually asked my son Leo: “What was your favorite part of the weekend?”

As I watched him stare blankly back at me, struggling to find an answer, I found myself wishing I could have a parental do-over. Why do we ask children this question so often? Would it make a difference if we asked it a different way?

Anyone who’s a parent knows what I’m talking about: we’re always asking kids to tell us their favorite color, pick their favorite TV show, or select their favorite relative. And our intentions are in the right place; after all, we’re trying to learn about how they see themselves and others, and to give them a chance to reflect on what feels good and pleasing.

But here’s the problem: children don’t see the world as a set of isolatable favorites; we make them see it this way.  Watching Leo’s face, I realized that for him, there was no single favorite memory – just a pastiche of happy experiences that blended together to make up a general feeling I’ll call “Weekend with Nana.” It wasn’t until I asked the “favorite” question that it even occurred to him he should decide which of his experiences with her was the best of all.

This distinction is not exclusive to Leo. All of us benefit greatly when we develop metacognition – or the skill to reflect on our own thoughts and feelings, see ourselves interacting with the environment and people around us, and become familiar with our own preferences and the preferences of others. Recent research even suggests this may be the most important skill of all when it comes to learning how to learn, both in school and in life. Yet the reality is that asking kids to pick favorites isn’t an optimal path toward helping them become more holistically self-aware; it’s an emotional short cut that teaches them to artificially divide their memories into preferred parts.

How might Leo have responded differently if I had asked this question instead: “What made you feel happy this weekend?”

The difference between the two questions is subtle but significant. With one, we’re asking children to rank the world. With the other, we’re inviting them to reflecton it.

Only one of those questions will actually help build the muscle memory of metacognition, and allow for a fuller understanding of the multiplicity of experiences that shape how we think and feel. And that’s not playing favorites.