You're right that this research isn't particularly pure CS. When I started it was very much focused on social games and whether the lessons from them could be applied to other software. It just so happens that motivation theory is a pretty wide body of literature, so you can spend a lot of thought cycles outside of CS.

You can spin it as a software engineering thing (architecting for retention), an HCI thing (how users respond to different interaction patterns) or even something outside CS altogether. Once you venture towards the user-end of the spectrum, you can put a lot of different hats on CS research. You just have to be careful that you don't end up either trying to end up spending lots of time learning another undergraduate major's worth, or worse, not knowing enough and being laughed at by a different academic group.