I went looking out of curiosity, and the first study my search turned up found:

We have established and replicated a clear experimental effect in which subjects randomly treated with political information shift their preferences toward the Democratic Party. We draw the conclusion that, in contrast to previous observational work, the current deficit of political knowledge in the electorate actually benefits the Republican Party.

Fowler, A., & Margolis, M. (2014). The political consequences of uninformed voters. Electoral Studies, Volume 34, 2014, 100-110. [https://doi.org/10.1016/j.electstud.2013.09.009]

I’m not trying to refute your statement or the assertion that you saw the study you describe; I just found this bit interesting and worth adding here.