On page 4 of this article there are some good examples of counteracting extraneous variables and really understanding biases. I think articles that summarize experiments like this are good for students to understand the importance of all the little details that could potentially confound their results on the IA.