I found this Quora discussion very informative.

2. Develop the ability to become an active reader. Don't just passively read material you are given. But pose questions, develop hypotheses and actively test them as you read through the material. I think this is what another poster referred to when he advised that you should develop a "mental model" of whatever concept they are teaching you. Having a mental model will give you the intuition and ability to answer a wider range of questions than would be otherwise possible if you lacked such a mental model.

Where do you get this model? You creatively develop one as you are reading to try to explain the facts as they are presented to you. Sometimes you have to guess the model based on scarce evidence. Sometimes it is handed to you. If your model is a good one it should at least be able to explain what you are reading.

Having a model also tells you what to look for to disprove it -- so you can be hypersensitive for this disconfirming evidence. In fact, while you are reading you should be making predictions (in the form of one or more scenarios of where the narrative could lead) and carefully checking if the narrative is going there. You should also be making predictions and seeking contradictions to these predictions -- so you can quickly find out if your model is wrong.

Sometimes you may have two or more different models that can explain the evidence, so you task will be to quickly formulate questions that can prove one model while disconfirming the others. I suggest focusing on raising questions that could confirm/disprove the mostly likely one while disproving the others (think: differential diagnoses in medicine).

But once you have such a model that (i) explains the evidence and (ii) passes all the disconfirming tests you can throw at it then you have something you can interpolate and extrapolate from to answer far more than was initially explained to you.

Such models also makes retention easier because you only need to remember the model as opposed to the endless array of facts it explains. Of course, your model could be wrong, but that is why you actively test it as you are reading, and adjust it as necessary. Think of this process as the scientific method being applied by you, to try to discover the truth as best you can..

Sometimes you will still be left with contradictions. I often found speaking to the professor after class was an efficient way of resolving them.

The author lists 8 other criteria, but this one had the biggest "light bulb" moment for me.

It was interesting to me because I intuitively would use this technique while listening/taking notes during lectures. But I never actually made a conscious decision to apply this consistently in all of my classes; it would only happen in classes I was interested in.