
Jeremiah J. answered 04/17/19
Science Communication Advisor specializing in Oral Presentations
My best advice is to think about the different points that all talks hit. This skirts around the need to be familiar with someone's research (and as someone who also went to a lot of talks outside of his field, I can say that these worked for me). Generally, you can ask a question that's still relevant if you fall into one of these three (or four) categories:
1) Background information.
2) The experiment/research.
3) The conclusions they draw.
*4) Questions other people may ask.
So, this is what I recommend. Have something to write on, and as the person giving the presentation moves through these three stages, jot down questions you may have about what they say. Don't worry if they sound silly or trivial -- these are your notes, and you don't have to show them to anyone.
So what are some examples?
1) Background information.
Before I joined my PI's lab, he was giving a talk at the annual departmental retreat -- the kind where no one was allowed to leave before at least one first-year graduate student (i.e., me and the rest of my cohort) asked a question. The specific focus of his talk was how a uni-cellular, aquatic organism swims by way of analyzing the shape of the wave generated by its flagella while beating. There was a lot of physics and math involved which was beyond my education at the time, but one thing I did think to ask was about the premise itself: how an organism swims.
So that's what I asked. "When I swim," I said, "I can let momentum carry me through the water. Is that how these guys swim too?" It turned out that that wasn't a trivial question after all (it hit the nail on the head of a topic called "low Reynolds number swimming"). It was a gamble that led to more discussion.
2) The Experiment/Research
I was attending a talk that my study program specifically had at the start of every month. One lab's grad student was presenting on tracking the growth of actin filaments using a MATLAB program. All of the videos they showed, however, were only of single filaments, and that seemed interesting to me. So, I thought of a question to ask.
"What does your program do if two filaments overlap in the video? Rather, is there a way for it to tell the difference between two filaments?"
The important thing here is to ask these questions in a non-accusatory way. I genuinely wanted to know about something that seemed to be a legitimate design challenge. I got the answer, and the presenter got a chance to shine.
3) The conclusions they draw
In college, I went to a talk where a researcher discovered you could reduce the number of seizures epileptic rats would have by playing a certain frequency close to their brains. I don't remember the exact details, but I do remember the question I asked:
"That frequency you found -- do you know what's going on? It's not something a rat would typically encounter, is it?" (I tried to avoid saying, "It's not biologically relevant? What's going on?")
The speaker thanked me and admitted that they were just as interested as I was. It was another one of those findings that indicates biology can be super weird.
4) Questions Other People May Ask
I don't have any good examples for this one (at least off the top of my head). But a good way to ask questions is to build off of someone else's. In other words, just because the talk is over, there may still be good chances to think of questions.
**
The bottom line is that it does take work to think up good things to ask, but if you're the kind of person who has a good deal of natural curiosity (and it sounds like you are), then trust the instincts you have. You never know when a question that sounds trivial to you is actually important.