Thread by @MCDuncanLab

I want to share a thread about an approach that I’ve been using to mentor people in my lab with the goal of developing “expert learners” in the research environment.

First off, what is an expert learner in research? An expert learner is someone who picks up new skills seemingly effortlessly.

This is in contrast to a novice learner who, when faced with learning a new skill, seems to make every mistake imaginable.

I spent a substantial amount of graduate school as a novice learner, but had excellent help from expert learners (Keith Kozminski, Asa Engvist-Goldstein, @RodalLab), who helped me develop expert learner skills.

A couple of years ago, I found educational literature that described the steps I stumbled through in graduate school. I’ve developed a mentoring approach based on this literature that I hope will help move trainees from novice learners to experts with less pain.

The approach focuses on three stages of learning a new skill. Self-awareness, planning, and evaluating. These terms come from the field of metacognition, which I think is a confusing name that means think about what you are going to do systematically.

Part 1 is Self-awareness: know your own strengths and more importantly weaknesses. For me these are things like, I tend to lose track of time, my handwriting is illegible, I get flustered if I have to work fast, I get bored, and will want to rush incubations.

Part 2 is Planning: plan out how you are going to learn this new skill. There are two levels: big picture and detailed.

The big picture is how will I learn this new skill? Can I shadow someone? Are there several protocols I can compare and contrast? Can I do a dry run?

The detailed level is once you have a protocol. Read through the protocol and identify every place where a step could go wrong, decide how likely that is to go wrong, how impactful that problem would be, and if needed identify mitigation strategies

For me, these are things like, because I tend to lose track of time, I might over permeabilize at the SDS step in an IF protocol. This would ruin the sample. For me, the best mitigation strategy was to switch to triton permeabilization which can be left on for hours.

Part 3 is Evaluating: reflect on the experiment after completion. Did you encounter problems that you did not anticipate? Did your mitigation strategies work? Did you need them, or did they add unnecessary steps? Are there new items to add to your self-awareness list?

For me, the planning stage and self-awareness differentiated novice me from expert me. I was unable to identify what might go wrong, because of lack of experience and of lack of self-awareness about how I personally was most likely to make a misstep.

It took me a ton of actual failures to realize common mistakes I made and identify good mitigation strategies.

Here’s where a mentor and my approach can help. When I have new people in the lab, I talk them through this approach. I tell them a list of common things that might be on a self-awareness list. I tell them about my common missteps and how I mitigate them.

I then ask them to review the protocol independently. I instruct them to think about whether it is like another experiment they have done in the past, and whether that informs on how they should plan for this new experiment.

I then ask them to independently identify things that could go wrong. Then I meet with them to discuss what they think could go wrong. Here, I often point out potential missteps that the trainee did not identify.

After the experiment, we do the same: independent evaluation and discussion. After the first couple of new skills, we skip the pre-step and just focus on the post-analysis. When they are comfortable, we stop the post-analysis.

I’ve been hesitant to post this. I can imagine someone thinking “Shouldn’t people be able to figure this out on their own?” Speaking from personal experience, in hindsight, yes, it should have been obvious, and I did eventually figure it out.

But I wasted a lot of time on experiments that failed unnecessarily. I’m also thinking about all the up-and-coming scientists who have missed out on the hands-on experience where they might have figured it out on their own.

Here are some resources I referred to when developing this. They are all from the topic of metacognition.
emeraldinsight.com/doi/full/10.11…
ncbi.nlm.nih.gov/pmc/articles/P…
link.springer.com/content/pdf/10…

It also draws a bit from "How to solve it" by George Pólya, which is about how to solve mathematical problems. Mainly the part where you try to identify similarities between the current experiment and something you've done in the past.