Consider choosing constraint

In my world, more is more. And yet, every productivity manual ever written recommends focusing on just one thing at a time. It’s been deeply confronting to consider constraint. My husband, an artist, uses things to construct visual landscapes (see image below). For my 14-year-old, visual and sensory things communicate who she is in the world –clothes, hair & makeup feature a lot right now. So, our home is chockablock full of things. My thing is curiosity, and what better place to indulge that than in the life sciences? As a student, all the secret hobby projects were only ever shared with my supervisor if they worked. As a postdoc, there were always side angles to explore and ‘just-for-fun’ collaborative rabbit holes to go down. So, for me, the things are in the mind and manifest in unfinished projects.

Sometimes, more really is more: Abagail by James Morrison (2018, oil on canvas). Image curtesy of Darren Knight Gallery

As a lab head, a more-is-more approach is manageable for the first few years. But there’s a tipping point. Thoughts like ‘concurrent projects increase the chance of finding the one that will take off’ switch to ‘oh my god, nothing is ever finished’. In my role as a supervisor, thoughts like ‘figuring things out is the very definition of a PhD’ turn to ‘my approach to supervision doesn’t seem to suit everyone’ and recently changed to ‘I need to teach my students the explicit skill of creating results and reporting them’. With hindsight, I’m a bit surprised that it took me so long to figure this out...

The week I wrote this blog started with ‘bean-dad’ being lambasted on Twitter for creating a ‘teachable moment’ for his daughter (1) and ended with a coup-attempt on the US government. Going into the latter is way beyond me, but the first got me thinking about how we train PhDs. I loved my time as a student, loved figuring stuff out, and my supervisor then is still a mentor now. But I don’t remember having been given any explicit instruction for writing up my work at all.

Australian PhDs have changed a lot since I was a student, but my impression is that students still mostly figure stuff out for themselves. It clearly works better for some than others, but with overall shorter PhD candidature and higher expectations with regards to how much data make a paper, I think we're at breaking point. Up until very recently, I’d tell every student in my lab, ‘find a paper that you like and deconstruct its elements so you can figure out how to do it yourself’. But no more, it turns out that the traditional ‘they’ll learn it best if they work it out for themselves’ approach (called constructivism) is too slow and returns patchy results. The alternative is explicit-teaching, breaking down and teaching the specific skills so students can spend more of their time on higher-order thinking (see notes for a nice explanation and further reading).

What’s the connection between explicit teaching and constraint then? This year I’m changing the way I supervise students by introducing the weekly one-pager. The one-pager template summarises the one main thing the student does toward their Honours/PhD thesis each week. I have a detailed template for my lab. It teaches students to constrain their thinking and to build a practice of professional results reporting. The main elements are these:

Descriptive title: The gist of what information is to come (~7-10 words)

Context: Clearly articulate the question you’re asking/ problem you’re solving (3-5 short sentences). Method: a hyperlink/reference to method/treatment etc… details of anything non-standard. Figure: Use a style guide to build all schematics/figures/ (I use a detailed description that is specific to my lab, eg fonts/scale/etc…) Figure legend: The title is the main take home-message (eg Mfg1 expression is linked to metastatic potential, NOT a fact, eg western blot of Mfg1). Then go into the details, remembering that some people only ever 'read' the figures in papers. Conclusion: What has been learned? What new question has this data provoked? Write down what you think this data means in the context of the question you were initially asking (3-5 short sentences). The one-sentence wrap-up gets paraphrased for the opening sentence (of a full paragraph) in the discussion section of your paper or thesis.

Ideally, these one-pagers will contribute all the figures to your final thesis (~12 multi-panel figures per chapter in a PhD, ~12 data figures for an Honours thesis, ~6 data figures for research in action students) I strongly recommend everyone take this free writing course:

The intent for the one-pager is that students learn from day-one how to structure their work and their thinking around one main idea, and to communicate it. Their working week finishes by sending me the one-pager. The next week starts with our Monday one-on-one where we constrain our discussion to what was in the one-pager and where to go next. Whereas in the past I would have had a free-ranging converation about the possibilities for a student's project, the one-pager forces me to constrain to the point, focusing on the next-best-step for them. The benefit to the student is that they get better practice in reporting their results, have less anxiety about the open-ended nature of candidature because they have tangible eveidence for progress, and they get better/more specific help from me.

Most businesses work by a money-in-exchange-for-value model. How value is measured in STEM is a subject for a separate blog, but I’ve come to believe that the universal skill of PhDs is the ability to produce valuable results. Once students learn how to harness this skill by effectively communicating their results, they will be valuable wherever they go, in or out of academia. The benefit to me of teaching students how to do this early in their careers is that I hope to see a better return on my investment (ie more complete projects) so it’s a win-win. Whether you're a student or a lab-head, I challenge you to consider constraint and to use the one-pager as a framework to manage the student-supervisor relationship.


(1 )

216 views0 comments