Tips for dealing with fickle advisor

Question

Field: Neuroimaging

Location: India

Current position: post doctoral fellow

Background: I am continuing to work in the same group where I did my PhD. My advisor/supervisor/mentor is a nice person who gives me good suggestions and/or throws in the right question at me which has helped shaped my thinking in the past few years.

I have been working on a couple of (almost) independent projects in the recent times. The plan that I follow is to come up with an idea, "design" the experiment (i.e., how the computation should be done), do a few pilot experiments, and get some proof of concept results. At this stage, I take them to my supervisor for feedback and/or help (say when I need additional resources to get the computation done, etc.).

While my supervisor appreciates the work that I do and is usually supportive of my independent projects, his usual tendency is to "jump" between projects. Concretely, say if I am presenting results of project A, he would start asking me about updates on project B, instead of finishing discussion on A. If I actually have an update on B, he would move on to C. Alternatively, he would just stall the discussion on various projects and inadvertently get me to work on one or the other project. However, after a point he goes back to jumping between projects.

As a consequence of this "fickle" behaviour, projects take a lifetime to complete. Even if a project reaches a stage where we start writing the first draft, the time it takes to submit the paper is usually many more months than is strictly necessary. Therefore, I have many ongoing projects but hardly any publication. I am planning to leave this lab next year and apply for postdoctoral fellowship abroad but I worry that I will not have enough to show for my time here (in terms of publications).

Question: Are there concrete tips on dealing with this kind of situation so that projects actually reach somewhere in a valid time frame (rather than be a never ending story)?

Other info:

• Even though some of these projects are primarily my ideas, it will not be possible to publish things without my advisor being on board
• As and when (if rarely!) a project reaches the paper stage, the paper does benefit from my advisor's comments (and he does contribute significantly)
• The reason I go to my advisor for feedback after an initial set of results is because if I wait till the final results and then present them, he might ask me to change something and redo the analyses which may not always be a good idea (for example, consider a machine learning case where I have tested my model on the final holdout set and the results are ready for publication; if the advisor asks me to change anything at this stage, it would introduce bias into my results)

Extra information (Edits):

• The meetings where I present these results are pre-arranged and the agenda is (almost always) to discuss the specific results of a specific project (for example, I will email my advisor to let him know that I have reached somewhere in project A and that I would like to discuss the results and the next steps at a time that works for him)
• With any request for such a meeting, I attach a PowerPoint presentation where all pertinent information (background material, methods, results, references, possible next steps, outstanding questions) are present
• The meetings always happen in person

Answer 1

You're already doing the things I would suggest as 'softer' tips: setting a specific agenda and preparing a presentation on a specific project is a pretty good cue to everyone in the room that it's the focus of discussion. You could be more direct and say "today I've prepared these slides to discuss project A, so I'd rather finish that before moving on to B and C", but I don't know if that'll really work in your case.

I think you should instead have a direct conversation with your advisor. Avoid being accusatory ("you're preventing me from being productive!") and stick to describing the problems you have ("I feel stretched between multiple projects so that I am making only modest progress on each of them rather than getting them completed; I need completed projects to advance in my career"). Seek their advice on how to manage this, and if that advice includes focusing on one of your projects, you can refer back to that advice when you are quizzed on something else. You have the benefit of working in a field where the standards for authorship and productivity mean that your productivity is your advisor's productivity, too: your coauthored papers with them count just as much for them as they do for you, so your goals of getting papers published should be in alignment.

I'd also consider, or perhaps ask, whether there are other outside factors behind your advisor's shifting goals. I'm in a bit of a similar situation, working as a scientist in a lab that is juggling a lot of different things. Some of those are collaborations where we've promised progress to various other people. Given those commitments, it's more acceptable to make progress on each of them rather than to tell certain collaborators that we won't get back to them until we finish XYZ. The reality is that XYZ project will probably never actually be finished, because even if papers are published from that project these will spawn new avenues to research. Until everything is known, nothing in science is really ever complete. There can also be grant deadlines or progress reports due for particular projects. I think if you know better what your advisor's motivations are, that may help you anticipate where their focus will be as well as understanding whether they are being fickle or just trying to manage various aspects of a heavy workload.