Mentorship Philosophy Statement
That document is a living document, liable to change, and so the following should be regarded as guidelines, dynamic ones at that, that reflect continuous reflection, learning, action, and improvement over time.
Everyone comes to the group from a different background, with different experiences, and I want to be a good mentor to everyone, which means adapting to help them individually, and being as supportive as possible. I would hate to think that something I say below, might limit someone’s imagination or make someone unnecessarily inconvenienced. We are all in this together: if you have a problem, then I have a problem too, and I want to help understand it and, if possible, address it. And if you have a question, or there is uncertainty, please come to me as well. Please reach out to me if you want to discuss anything – I want to be as supportive as possible.
Individual meetings (typically 30 minutes, but can be longer if needed) are, by default, once per 3-week cycle. At individual meetings, we will go over ideas, plans, raw data, analyzed data, interpretations of data, and writing and questions since the last individual meeting, as well as future plans and action items for the next few weeks. We’ll evaluate plans and data together, modifying the plans if needed, and also discuss any ways I can be more helpful or supportive, e.g. bringing in help from others, or new resources. Many people in the group choose to bring slide decks (often hosted on a Dropbox shared with me, so we can see the trajectory of past meetings) to these meetings to organize their data and plans; some bring data and plans, or agendas, in text format. I think this is a very good practice. If someone wants to meet more often than once every 3 weeks (common when one is new, transitioning projects, working on fellowship applications, etc. – or just if needed), this is fine too: after finishing our first meeting in a 3-week cycle, we can try to schedule a follow-up meeting earlier than it would occur on the traditional 3-week calendar. Of course, you do not need to wait for a meeting to email me data, plans, questions, requests for feedback, or other information. Email is often a way to get more rapid feedback.
Once a year, we will have a (voluntary!) individual development plan meeting. We’ll go over skills to acquire, networking and presentation opportunities, writing and reviewing opportunities, teaching and education activities, fellowship and grant opportunities, collaboration and leadership opportunities, and in general any opportunities or activities (e.g., internships at startups or companies) that may be helpful for an individual’s career goals. Non-lab activities can be extremely valuable for boosting individuals’ intellectual and professional development. Of course, all of this can be dynamic, but it can be good to make a plan, even if we are free to change it!
Of course, one can always request a meeting, whether for an individual or for a group of people. Just reach out via email.
Email is the best way to reach me; I try to get through my whole inbox every 24 hours. If a subject line indicates urgency (e.g., an order approval), I try to prioritize those emails first. If more urgent, I can me instant-messaged on Google Chat/Hangout. In a very urgent situation, I can be texted at my cell phone number. (Please reserve my cell phone for true emergencies, since if cell phone texts become too common, then this communication channel just becomes another ordinary channel.) But of course if something is a true emergency, I want to know about it and help! I tend to respond to messages fastest during daylight hours on weekdays; I sometimes check messages at other times (e.g., on the weekend, especially Sunday afternoon) but not as often. I tend not to write emails at night, or on weekends. (In the event that I do send you an email on the off-hours, perhaps since I’m just up working: I do not expect replies in these off-hours; feel free to wait until your next working hours to reply.) If something has an urgent deadline, of course, we can make arrangements in advance to correspond in the off-hours. And if someone must meet at night or on the weekends, e.g. someone has clinical duties during daylight hours during the week, or we are going for an urgent deadline, we can often work this out, if we plan in advance.
If someone in the group has a medical or health issue, or other life issue that may change how you do work, feel free to let me know how I can be supportive. You are not obligated to disclose any details, unless you want to, of course. MIT also has many mental health, student and staff support services, and other resources (such as the MIT ombudsperson) if you want to seek help from others at MIT. I also encourage people to talk to senior administrators and other faculty in their home departments, and group alumni, for perspective and advice.
I always want to hear feedback from people in the group. The group, and I myself, have evolved a lot since the beginning of the group. I’m always trying to be a better mentor, and to become better at all aspects of my job. I’ll be the first to admit that I’m always learning.
I like to make plans, whether for a project or for an academic path in the group and beyond. We should be free to reject the plan for a better one, but it’s nice to know that the worst case is still potentially amazing. I also think that the big ideas often require serendipity, working together, and following chance leads (see our group’s essays on “Assumption-Proof Brain Mapping” and “Architecting Discovery” for examples of approaches we’ve used in the past). This is a delicate art, of course, because some paths might be dead ends. I like to work, in 1-on-1 and subgroup meetings, on how to maximize impact and creativity while minimizing risk and time – something that I think can be a “learnable, teachable” skill. Thus, projects usually arise out of series of discussions between group members and myself, rather than from an overly prescribed algorithm. Of course, I can always suggest projects to try out – we usually have a very long list of cool projects we could explore. And I love to hear about new ideas and results – many of the new directions our group has headed in, arose from the initiative of a group member, or out of a group discussion. One issue, of course, is that new directions sometimes require new funding, so we may need to work together to secure funding for a new direction, but this is a good skill to have in any case. A great project will be both of interest to me, and to the people in the group who will work on it – without that dual motivation, a project will not likely succeed. Finding such a great project can require iteration, going back-and-forth, and revisiting our path and our assumptions, across many meetings. Of course, funders will often want deliverables at specific times, e.g. quarterly and yearly, and we all will need to do at least some work that keeps funders happy, in order to sustain the group.
Graduate students, postdocs, and research scientists in the group are from different programs or in different departments, which makes it hard to come up with a one-size-fits all description of academic progression. A yearly meeting to look back over the previous year, and to look ahead to the next year, to review data, inventions, ideas, plans, and so forth, should be held. Graduate students should have committee meetings yearly, to achieve this goal; postdocs and research scientists will have a yearly review process to go over such things. This is a different meeting than the “individual development plan” meeting described above, just to be clear. My hope is that such meetings are room to explore where to go, and what to do, and are not seen as “grading” – rather, they are used to take a step back, to look at the big picture of the project and what we are doing, and think constructively about what we have done, and what we can do going forward.
What makes for a successful academic time in the lab? At the substance level – it should result in a new discovery, invention, process, idea, or approach, ideally one that gets to something fundamental, or that has broad-ranging application, which yields impact. The topic is, in our group, usually about understanding or repairing the brain, although forays into other topics of course have occurred over the years, and continue to occur. For the latter, we should discuss this carefully, since there is danger in a group becoming spread too thin, or in people becoming isolated from others. One possibility is co-advising – someone working on, say, a tool in our group and applying it to a problem in another group. I’m open to anything that both has great positive impact on the world, and that furthers an individual’s career progression. I’m pretty flexible about trajectories in the lab, and if someone tries something out that is risky, and then after a year of trying that out, wants to return to a project closer to the core of the group, that is very possible. Indeed, we almost always have tons of work available on projects close to the core of the group – which, in some ways, may provide some “insurance” that enables people to try risky things, because we have backup projects ready to go at any time. Some very cool projects have resulted from major project pivots in a later year in the lab. Many people, at some point, converge upon 1 main project, and perhaps a few side projects where they collaborate with or help others in the group. I think that “constructive failures”, e.g. trying things out that don’t succeed, but that point you in the right direction, perhaps because you’ve seen something that nobody’s seen before, and now you know what to do next, should be celebrated as successes.
Often, side collaborations arise, for example if a tool someone is working on could have great impact by being applied to someone else’s problem. It’s important not to be overwhelmed by side projects, since finishing main projects is important, but they help science move forward and help build a good network for one’s career. We also collaborate strategically with many other groups. As examples: we work with another group, side-by-side, on a tool, and the collaboration helps us make the tool better than we could do alone; we work with another group, providing a tool, but going beyond the standard teaching-and-support expected after a paper comes out, and end up collaborating; we generate new science, and need the tools or computational help of someone else, to facilitate or interpret our science. But many other combinations are possible. The main point, of course, is that the collaboration should be greatly positively impactful for the world, and boost people’s personal careers or paths.
At the procedural level: typically graduate students will have one or more first-author or co-first author papers well along the path to being published (written, submitted, and/or fully published; the details are often worked out in discussion with the thesis committee) by the time of the PhD defense. (Some programs, like MAS and EECS, require a Master’s thesis on the way to the PhD; because bioengineering and neuroscience work takes so long, I usually regard the Master’s thesis as a “checkpoint” or “progress report” along the way to the ultimate goal of the PhD, where we report on what we’ve done to date (quite possibly, things that didn’t work).) I personally do not believe in keeping graduate students longer than they think they should stay, but I do think it’s good to talk about how to maximize the positive impact of the PhD experience on the student and on the work – e.g., how to prepare a student for their future career, how to keep the research going (for example by training new group members on key skills), and so forth, as parameters to be considered. PhD theses in our group have no standard organization; they are sometimes just compilations of peer-reviewed papers and/or paper drafts, one per chapter, flanked by an introductory chapter and concluding chapter. Typically postdoctoral scholars will have one or more first-author or co-first author papers well along the path to being published (written, submitted, and/or fully published) at the time of applying for a faculty job, if that is the goal. Sometimes people in the group will do a small faculty search around the time of a first paper submission, with the idea of doing a broader search a year later when papers have ideally come out in print. Research scientists in our group may be either “permanent” staff members, or transitioning to other target careers, so trajectories are more individual. This paragraph is necessarily a bit vague, alas, because in the end we are all on individual tracks, and thus academic progression should be viewed through the lens of the individual development plan, mentioned above, as well.
As long as people are doing well, I don’t regulate hours in the lab or vacation time. People are also free to work when they feel most appropriate (although collaborative work may require coordination and consensus-building). I think that people should balance work and life, and take care of their needs. In addition, everyone is different, and has different personal constraints and needs, and are creative and productive in different ways. Science is hard, and we work very hard, but it’s important that our motivation for work match the work that we must do. That doesn’t necessarily mean the answer is to work more; perhaps if the motivation is low, we should ask why – for example, maybe the work isn’t important enough to justify the amount of work required, or we should switch directions. Or maybe the work is too hard because we lack a key tool, and we should switch to working on that instead. Ideally in science, our work would synergize with our deep curiosities, goals, and dreams. Sometimes we will have to work more, and sometimes we will have to work less. Personal lives are important, and not spending enough time on personal lives can be a source of stress. Time management is also a skill. As with science mentorship, I like to focus on “learnable teachable skills” – happy to work with people on this skill.
In closing, these guidelines are just that – guidelines. Our group works on so many different things, and we have such a diverse group, that I don’t know if there is, nor should be, a “one size fits all” recipe of how to work and succeed in the lab. I may have a more flexible view on success than most faculty: indeed, some group members who dropped out of their degree program and then started a company or pursued other opportunities are generating great impact and changing the world for the better, and I consider that a success. In summary, my hope is that we can all, working together, help everyone in the group maximize their positive impact on the world, and to help them achieve their personal career and life goals.
This document began as a summary of our earlier 39-page lab guidelines, and thus incorporates many ideas that were added to that document by many earlier group members. It probably incorporates a zillion documents that all fed into that document, so no claims of originality or authorship are made here. It also incorporates ideas prompted by MAS and BCS department members, and many other MIT staff, along with references that they sent along, and many expectation documents that different groups have circulated. Thus there is no author of this document, so to speak; when I use the word “I”, it’s simply a practical matter, to have a point of contact.