Notes from my graduate studies at the University of Toronto in the Department of Computer Science.

Groups and Teams

Saturday, August 25, 2012

Over the last year I've been working part time in the Translational Imaging-Genetics Research lab at the Centre for Addiction and Mental Health.  It's been a significant change of pace for me, and an entirely different domain of research.  I didn't think I would be qualified to work in such a place, but it turns out that with a very understanding supervisor and my aptitude for all things computery, I've found my way into the niche of IT-guy/command-line-wizard/software-developer/research-assistant.  I'd like to take some time, eventually, to describe more about my work in the lab, but in this post I'd want to take you through some of my experiences and observations of the lab culture, because it is significantly different from other places I've worked in academia and I'm enjoying it.

The title of this post is the difference in a nutshell.  In the past, and in particular in during my time in the CS department, I have worked in a group of other researchers.  We each had our own research projects which we worked on mostly independently, and we were assembled in a physical space because of the expansive notion of "software engineering research" we all claimed to be doing.  There was not much in the way of collaboration with other students (that I was aware of at least) in the true spirit of the word though.  That is, it wasn't often that we worked on projects together: where we sat at the same desk and brainstormed or wrote code, co-wrote papers, or chatted about what we were up to specifically that day and asked for a hand with it.  We did regularly have meetings, and sometimes presentations, to update one another on our projects, and we might (sometimes) talk about research over lunch in the lunch room, so we weren't completely isolated from one another but most days passed rather quietly, working alone in our cubicles.  Of course, sharing a physical space meant that I made some friends with the other students, and would occasionally have discussions about research (shout outs to Jorge, Neil and Jono, with whom I schemed several side projects, and Alecia F who was often there to bounce ideas off of/proof read/pester me).  In summary, I felt I had the support of others if I needed it and could ask for it, but overall support wasn't something that was often forthcoming, and the sense I retain from my time there is of working amongst others and not so much with others.

In the TIGR lab the culture is rather different.  We are more like a team, than a group. Although the students in the lab each have their individual research projects, there is lots of overlap between them.  Maybe it's that one person is investigating a different aspect of the same dataset as someone else, or that they have experience with a particular kind of analysis technique that another person needs to use, or  one person may be more familiar with a statistical method or handy with a bit of software that someone else needs.  Several times a day someone will ask for help when they're stuck trying to make heads or tails of their results, or people will spend time discussing (or arguing about) an approach or piece of writing, for instance.  More concretely, for myself, I'm often asked if I can help fix a script or write a bit of bash code to automate a task someone has to do for their work (the neuroimaging world, I've discovered, is rife with pesky file conversions and glue code).  I'm also always asking people to help me use the neuroimaging software (since it's all new to me) or for someone to explain to me how brains/anatomy/MRIs/image-processing/etc.. works.  Additionally, various people in the lab have taken on roles in doing weekly data processing tasks for long-running studies -- often which aren't directly related to their primary research but the study is important to the lab as a whole and the work is just too much to be done by one person.  As well, the PIs in the lab will sometimes gently delegate work to get pilot data ready for a grant proposal, or ask for feedback on the proposals themselves.

The point I'm trying to illustrate is that the lab environment is extremely collaborative, and for the most part this collaboration is seen as a necessary part of the functioning of the lab.  I don't think we could do the work that we do without having the expertise and help of everyone.  As you might suspect, the lab at CAMH is also a much more talkative and social place.  It seems to me that part of this is just the nature of team work -- you need to communicate to do it.  But, there's also a genuine friendliness and camaraderie that has arisen, maybe because of our need for teamwork, and it certainly also supports it.

This distinction I'm making between a group and a team is something I've probably had a sense of in the past but my sudden shift from one environment to another has brought it into focus for me.  You can probably tell just from my biased language that I prefer working in a team.  For research at least, I really seem to thrive when I'm working with and for others (this, despite my introvert tendencies) with a shared purpose.  I'd bet that a more isolated, group environment works well for those people that have a clear idea of what they want to do and are happy to just put their head down and do it.  That, I'm finding, is just not me...  at least not when it comes to research (it can describe me pretty well when I'm on a solo bike trip, or doing field work whilst farming).

I was recently asked to speculate on what makes one laboratory more team-like and one more group-like.  This is probably best left to someone else who has more experience (... or who has done the research... *cough* *cough* Dr. Jorge Aranda *cough*).  But, with my two and half data points in hand (i.e. the different labs I've worked in for any significant amount of time), I'll jot down some of my conjectures as to factors influencing teaminess in the TIGR lab:
  • an overlapping set of research goals and and divergent skillsets amongst the lab members seems to make a lab more apt to be team-y, as I've described above (and see interdisciplinary research below). 
  • a shared, open space.  The TIGR lab is one windowless, smallish room (say, a little over three Jon's wide by three Jon's long) in the basement of CAMH.  There is essentially one desk that wraps around the room, with no divisions, and all six or seven of us students sit side by each.  The two PIs each have adjoining offices that open into the room.  It's a tight space, but it makes spinning around in your chair to ask someone else for help possible, and the norm.  I also like having the PIs share the working space since it adds a we're-all-in-this-together element.  (I'm sure the PIs might sometimes prefer to have a separate space to escape to though!).  In contrast, my lab space in the CS department consisted of semi-cubicles that divided up the room with the professors sitting in offices in a different wing of the lab. 
  • conversations happen, and some of it isn't always about research (like in real life, sometimes youtube parties happen). It can get noisy at times in the lab, but overall people are extremely respectful of the working environment, and also tolerant of discussions going on around them.  This is, of course, a necessity in a lab as tight as ours is.   But the advantage of such freely flowing conversations is it encourages (sometimes serendipitous) opportunities for collaboration.  
  • shared breaks and socialising.  Very often when someone leaves the lab for a break or to get food they ask if anyone else wants to join them.  This includes the PIs.  Even if we just go walk outside to stretch our legs and see the sun shine or get a coffee.  And informal group lunches happen often, especially if someone is excited about a new restaurant, say.  Obviously some of this behaviour happens because we all have bonded and want to hang out, but it's also just a respectful thing to do as a way to include everyone and maybe make sure people are taking care of themselves (read: those-of-us-in-workaholic-mode-who-seem-strapped-to-our-seats).  Of course, we have more organised group outings if someone is leaving the lab, or after conferences, or just at milestones during the year.  
  • personalities matter. This is something interesting: the PIs have made a point of asking those of us in the lab to interview prospective newcomers and give our impressions.  Not so much about a newcomer's qualifications but more about their fit with the lab -- i.e. would they make good lab mates?  It's a rather nice thing to do.  To me it means they recognise the stake we all have in the decision, and that group dynamics in the lab matter because (among other things) they affect the kind of research we all do.
  • the expectation of working together. As I mentioned, it's been an expectation from the PIs that we share in certain lab duties. In group meetings one person may be directed, or volunteer, to help another person.  I don't want to overstate this though; it isn't as if we do all or even most of our work together, or that we're all spending lots of time working on each other's projects. No.  It's just that there is a certain willingness to help out when necessary that isn't really questioned or even maybe obvious, but I notice it as difference and I think it helps build a sense of the team. 
  • shared hardship? I wonder if some of what fosters a team is sharing in the struggle against hardship.  Our lab is newly established and growing.  And that means we regularly experience the growing pains of new hardware and software, for instance.  And as I say, we work in a windowless and somewhat cramped workspace.  It's not ideal in some sense, but maybe this kind of resource scarcity fosters group cohesion in order for survival?  We really do have to tolerate noise and conversation, for instance -- there isn't any choice!  And maybe looking out for one another by going on breaks or getting food is also a way of keeping things friendly and peaceful. 
  • different authorship rules? Here's another really speculative one.  In computer science it seems that the threshold for authorship is set much higher in terms of contribution to a project.  In the the slice of the neuroscience world I'm involved in now, authorship is often given to people who contribute in ways that may only get a thank-you note in a CS paper.  For instance, early on at the TIGR lab I learned how to operate an analysis pipeline which I used to processes a corpus of MRI images our lab had collected.  I put the results of that into our database, and later on someone else used them in their work and I was named on their poster as an author.  Now, I don't think I even deserved a mention because what I didn't feel very significant... but, instead of a thank-you (something I haven't ever seen in neuroimaging papers, by the way)... authorship.  So maybe, subtly, this kind of generous recognition of contribution leads to more collaboration?  I don't know, since I haven't worked in any other neuroscience labs (remember, ~2 data points and all). 
  • interdisciplinary research.  The work I'm doing in the TIGR lab seems just to be inherently more interdisciplinary.  It's even in the title of our lab: translational imaging-genetics (i.e. we investigate   the relationship between genetics, brain structure and function (via imaging), and behaviours and abilities -- mental illness).   Roughly speaking, our lab has people from biomedical engineering, psychiatry, genetics, psychology, neuroscience, physics, chemical engineering, clinical psychology, and computer science.  The problems we're working on are just too big to be tackled alone. 
  • being wrong is okay? so this point was suggested by one of my lab mates who read an early draft of this post.  The idea here is that maybe in a team setting we're sharing more of the day-to-day slog of doing science: we see each other try things that don't pan out, or screw up doing something that eventually does work (or do both at once!) every day. The mundane successes and failures are shared a little more, and that makes for a kind of comfort and willingness to work together.  


Leash said...

You missed an important factor...that I know to be highly influential with you and your definition of team. Wether one gladly shares hunks of bread and cheese and a bowl of olives with you during lunch breaks.

There should also be a clown on every team...or else I serve no purpose ;)

About Me