Experimental Error banner

Read our COVID-19 research and news.

Experi-mentoring: Where to turn when your PI can’t help

I’m not sure whether everyone else knew what to expect when they started graduate school, but I’m going to go ahead and volunteer that I sure as heck didn’t. I had so many questions right from the start: Which journal articles should I read? How do we order supplies? Why does everyone laugh when I ask them how many years it’ll be until graduation?

Naturally, you might assume that your principal investigator (PI) is the one to ask. After all, you’ve chosen this person for their helpfulness, erudition, and nonzero grant funding. And for many years, that’s how everyone assumed the traditional quasi-genealogical academic model worked: The PI imparted wisdom, and the grad student gleaned that wisdom until the PI either determined the grad student was ready to stand on their own—or needed to free up some lab space.

Yoda once said, “Always two there are, no more, no less: a master and an apprentice.” But the reality—in Star Wars, as in real life—is much more complicated. Recall, for example, that in the same film, Kitster said, “This is so wizard.” Wise words.

In academia, the principle of “always two there are” makes it sound like you’ll train at your PI’s elbow, shadowing them as they perform laboratory experiments and sleeping on a pallet of hay in a corner, until your PI can stand back, years later, hands on hips, and watch you secure their scientific legacy. There are several things wrong with this picture.

First, your PI has far more people to train than just you. Your lab may include postdocs, undergrads, research scientists, and lab technicians. Yoda’s impression is therefore more accurately phrased as “Always about four to 20 there are: a master and a bunch of apprentices of assorted rankings, most earning less than a living wage.” Not only will your PI not train you from dawn to dusk, you could go weeks without even interacting with them. Many labs give the impression that you shouldn’t bother your PI unless your question is legitimately important. For example, you wouldn’t interrupt your PI to ask, “How much magnesium should I add?” but it might be acceptable to ask, “How do I extinguish a chemical fire that I started by adding too much magnesium?”

Second, your PI may be in charge of your training, but your PI isn’t in charge of training. Your PI’s job is to write grants, teach classes, oversee the lab work, submit papers, sit on committees, and maintain a deep scowl. Any PI who solely fills their time training students won’t have a lab for long.

Third, it’s a rare PI who physically performs lab work. So, although your PI may be nominally training you, the idea that you’ll watch them in the lab, elbow-deep in soiled glassware, might be a fantasy. More likely your PI will direct you to the nearest resentful postdoc.

Fourth, hay is itchy.

When your PI is unavailable to mentor you—presumably because they’re inside a vault, cackling as they count their stacks of money—it’s helpful to know where else you can turn. Here are a few possible alternative sources of mentorship:

Older Grad Students

Ah, the divine wisdom that comes from having experienced something only a year or two earlier. Older grad students in your lab have the advantage of knowing most closely what you’re going through. But, on the other hand, if they truly had terrific insights, they wouldn’t still be in grad school. Take their advice with a grain of salt, and watch out for pranks; many younger grad students have made fools of themselves when told to go find something that sounds plausible but doesn’t actually exist, like a left-handed smoke shifter—or a university administrator who supports grad student unionization.


If the postdocs in your lab want to become PIs someday, they’ll have to learn how to mentor students; it’s an important skill for them to develop so that it can eventually be ignored and neglected. Help them along that path while they help you learn, but don’t worry if they’re bad at it—most likely it won’t be a lack of mentorship skills that keeps them from becoming a PI; it’ll be the abysmal academic job market.

Research Scientists

For those unfamiliar, a research scientist is a step above a postdoc but a step below a PI. In other words, they may be just as experienced as your PI, but without students to train, grants, university recognition, tenure, a shot at tenure, or an enviable salary for someone so advanced. You might think it’s a mistake to hitch your wagon to this somewhat dim star, but research scientists are often the best-informed members of the lab. Like your PI, they have many years of valuable experience, but unlike your PI, they actually work in the lab.


Asking the lab’s undergraduate intern for help shows that you’re not so proud that you’d value petty hierarchies above knowledge. It also gives them a dangerous taste of power that they do not yet know how to responsibly wield, and you’ll find yourself dodging their “advice” for the rest of the year.


If you have questions about your research, and you’ve either exhausted your possible human mentors or you’re too introverted to ask them in the first place, fear not: Google can be your mentor! Google knows everything! This is a good plan. In fact, according to autocomplete, this plan is a hybrid of hmo and ppo plans in my area urgent care hours.

Other Faculty

Asking underlings or search algorithms for answers is one thing, but seeking advice from other faculty members will make you feel like you’re cheating on your PI. You’ll fear that someday they’ll take your PI aside and whisper, “So, um, when you were out of town, one of your students came and asked me about a lab protocol. I swear, it only happened once, and I had no idea they were your student.” There’s nothing wrong, however, with asking someone else’s PI for help. The worst they can say is “no,” and the best they can say is “here’s $100 million.”

This may all sound like a drawn-out way of saying, “Ask people other than your PI for help.” And that is indeed what I’m saying. The fear of asking for help is a pervasive problem in graduate schools, and reaching out beyond your PI is harder than it sounds.

Don’t just take it from me that experi-mentoring is a good idea. Scientists have studied the mentorship question with academic rigor, and in a study published last month in the Proceedings of the National Academy of Sciences, researchers reported that the skill development of biology Ph.D. students correlated well with active discussions involving senior grad students and postdocs—but not with mentoring from the students’ PIs. Seriously, that’s a real study. Scientists will study anything.

So, if you’re struggling in grad school and unsure where to turn, consider seeking mentorship from sources who aren’t your PI. Build a network of colleagues who can help you, and once you learn a little more, make yourself part of someone else’s network. But most importantly, as a wise boy once said: “This is so wizard.”

Read more Experimental Error stories

Follow Science Careers

Search Jobs

Enter keywords, locations or job types to start searching for your new science career.

Top articles in Careers