I swear I just had this conversation with my new advisor, complete with 20 minutes of rambling. Is “trial and error” why a PhD takes 5+ years, or is it because of excessive advisor rambling?
RM, let’s frame it this way: it’s very common to spend most of grad school figuring out which experiments you should do, and bring trained (or training yourself) how to do them well. Then the rest of it is spent actually doing them. My PhD took 6+ years, but if I’d known from the get-go what I needed to do and how to do it, I probably could have wrapped the whole thing up in 2.5-3.
Wow. This is becoming unbearably depressing. It seems as if the one comment left is: “I gonna kill myself”. C’mon guys,!. There must be unexplored solutions/alternatives!
I don’t understand – isn’t the thrill of being a scientist actually doing experiments, aka trial-and-error? Why the hell would you want to be a scientist if you just follow directions? And I’ve found its amazing how frequently some random observation from a “trial-and-error” period in the lab comes back to be eminently useful months or years later.
Vertically ascending scientists do not do “trial and error”, they reveal “Truth” that was obvious to them from the onset and demonstrate that truth with one perfect unassailable experiment, whereas the riffraff fumble around with trying different things and wasting everyone’s time and money. That is what makes great scientists different from the unwashed masses and why only they should funded.
Hey, I’ve got to get my jabs in at the established folks when I can, especially when DM’s previous post tells me that I’m “cannon fodder”. The cannon fodder would prefer to remain blissfully ignorant for now, thanks.
I’ve gotten advice to hide the “trial and error” parts of experiments from papers and talks. Meaning, basically, present it like a non-facetious version of Philapodias post. Observation, hypothesis, answer.
I get we often have to do this for coherance and clarity, but I wonder if it influences the attitudes of beginning trainees and causes them to mistakenly think of science as a straight line instead of meandering concentric circles.
@jmz4
This is why all graduate students should read classic papers like Katz and Hodgkin (in my world, surely there are other field-specific versions) when the style of writing allowed for thought processes to be written out. False starts and blind alleys were not omitted from the publication, but rather discussed. Those papers reveal the reality of the scientific process and are hugely informative in that way. They are also sometimes tedious to read for all the same reasons. Biographies of scientists also often describe the serpentine path of discovery and can be both inspirational and provide reality checks and should be recommended to all of us.
I wonder if it influences the attitudes of beginning trainees and causes them to mistakenly think of science as a straight line instead of meandering concentric circles.
Of course it does. It also influences people in a removed subfield to think that your approaches* are easy as pie and it is only their models/systems/assays that are a royal pain to accomplish.
*especially if you rely on a machine that goes “ping!”.
A wonderful book to read by Sir Peter Medawar is: “Advice to a young scientist”. I read it when I was a postdoc and it was very useful, reinvigorating and refreshing. Give it a try young guys!!!.
I would say to the grad student: you have to enjoy the process. Even as a full professor I really like doing experiments. I just want to see what happens if I do x. Is my idea right? Did I get some new information (even if sometimes only about a reagent or a method)? It’s a great day when I can come into work and know I will learn something new.
I often wish I were a grad student or a postdoc with the freedom to set up experiments every day..but I have to bring in the $$, so alas it is only about once a week right now, and much less before grant deadlines. Still, I think I will never lose “the pleasure of finding things out” (Richard Feynman). It’s the best!
“I just want to see what happens if I do x. Is my idea right?”
Is this trail and error? If so, it’s a very different brand from just blindly trying X, or Y, or Z, etc., without any clue or idea about why one might work and not the other. I can understand why grad students would hate that form of trail and error. That doesn’t sound like science to me — at least not the way I do it.
In order to narrow the possible things to try to X, Y, and Z, you had to have some sort of initial ideas.
narrowing of possibilities is based on an intuition that is often built off of years of experience. So yeah it’s not easy for a new grad student to understand why they should try those ten different things until they do it a bunch and develop their own intuition.
The secret of BSD success is like classic Michaelis Menton enzyme kinetics. Adding more substrate to a reaction (Funds/space/warm-bodies) increases the velocity of the reaction (productivity) until to approach the V-Max (limiting lab-space/funds/competition) and have a reduced velocity. Finding an optimal point on the curve with maximum velocity give you maximum productivity with minimal costs. The equation is thus:
V(productivity)= ([Funds/Space/Minions]* Average IF )/([Riffraff] + [Minions who can eclipse you])
The more minions you have, the less effect trial-and-error has on the equation since on average each minion will produce at a certain rate.
Grumpy is right. I have hypotheses before I start, but these are supported more by intuition than by actual data. Before I ask anyone in the lab to take on a new and risky project I like to do a first experiment myself- especially if reasonably easy (and not too expensive). The downside is that many if not most of these initial experiments negate the hypothesis. The bonus is that I am the first one to know the result, which is still- even after all these years- quite exhilarating. As a behavioral scientist will tell you, variable reinforcement is a powerful reward system…I remember every one of my Eureka moments.
Philapodia, is there an equation for risk vs reward?
Reward = (PI h-index)*(total University IDC)*(# Nobel Laureates on Faculty)
Risk = (PI Student Loan Debt)*(1/(Average number of Deanlets per college))*(classes taught by PI per semester)^(times University football team has won NCAA Championship)
@Grumpy and baltogirl: Yes, *you* are the ones with the intuition, and for you it’s rewarding to follow up on that. You know you’re not just taking shots in the dark.
But since you can’t explain your intuition to your new grad students in a way that would satisfy a scientist, i.e., since you can’t make scientific arguments for your gut, to them it *is* just shots in the dark.
TLDR: Following up on your own intuition is fun. Following up on someone else’s, not so much. The latter in no way suggests that someone is not cut out for science, which is the idea being floated here.
If a new grad student can’t trust their PI’s expertise enough to try a few experiments out based on their PI’s intuition, where the outcome is not guaranteed on the first shot, then yeah grad school is gonna be really brutal for them. (Restating DM’s original post)
if they instead try out a bunch of other stuff based on their own (likely unfounded) “intuition” because thats more “fun”, then they are going to be wasting a whole lot of ppl’s time, including their own.
I’m not picking on anonymous, I see this all the time with junior trainees.
Am I the only one who remembers what it’s like to be a grad student?
No one is saying that grad students shouldn’t listen to their advisor, Grumpy. All I’m saying is that we shouldn’t assume grad students won’t enjoy being scientists if they don’t enjoy the parts of grad school that are just not that much fun. Like doing stuff just because someone else told you to.
Did you become a PI because you enjoy doing what other people tell you to?
DrugMonkey 
April 6, 2016 at 5:05 pm
I think steve shea is worth following. no?
LikeLike
April 6, 2016 at 7:09 pm
I swear I just had this conversation with my new advisor, complete with 20 minutes of rambling. Is “trial and error” why a PhD takes 5+ years, or is it because of excessive advisor rambling?
LikeLike
April 6, 2016 at 9:09 pm
RM – no it’s because of smart-ass students:)
LikeLike
April 6, 2016 at 9:09 pm
Some people need to ramble in order to think. It’s a very common thing these days.
LikeLike
April 7, 2016 at 8:28 am
Remember the old joke that if it wasn’t for grad students the pace of science would be unbearably fast 🙂
LikeLike
April 7, 2016 at 9:14 am
“…bad news for them about grad school” – well, actually, the bad news is about science itself.
LikeLike
April 7, 2016 at 9:55 am
RM, let’s frame it this way: it’s very common to spend most of grad school figuring out which experiments you should do, and bring trained (or training yourself) how to do them well. Then the rest of it is spent actually doing them. My PhD took 6+ years, but if I’d known from the get-go what I needed to do and how to do it, I probably could have wrapped the whole thing up in 2.5-3.
LikeLike
April 7, 2016 at 10:05 am
Actually the bad news is about life, really.
Welcome to your mid-20s, grad student!
LikeLike
April 7, 2016 at 10:16 am
Wow. This is becoming unbearably depressing. It seems as if the one comment left is: “I gonna kill myself”. C’mon guys,!. There must be unexplored solutions/alternatives!
LikeLike
April 7, 2016 at 10:28 am
I don’t understand – isn’t the thrill of being a scientist actually doing experiments, aka trial-and-error? Why the hell would you want to be a scientist if you just follow directions? And I’ve found its amazing how frequently some random observation from a “trial-and-error” period in the lab comes back to be eminently useful months or years later.
LikeLike
April 7, 2016 at 10:38 am
Vertically ascending scientists do not do “trial and error”, they reveal “Truth” that was obvious to them from the onset and demonstrate that truth with one perfect unassailable experiment, whereas the riffraff fumble around with trying different things and wasting everyone’s time and money. That is what makes great scientists different from the unwashed masses and why only they should funded.
LikeLike
April 7, 2016 at 10:40 am
isn’t the thrill of being a scientist actually doing experiments, aka trial-and-error?
Yeh, but one doesn’t necessarily realize that until later on I think.
LikeLike
April 7, 2016 at 10:54 am
Hey, I’ve got to get my jabs in at the established folks when I can, especially when DM’s previous post tells me that I’m “cannon fodder”. The cannon fodder would prefer to remain blissfully ignorant for now, thanks.
LikeLike
April 7, 2016 at 11:54 am
It’s not “rambling”, it’s “information processing with realtime verbal output”.
🙂
LikeLike
April 7, 2016 at 12:15 pm
And then there’s parenting: the ultimate in trial and error. And error, and error, and error.
LikeLike
April 7, 2016 at 12:51 pm
I’ve gotten advice to hide the “trial and error” parts of experiments from papers and talks. Meaning, basically, present it like a non-facetious version of Philapodias post. Observation, hypothesis, answer.
I get we often have to do this for coherance and clarity, but I wonder if it influences the attitudes of beginning trainees and causes them to mistakenly think of science as a straight line instead of meandering concentric circles.
LikeLike
April 7, 2016 at 1:47 pm
Of course it does, which is why Philapodia’s post was ironic, not facetious.
LikeLike
April 7, 2016 at 3:27 pm
@jmz4
This is why all graduate students should read classic papers like Katz and Hodgkin (in my world, surely there are other field-specific versions) when the style of writing allowed for thought processes to be written out. False starts and blind alleys were not omitted from the publication, but rather discussed. Those papers reveal the reality of the scientific process and are hugely informative in that way. They are also sometimes tedious to read for all the same reasons. Biographies of scientists also often describe the serpentine path of discovery and can be both inspirational and provide reality checks and should be recommended to all of us.
LikeLike
April 7, 2016 at 4:27 pm
I wonder if it influences the attitudes of beginning trainees and causes them to mistakenly think of science as a straight line instead of meandering concentric circles.
Of course it does. It also influences people in a removed subfield to think that your approaches* are easy as pie and it is only their models/systems/assays that are a royal pain to accomplish.
*especially if you rely on a machine that goes “ping!”.
LikeLike
April 7, 2016 at 4:33 pm
A wonderful book to read by Sir Peter Medawar is: “Advice to a young scientist”. I read it when I was a postdoc and it was very useful, reinvigorating and refreshing. Give it a try young guys!!!.
LikeLike
April 7, 2016 at 8:20 pm
I would say to the grad student: you have to enjoy the process. Even as a full professor I really like doing experiments. I just want to see what happens if I do x. Is my idea right? Did I get some new information (even if sometimes only about a reagent or a method)? It’s a great day when I can come into work and know I will learn something new.
I often wish I were a grad student or a postdoc with the freedom to set up experiments every day..but I have to bring in the $$, so alas it is only about once a week right now, and much less before grant deadlines. Still, I think I will never lose “the pleasure of finding things out” (Richard Feynman). It’s the best!
LikeLike
April 7, 2016 at 9:29 pm
“I just want to see what happens if I do x. Is my idea right?”
Is this trail and error? If so, it’s a very different brand from just blindly trying X, or Y, or Z, etc., without any clue or idea about why one might work and not the other. I can understand why grad students would hate that form of trail and error. That doesn’t sound like science to me — at least not the way I do it.
LikeLike
April 8, 2016 at 1:32 am
In order to narrow the possible things to try to X, Y, and Z, you had to have some sort of initial ideas.
narrowing of possibilities is based on an intuition that is often built off of years of experience. So yeah it’s not easy for a new grad student to understand why they should try those ten different things until they do it a bunch and develop their own intuition.
LikeLike
April 8, 2016 at 9:08 am
The secret of BSD success is like classic Michaelis Menton enzyme kinetics. Adding more substrate to a reaction (Funds/space/warm-bodies) increases the velocity of the reaction (productivity) until to approach the V-Max (limiting lab-space/funds/competition) and have a reduced velocity. Finding an optimal point on the curve with maximum velocity give you maximum productivity with minimal costs. The equation is thus:
V(productivity)= ([Funds/Space/Minions]* Average IF )/([Riffraff] + [Minions who can eclipse you])
The more minions you have, the less effect trial-and-error has on the equation since on average each minion will produce at a certain rate.
LikeLike
April 8, 2016 at 12:56 pm
Grumpy is right. I have hypotheses before I start, but these are supported more by intuition than by actual data. Before I ask anyone in the lab to take on a new and risky project I like to do a first experiment myself- especially if reasonably easy (and not too expensive). The downside is that many if not most of these initial experiments negate the hypothesis. The bonus is that I am the first one to know the result, which is still- even after all these years- quite exhilarating. As a behavioral scientist will tell you, variable reinforcement is a powerful reward system…I remember every one of my Eureka moments.
Philapodia, is there an equation for risk vs reward?
LikeLike
April 8, 2016 at 1:57 pm
Reward = (PI h-index)*(total University IDC)*(# Nobel Laureates on Faculty)
Risk = (PI Student Loan Debt)*(1/(Average number of Deanlets per college))*(classes taught by PI per semester)^(times University football team has won NCAA Championship)
LikeLike
April 8, 2016 at 6:43 pm
@Grumpy and baltogirl: Yes, *you* are the ones with the intuition, and for you it’s rewarding to follow up on that. You know you’re not just taking shots in the dark.
But since you can’t explain your intuition to your new grad students in a way that would satisfy a scientist, i.e., since you can’t make scientific arguments for your gut, to them it *is* just shots in the dark.
TLDR: Following up on your own intuition is fun. Following up on someone else’s, not so much. The latter in no way suggests that someone is not cut out for science, which is the idea being floated here.
LikeLike
April 8, 2016 at 7:42 pm
If a new grad student can’t trust their PI’s expertise enough to try a few experiments out based on their PI’s intuition, where the outcome is not guaranteed on the first shot, then yeah grad school is gonna be really brutal for them. (Restating DM’s original post)
if they instead try out a bunch of other stuff based on their own (likely unfounded) “intuition” because thats more “fun”, then they are going to be wasting a whole lot of ppl’s time, including their own.
I’m not picking on anonymous, I see this all the time with junior trainees.
LikeLike
April 8, 2016 at 9:38 pm
Am I the only one who remembers what it’s like to be a grad student?
No one is saying that grad students shouldn’t listen to their advisor, Grumpy. All I’m saying is that we shouldn’t assume grad students won’t enjoy being scientists if they don’t enjoy the parts of grad school that are just not that much fun. Like doing stuff just because someone else told you to.
Did you become a PI because you enjoy doing what other people tell you to?
LikeLike