It’s no secret that to be successful in science requires a wide range of skills. Some of these combine physical dexterity with attention to detail, like accurate pipetting, good sterile technique, and precise dissection skills. Others are based more on habits and expected behaviors, such as detailed note-taking, strong organizational skills, and clear communication. Additional examples include learning the nuances of experimental systems, operating complex instruments, and navigating sophisticated software. But at its heart, science is mostly about the ability of a researcher to think critically and creatively. Below, I outline several broad stages of scientific development, acknowledging that these are not entirely separate. These “levels” of scientific maturity encompass everything from ironing out the all-important details of an experimental design to seeing the big picture and conceiving of new research directions. For trainees, all of these will be essential if your goal is to develop into a full-fledged independent scientist.

I’ve discussed these stages below in order from early-stage scientists (i.e., undergraduates or first- or second-year graduate students) on through principal investigators. 

1. Being able to carry out and interpret a set of studies that have been largely dictated to you in some detail by someone else. This may sound trivial, but it’s not. There will still be a good deal of thinking and awareness required to be successful. Most importantly, there is no partial credit in the world of science (or in most skilled jobs). If you do nine steps correctly but mess up the tenth, that’s not an ‘A–’ it’s a ‘D’ (or maybe even an ‘F’). The outcome is likely unusable. Executing each step correctly requires attention to detail and an engaged, capable intellect. Initially, some of the required thinking may involve noticing if the protocol you’ve been handed is incomplete or contains a possible error or if there’s a problem with the starting materials. It’s also noticing if something goes wrong during the experimentation process. Rather than ignoring or glossing over anomalies, you note them down and minimally bring them to the attention of others. And then, once you have the data, you have to think about what conclusions, if any, can be drawn. Know that no matter where you are in your career, bringing data to a mentor without having made a reasonable attempt to rigorously interpret your own findings is never a good idea. Right from the beginning, you’re expected to be engaged and informed enough to draw at least some initial conclusions. In reality, a good number of aspiring students don’t make it past this first step. This may occur because they ultimately fail to summon the required focus and attention to detail or because they lack some essential ingredient, such as scientific curiosity or drive.
2. Being able to carry out and interpret a set of studies that have been outlined for you—but without extensive details. This requires considerably more independence, as you’ll need to figure out the best way to accomplish the set goals. Typically, this will necessitate searching the internet, reading the literature, communicating with knowledgeable colleagues, weighing the pros and cons of different strategies, and ultimately coming up with a very specific plan. There may also be considerable troubleshooting, because experimental science often doesn’t work the first time (or the second). It also requires coming up with the necessary controls to make any findings meaningful. The better you can anticipate the right controls, the more efficient you’ll be. But recognize that sometimes it just takes trying things and possibly doing some imperfect experiments before you can fully wrap your brain around the inherent issues, variables, and necessary controls. This level of science also demands an ability to be insightful, objective, and highly critical of your own work and data. You’ll want to minimize the extent to which others are required to point out the flaws in your data or reasoning. Still, it’s perfectly acceptable for you to point out your own mistakes, provided you gradually learn to keep these to a reasonable minimum. And if you happen to have overlooked something, which happens to everyone, own your mistakes and learn from them.
   If you can carry out Stages 1 and 2 consistently, you are well along the path toward a PhD.
3. Being able to independently come up with THE NEXT logical set of studies to fully answer a question and build a complete story. This is where real scientific independence starts. You might have been handed a solid initial result or a foundation for a project, and now you’ll need to see it through to completion. Many graduate student projects follow this kind of trajectory. To be successful will require a much deeper command of the literature, real intellectual engagement, and a dollop of creativity to figure out what experiments and general directions you’ll want to go in. The story already has a beginning and perhaps some suggestion of an arc, but it’s up to you to determine the middle and the end—namely, most of it. Critical questions likely include the following: what is already known, what is unknown but worth answering, and which experiments will be most informative in addressing these questions? And when you’re carrying out these studies, how adept will you be at identifying unexpected but intriguing results—findings that may ultimately prove more interesting than your original questions? This process requires precise logic, an engaged intellect, an ability to make connections between what you’ve done and what others have done, and the drive to fully immerse yourself in a problem and work your way to a successful outcome. It requires distinguishing experiments that could be done from those that should be done. And it requires a growing sense of what may constitute a legitimate dead end, one necessitating a shift to another more-fruitful line of studies. For this you will need to develop sound scientific judgment, a rather thick skin, and, most importantly, a willingness to accept that you alone are responsible for your ultimate success or failure. If you can do all of this, you deserve a PhD.
4. Being able to come up with a largely or entirely new direction within the general confines of a research group’s broad interests. This is getting into more advanced grad student, postdoc, and professor level territory. Here you’ll need to see the big picture and identify the most interesting questions and important gaps in the knowledge. You’ll also need to have the judgment to estimate some likelihood of success or failure and avoid projects that are perhaps too risky relative to their perceived payoff. Rather than being handed a foundation, you’ll need to create one. And once you’ve succeeded in doing so, you’ll need to efficiently execute Stages 1–3 to build the story and develop a potentially new direction of research. How you establish that initial toehold can vary. Sometimes, it’s fairly simple. For example, you’ve read about gene X and technique Y, so you think, “It would be interesting to see what happens if I do technique Y on gene X”. Obviously, you’d need to have a sound justification for doing this—good science isn’t about rote mix and match experiments that require little imagination or insight. Alternatively, you may decide to develop or refine a novel technique or adapt an established protocol to a new system, provided there is a clear reason to do so. Or a novel idea may crystallize in your mind after digesting the scientific literature, perhaps in combination with hearing a talk or attending a meeting. All these scenarios will likely come with some potential for failure. Therefore, it’s also important to know when a project isn’t bearing fruit and when it’s time to cut your losses. Often you may still be able to extract something worth publishing, but it should no longer be your main focus. In such cases, having several projects going simultaneously can provide a critical safety net. Admittedly, without people taking chances and delving in new directions, science would rapidly stagnate. Just make sure you can justify the risk of a new project with its potential payoff.
5. Being able to think under pressure on the fly. Doing science is admittedly more of a marathon than a sprint. (Although addressing reviewer’s concerns during the publication process can often feel like a sprint at the end of a tiring marathon.) Rapid response–type thinking probably isn’t an essential skill on most days. But what happens when you give a talk and get a question? Do you freeze up or can you relax enough to come up with an informed, measured response? Of course, being relaxed won’t help much if you haven’t done your homework. If you are intellectually lackadaisical or incurious, questions that require thinking will always rattle you. And while some people may naturally function better in pressure situations, it’s mostly a matter of practice. Virtually all scientists (myself included!) have some painful memories of falling short in this area, especially early in our careers. But the more you read and think on your own, the better you’ll be able to handle these situations. Moreover, the more you practice a presentation, the more relaxed you’ll feel while delivering it, which can do wonders for your ability to grasp and answer questions. With time you’ll come to welcome and enjoy tough questions. But like writing and most other skills, it takes practice, time, and a willingness to put yourself out there to get better.

Some additional brief thoughts:

• Again, just because you can think of an experiment doesn’t mean that you should do an experiment. This is also relevant when suggesting experiments to others, either in group meetings or when reviewing papers. Always consider why the experiment is important.
• Never bullshit. If you don’t have a decent answer to a question, just say something like, “I don’t have a good answer for you right now, but I will think about it and try to get back to you.” Spewing obvious garbage makes people lose respect for you. It’s game over in their mind.
• Reading papers, including those outside your immediate field, is probably the best way to come up with new ideas. And it’s something we should all probably do more of. It stimulates the mind in ways that can’t be anticipated.
• Don’t wait for your supervisor to send you papers to read. You should be sending relevant papers to your supervisor.
• Develop the ability to use your intellect like a powerful zoom lens, shifting your focus rapidly from the relevant scientific minutiae, to the assembly of a publishable unit, and then to the wide-angle view of where your work fits into a bigger picture. 
• Trust your instincts, logic, and moral compass. If something seems wrong, then it probably is wrong and needs to be addressed.
• Working hard is important, but also take time out for other interests. Breaks can be helpful for your science—in fact, they can be essential. 
• Self-care (e.g., exercise, sleep), hobbies (e.g., showing pet llamas), and relationships (friends, family, etc.) are critical for scientific longevity. You’re in this for the long haul—avoid burnout.
• Aha moments can come at the weirdest times, often when we’re doing something else, like taking a hike or cooking or when we’re in the middle of a morning shower.
• Talking to your friends and colleagues about your science, such as others in your research group or department, can have huge benefits. Sometimes they may provide you with a great idea or at least plant a seed that will grow into a great idea. At other times, you will spontaneously come up with an answer while describing your problem to someone else.
• See the paper. What’s the story you’re going to tell? Always frame your work and your thoughts in the context of having to eventually publish your studies. Visualize the paper starting on day one of your experiments. 
• Make clear formal figures or panels as you’re going along. Don’t wait until the end when it’s time to write the paper. There is something about seeing your data in a format that looks like the figure in a paper that will make you view it more objectively and critically.
• Multitasking and good organizational skills are essential. Among other things, having multiple irons in the fire means that your chances of getting some positive results on a regular basis are much greater. This also makes it much easier to ditch projects that aren’t panning out.
• Take needed time to think. You are a sentient being, not a robot. That said, without a lot of doing, thinking becomes pointless, and robots are quite admirable for their consistency.
• Avoid the trap of becoming too attached to a single tool. As the saying goes, if you only know how to use a hammer, then all problems start to look like nails. This eventually leads to dull, unimaginative science, where the why is subjugated by the how.
• Build meaningful relationships and be generous with others. I promise this will come back to you in a good way.