The following is a transcript of the presentation video, edited for clarity.

Strong Aims

Strong aims should be memorable. A great way to see whether you’ve done your job right is to hand your aims to somebody and take them away and say, “What were my aims?” See if they can remember what the heck you’re talking about. If you write these in a compelling way, people will really be able to remember them. But if you just sort of “blah, blah, blah” on the page, they’ll look at it and won’t be able to recall it.

The reason that’s so important is because while they’re reading your application, you want them to have those things in their immediate recall. You want that to be their touchstone for everything they read. So they really should be memorable.

Make them easy to read. One thing I really pay attention to in specific aims when I’m writing them myself is to really simplify the syntax. Make sure these sentences are easy to process. You can get through them quickly, and that the memory load— I am not joking about this—that the short term memory load on these is very light. You want people to just be able to skip along through that, so they’re working through these ideas quickly and succinctly, and not in a way that just goes increasingly deeper and deeper and deeper where you have to keep referring back up to the paragraph before where it can be very deep and complex ideas. That’s not where you want to be on the specific aims. You want to be succinct and memorable.
Your experiments can have expected outcomes, not known outcomes. That’s not good.

Your aims have to be programmatic and linked. This is a real challenge, to make them all address a common problem, and have them be linked—it’s nice if they’re linked in terms of even having the same subjects. But you have to be very, very cautious that one experiment, one aim doesn’t depend on another aim. Those are called “domino aims” like if aim one fails, if for some reason you’re unable to do aim one, aims two and three can’t be done or they’re no longer important, or no longer relevant. You’ve opened up the possibility of this whole thing being a waste. Reviewers are very, very sensitive to those domino aims.

They have to be programmatic and linked—there has to be a reason why you have these three aims all on the same piece of paper—but they can’t be so interdependent that any one of them can bring the others down. So that’s a challenge.

It’s cool if you can challenge an existing paradigm. But you have to be careful, because they’ve been doing that existing paradigm for 500 years and you’re a whippersnapper.

If it can address a critical barrier—if there’s some kind of thing that’s just this one breakthrough is needed, and you can address that, that’s great. Developing a new technology is good—it’s challenging to make developing a new technology a good aim, unless it’s like one aim where that technology is going to contribute to the other aims, but the other aims don’t depend on that technology, and if this technology actually allows you to look at something theoretical, if it somehow contributes to the theoretical framework.

Feasible, I talked about that already. Formal statements of a hypothesis.

This is the same idea again, but another way of saying it: Your aims draw a clear line through what is known, to your contributions to that knowledge (either your past contributions or your planned contributions), to the gaps in knowledge that this work is going to bridge, to how this study is just what needs to be done to complete that continuum.

Challenges and Problems in Constructing Your Specific Aims

People are often not specific in their aims, surprisingly. That’s the most common problem.

“Looking at X”—that’s right up there with “a paucity of extant literature”— this experiment is going to “look at” something. That is just the worst. That is not a motivation. You see that, it’s just problematic. There needs to be an experiment. Looking at something is not an experiment.

There has to be a theoretical framework. Don’t even think you can do this without a theoretical framework—that’s just a silly idea. The reviewers will look at something that doesn’t have a theoretical framework, and they’ll pull out this phrase which is just as bad as “Overambitious” and that’s “Fishing Expedition.” That is just throwing a toxic mess all over the review. You’re all done. “Well, it seems like a fishing expedition.” “Okay, next grant. Moving on.” The framework is what keeps it interesting, it gives you something to build this into. Otherwise you’re looking for an effect where there might be one—but probably not. You have to watch out for that.

I think I mentioned overambitious. Aims that aren’t specific. You don’t show a complete understanding of the area. These are all just common, common problems. Missing key citations.

Reviewers are prone to read something and see the statement that you make as some sort of statement of your belief. When you say something in first person, that makes that pressure even stronger. If you say outright “we believe”—then you’ve really lost them. Science isn’t a matter of belief. In fact, it’s kind of the opposite of belief. What you’re doing is you’re leaning on: There’s this fact, there’s this fact, there’s that fact—which would lead to the conclusion that this is … whatever. Probable, possible, likely, concerning, whatever. I work hard to stay away from first person language on that page. Through the whole grant if I can.

Then, a hypothesis that’s not falsifiable is the worst. If you have a hypothesis, and there’s no way for the result of the experiment that you propose to falsify that hypothesis, then you’ve proposed something that’s very weak.

I want you to watch out for this one: You’ve included way too many words, including many that are superfluous, unnecessary, and fail to add actual content to the narrative that you are building. And you could just say “too many words.” This goes back to lightening the load on your reviewers. Just say what you mean in as concise a way as you can. Make this easy. Make this the easiest thing they ever read. Don’t show how smart you are, show how considerate you are. Still reveal all your ideas. Get all your intellect out there, your intellectual firepower. But do not make them work hard to get it. Make it easy with bullets and figures and subheads and everything. Lead them carefully through it by the hand.

You are allowed two typos in your entire application. Five typos and people will say “This was carelessly prepared. I think they will probably do careless science as well.” I am not kidding. This is one of the most important documents you’ll write. It should be flawless.

Then, I’ll say this about writing to your target audience: Be very thoughtful in orienting yourself to this task. You can shoot too high. You probably are the expert on this technique. You probably do know more about this than anybody in that room. You can write this in a way that can shut people out.

If you write something in such an inaccessible way, this is actually a very insecure approach, I think. You’re hoping you’ll make the reviewers feel like if they just knew a little bit more, they’d realize how brilliant this was. That’s not the way they react. The way they react is, “I’m a smart person. I’ve read a lot of grants. They did not bring this idea to me. I will just keep turning to the next one.” You really are trying to make it accessible.

You’re really trying to make this idea accessible to reviewers—but not too broad. There’s a sweet spot. That’s the challenge. If you shoot too low and just describe everything in broad strokes, “We’re going to do the such-and-such paradigm on a bunch of kids, and we’re going to see if they do it.” Then it sounds like you’re counting on this halo that you’ve got. That you as a reviewer are just going to trust these folks. They must know what they’re doing if they’re able to be so casual about it. That is the “Just Send Ca$h” approach.

Finally, being vague about the populations you describe. If you’re going to put your population into your aims—which I would recommend you do—add a little bit of specificity. It helps to know if you’re doing something like language development, and you’re talking about two-year-olds versus three-year-olds, if you’re talking about kids in one educational context versus another, if you’re talking about different populations of motor disorder to know exactly what those populations are. It really makes a big difference. Use the extra words and stick them in there.

Audience Comments:

• One thing we’ve discussed in other meetings like this is to avoid verbs that lead you to think it’s a fishing expedition like “explore” or “look at” or even sometimes “investigate” if it’s just used kind of in a blanket non-specific manner. As opposed to verbs like “identify”, “isolate”, “correlate”, “test”, “determine”, “differentiate”. Then you feel good about experiments happening, as opposed to walks through the park where you look and explore.

• One of the challenges I have in reading specific aims is rarefied vocabulary. Vocabulary that is not defined. You can use a lot of words to sort of walk around your backhand, or walk around the vocabulary, but it is a challenge. Especially if you’re trying to do stuff pushing up against that corral fence.

Chris Moore:

That’s actually time to think about the third reviewer. You can adopt scientific jargon to show that you’re inside that club, but you can really alienate people at that same time. The same thing is maybe even more of a problem with acronyms. Acronyms just slow you down. I always recommend minimizing the use of acronyms.

• You were talking about “no domino effect”—but with intervention studies, when you do intervention, maintenance, generalization. You have no maintenance or generalization if you don’t have an intervention that makes its way through. So I’m always concerned about that.

Chris Moore:

You can address too much of the treatment continuum at one time. It would be appropriate to have concerns about an experiment that depended for a large part on an intervention in the early part that hadn’t been demonstrated yet. That experiment on the efficacy of the intervention has to be done first. That’s actually a perfect example of domino aims.

Preparing for Your First NIH Grant: More Videos in This Series

1. Who Is the Target Audience for Your Grant?

2. How Do I Determine an Appropriate Scope, Size, and Topic for My Research Project?

3. Common Challenges and Problems in Constructing Specific Aims

4. Are You Ready to Write Your First NIH Grant? Really?

5. Demystifying the Logistics of the Grant Application Process

6. Identifying Time and Budgetary Commitments for Your Research Project

7. Anatomy of the SF424: A Formula for NIH Research Grants

8. Common Strengths and Weaknesses in Grant Applications