NIH Specific Aims Page (1 page)

Name: skills/research-grants/specific-aims
Author: pradyumnasagar

A working framework for drafting and revising the NIH Specific Aims page. The Specific Aims page is the single most-revised document in any NIH submission; it is the first thing the reviewer reads, the only thing many reviewers re-read, and the document that drives the Approach. A focused 1-page Aims is the difference between a focused 12-page Approach and a 12-page shopping list.

When to use

Drafting the 1-page Specific Aims page for an NIH R01, R21, R03, R33, U01, or similar research grant.
Drafting the 1-page Specific Aims for an F-series (F30, F31, F32) or K-series application.
Self-reviewing a draft Aims page and identifying whether it is a 5-year plan or a 1-page wish list.
Iterating from v1 to a publishable v5.
Reviewing a colleague's Aims and giving structured feedback.

When NOT to use

This is not the Research Strategy. The Aims page is the cover page for the Research Strategy. The Research Strategy (Significance / Innovation / Approach) is a separate document.
This is not the Abstract. The Abstract is 30 lines, contains the public-health relevance, and is a separate attachment. The Aims page is a 1-page document.
This is not a Specific Aims page for a non-NIH mechanism (NSF, ERC, DARPA). Other mechanisms have their own structural conventions; the structural points in this skill transfer, the rhetoric does not.
This is not the right skill for a project narrative, a project summary, or a public-health-relevance statement (each is a separate NIH attachment with its own constraints).

Prerequisites

A working knowledge of the FOA's Specific Aims page limit (1 page is the default; some FOAs differ).
A central hypothesis or a central aim — the single sentence the project is about.
A 4–5 year timeline (for R01) or a 3–5 year timeline (for K / F).
Preliminary data that supports the central hypothesis, even if modest.

Core workflow

Write the take-home sentence first. The take-home sentence is the central claim of the project. If the take-home sentence cannot be written in one sentence, the Aims are not focused. Test the sentence against three rules: (1) it is a claim, not a topic; (2) it has a verb that asserts; (3) it is falsifiable.
Choose the number of Aims. The default is 2–3 Aims. One Aim is sometimes appropriate (small R03, R21, F30, F31, F32). Four Aims is rare and risky. The number of Aims is a strategic choice: too many Aims diffuse the project; too few Aims leave the reviewer wanting more.
Order the Aims. Aim 1 should be the most-supportive Aim — the Aim that, if successful, most strongly supports the central hypothesis. Aim 2 builds on Aim 1. Aim 3 is the most ambitious or the most risky. A reviewer reading only the first Aim should still be convinced the project is worth funding.
For each Aim, write the rationale, the approach, and the decision rule.
- Rationale (2 sentences): what is known, what is missing.
- Approach (4–6 sentences): model, perturbation, readout, n, analysis, expected outcome.
- Decision rule (1 sentence): under what observation is the Aim considered "supported" or "rejected."
Write the opening paragraph (gap + centrality). The opening paragraph is the most-revised part of the Aims page. It must:
- Open with a clinical or biological problem the reviewer cares about.
- State the gap in current knowledge.
- State the central hypothesis.
- State the long-term goal and the project's role in it.
Write the closing paragraph (payoff + team + why now). The closing paragraph is short (3–4 sentences):
- What advances if the project succeeds.
- Why the lab is positioned.
- What the next-stage project is (R01 → R01; K → R; F → K).
Self-review against the page budget. A 1-page Aims page is approximately 550–650 words including the title. Anything above 700 words is too long. Cut.
Read aloud. Read the Aims aloud at a measured pace. The reading time should be ~3.5–4 minutes. If it is shorter, the Aims is too thin. If it is longer, the Aims is too long.
Iterate. The Aims is the most-revised document. Plan 3–5 drafts before submission.

Document patterns

Specific Aims page template (1 page)

# Title (the title itself is part of the 1 page)

## Opening paragraph (gap + centrality; ~7–10 sentences)
- Sentence 1: The clinical or biological problem.
- Sentence 2: Why it is unsolved.
- Sentence 3: The barrier (a specific mechanistic or technical gap).
- Sentence 4: The central hypothesis.
- Sentence 5: The long-term goal.
- Sentence 6: The objective of this application.
- Sentence 7: The team's prior work that grounds this proposal.
- Sentence 8 (optional): The institute / FOA alignment.

## Aim 1 (3–5 lines each: rationale, approach, decision rule)
**Aim 1: [verb-driven title, e.g., Determine whether X regulates Y in Z].**
- Rationale: ...
- Approach: ...
- Expected outcome / decision rule: ...

## Aim 2 ...
## Aim 3 (optional) ...

## Closing paragraph (payoff + team + why now; ~4 sentences)
- Payoff: if the aims work, the field will have ...
- Team: we are positioned because ...
- Next stage: this R01 lays the groundwork for ...
- Why now: a specific recent advance (a paper, a reagent, a dataset) makes this feasible.

Aim 1 construction (concrete)

**Aim 1: Test whether [molecule X] regulates [process Y] in [model Z].**
- Rationale: Prior work shows X is required for Y (Ref 1, 2), but
  whether X is sufficient, and the mechanism by which X acts on Y,
  is unknown. We will test sufficiency using a gain-of-function
  perturbation in Z.
- Approach: We will use a doxycycline-inducible X transgene in Z
  (n=12 per group, 4 cohorts). We will measure Y by [assay] at
  0/1/2/4 weeks and quantify with a [linear mixed model]. We will
  test sufficiency (Aim 1a) and necessity via conditional knockout
  (Aim 1b).
- Expected outcome / decision rule: Aim 1 will be considered
  supported if X overexpression increases Y by ≥1.5x (p<0.01) and
  X knockout reduces Y by ≥50% in the same model. If Aim 1 is
  supported, we will proceed to Aim 2 (mechanism).

Central hypothesis test (3-rule test)

Rule 1 (claim, not topic):
  FAIL: "We are studying mRNA methylation in glioblastoma."
  PASS: "N6-methyladenosine on MYC mRNA drives treatment resistance
        in glioblastoma, and targeting the methyltransferase reverses
        it in xenografts."

Rule 2 (asserting verb):
  FAIL: "We will explore / discuss / investigate / look at..."
  PASS: "We will determine / test / show / demonstrate / cause / predict / establish."

Rule 3 (falsifiable):
  FAIL: "We will study how cells respond to stress."
  PASS: "Heat shock factor 1 activation is required for survival
        after 42°C for 30 minutes in primary cortical neurons."

Opening paragraph formula (in order)

[Sentence 1: The clinical or biological problem, in 1 sentence.]
[Sentence 2: The state of current practice or knowledge.]
[Sentence 3: The specific gap (a missing mechanism, a missing dataset,
            a missing reagent).]
[Sentence 4: The barrier the gap imposes (why the field cannot progress
            past this).]
[Sentence 5: The central hypothesis, falsifiable.]
[Sentence 6: The long-term goal of the lab (programmatic direction).]
[Sentence 7: The objective of this application (the specific question
            addressed in this proposal).]
[Sentence 8: The team's prior work (one or two citations).]

Closing paragraph formula

[Sentence 1: Payoff — if the aims work, the field will have X, Y, Z.]
[Sentence 2: The team's positioning — we are positioned because of
            [prior work, expertise, infrastructure, collaborators].]
[Sentence 3: The next stage — this R01 lays the groundwork for
            [next R01, next translation, next application].]
[Sentence 4: Why now — a specific recent advance makes this feasible
            (cite a paper, a reagent, a dataset).]

Page-budget check (1 page)

- Title (line 1): 1 line.
- Opening paragraph: 6–10 sentences.
- Aims: 3–5 sentences each; 2–3 Aims; total ~10–18 sentences.
- Closing paragraph: 3–4 sentences.
- Total: ~600 words.
- Aims page that is shorter than 500 words is too thin.
- Aims page that is longer than 700 words is too long.

Common pitfalls

Topic, not claim. "We will study X" is a topic. "We will determine whether X causes Y in Z" is a claim. The reviewer is buying a claim, not a topic.
Three Aims with no decision rule. Aims without decision rules are aspirations, not experiments. A reviewer cannot tell when an Aim has succeeded or failed.
Aim 1 is the most ambitious Aim. Aims should be ordered from the most-supportive to the most-ambitious. Aim 1 should be the Aim that, if it works, makes the rest of the project worth doing.
Opening paragraph that opens with the lab. The opening paragraph opens with the problem, not with the lab. "Our lab has studied..." wastes the first sentence.
Aims that depend on Aim 2. Aims must be independently fundable. If Aim 1 cannot be evaluated without Aim 2's results, the Aims are not independent.
Closing paragraph that does not say "why now." "Why now" is the discriminator between an application and a wish list. A specific recent paper, reagent, or dataset must be cited.
Title that wastes a noun. "Studies on..." titles waste a noun that could be a claim. Use a verb-driven title.
Aims page that reads as a method list. The Aims is about questions, not about assays. The Approach is where the methods live.
Inconsistent language between Aims page and Approach. If the Aims page promises "we will determine X causes Y," the Approach must include a causal test. Aims and Approach should be co-revisioned.

Validation

The Aims page is 1 page (≈600 words, single-spaced, 0.5"–1" margins, NIH-compliant font).
The central hypothesis passes the 3-rule test (claim, asserting verb, falsifiable).
Each Aim has a rationale, an approach, and a decision rule.
Aim 1 is the most-supportive Aim (the Aim that, if it works, makes the rest worth doing).
Aims are independently fundable.
The opening paragraph opens with the problem, not with the lab.
The closing paragraph includes a "why now" sentence.
Reading the Aims aloud takes 3.5–4 minutes.
A reviewer who reads only the Aims can summarize the project in one sentence.

Open alternatives

NIH Forms-H/I vs. older Forms-G. Forms-H/I is the current package; Forms-G had slightly different page conventions. Always use the current package per the FOA.
1-page vs. 2-page Aims for K-series. Some K-series FOAs allow a 1-page Aims; verify the FOA. The structure is the same; the budget is different.
Alternative: Specific Aims is not a substitute for the abstract. The abstract is a 30-line, separately-scored attachment. The Aims page is the substantive document. Draft both.

References

NIH: Forms-H/I application package — Specific Aims page is the first attachment after the Project Narrative.
NIH: SF424 (R&R) Application Guide — Specific Aims section.
NIH: Grants Policy Statement (NIH GPS) — current revision.
NIH: RePORTER and RePORT matchmaker — funded-Aim examples and study section mapping.
NIH: NIH Peer Review Process — review criteria (Significance, Investigator(s), Innovation, Approach, Environment).

Changelog

1.0.0 (2026-06-10): Initial adaptation by Pradyumna Jayaram.