Science Communication for Non-Scientist Funders: Closing the Gap

NIH's Plain Language policy, implemented under the Plain Writing Act of 2010, requires that grant application abstracts and project summaries be written in language accessible to a general audience. The requirement is not stylistic — it is structural. NIH program staff, Congressional appropriators, and oversight bodies use these summaries to understand and justify funding decisions to audiences who are not domain scientists. A project summary that is only legible to specialists in the funded investigator's subfield cannot do this job, which means it cannot serve the function the requirement was designed to serve.

Private foundations take an even more direct approach. The Wellcome Trust, the Bill and Melinda Gates Foundation, the Simons Foundation, and most major philanthropic science funders employ program officers whose scientific backgrounds range from deep expertise in the funded domain to adjacent science training to non-scientific policy or philanthropy backgrounds. These officers manage grant portfolios, conduct site visits, evaluate progress reports, and make recommendations to program leadership — all of which requires understanding the science at a level that is often above lay audience comprehension but well below the specialist level at which most scientific applications are written.

The gap between how researchers write and how non-scientist funders read is not a communication style problem. It is a problem of audience modeling: most researchers writing grant applications model their audience as a peer — a specialist who shares their background and can fill in unstated context. The non-scientist funder is not that audience, and writing for the wrong audience costs funding.

The Backgrounds of Program Officers at Major Foundations

Understanding who reads your application is the prerequisite for writing it effectively. Program officer backgrounds at major research-funding foundations vary considerably and are often publicly available through foundation staff directories and annual reports.

The Wellcome Trust's program officers in biomedical science typically have doctoral-level training in biological sciences but manage portfolios that span dozens of subfields — a single officer may oversee grants in immunology, infectious disease, and developmental biology simultaneously. Their scientific literacy is high but their subfield specificity is lower than an investigator who has spent a decade in one research area. They can evaluate scientific quality at the level of study design and methodological approach, but they depend on clear writing to understand the significance of findings in specific subfields.

The Gates Foundation's Global Health program employs program officers with backgrounds in global health policy, public health, medicine, and adjacent social sciences. Their evaluation of grant applications emphasizes implementation relevance, equity considerations, and measurable population-level outcomes. An application written primarily for an immunological audience — emphasizing mechanism and cellular biology — will communicate poorly to a program officer whose evaluation framework is built around disease burden reduction and health system strengthening.

The Simons Foundation funds mathematics and physical sciences as well as life sciences and autism research. Their program officers in mathematics and theoretical physics tend to have stronger technical backgrounds, but their autism research program serves a community of funders — parents and family members with direct experience of autism — who use program officer summaries to understand how grants connect to their interests. Writing that reaches only the scientific program officers fails the broader stakeholder audience that ultimately sustains foundation funding.

How Non-Scientist Funders Make Portfolio Decisions

Research funders — whether federal agencies, private foundations, or corporate research programs — make portfolio decisions based on explicit and implicit criteria that are not always visible in the published solicitation. Explicit criteria are stated in the request for proposals: the eligible mechanisms, the topic areas, the review criteria, the maximum award amounts. Implicit criteria are the portfolio-level considerations that program leadership uses to allocate resources across competing applications that have all met the explicit threshold.

For non-scientist funders, implicit portfolio criteria typically include: balance across topic areas within the funded domain; representation of early-career investigators; geographic or institutional diversity; complementarity between new grants and the existing portfolio; and alignment with the funder's current strategic priorities. These criteria are applied by program officers who are reading applications and making recommendations — which means they are applied by people who are interpreting scientific content through a non-specialist lens.

An application that communicates its significance, approach, and expected outcomes clearly to a non-specialist reader is an application that program officers can advocate for in portfolio discussions with program leadership and board members. An application that requires specialist knowledge to understand cannot be advocated for by a non-specialist program officer, regardless of its scientific merit. The advocacy gap is a funding gap.

What Plain Language Actually Requires

NIH's plain language guidance specifies several concrete requirements: use the active voice; write short sentences (average under 25 words); use common words instead of technical vocabulary wherever a common word is accurate; explain acronyms on first use; organize information from most to least important; and use headers and lists to support scanning. These are mechanical requirements, but they are grounded in readability research showing that documents meeting these criteria are comprehended more accurately by general audiences than documents that do not.

What plain language does not require — and where many researchers misunderstand it — is that the science be simplified or that technical claims be made imprecise. A plain language project summary can accurately convey a complex mechanistic finding, a sophisticated statistical approach, or a novel methodological contribution. What it cannot do is assume that the reader knows why the finding is significant, what problem the method is designed to solve, or why the methodological contribution advances on existing approaches. Those connections must be stated, not assumed.

The test for plain language in a research document is not whether a lay person would find it easy to read. It is whether a non-specialist scientist — someone with graduate training in an adjacent field — could read it without needing to look anything up, could explain the significance of the work to a colleague in a non-scientific role, and could accurately describe what the investigators will do and what they expect to learn. If yes, the document is doing its job. If no, the problem is almost always that significance and rationale have been left implicit rather than made explicit.

Writing for Mixed Expert and Non-Expert Review Panels

Some funding mechanisms use review panels that explicitly combine expert scientific reviewers with non-scientist evaluators — patient advocates in clinical research, community representatives in public health research, policy advisors in education research. These mixed panels are most common at foundation funders and in federal programs with explicit public benefit mandates, but they are increasingly common in federal health agencies as well.

Writing for a mixed panel requires layering the communication: the technical content must satisfy expert reviewers, while the framing and significance must be accessible to non-expert panelists. The practical approach is to ensure that every technical claim is preceded by a context sentence that establishes why the technical point matters, and followed by a consequence sentence that explains what the technical finding enables. Expert reviewers skip through context and consequence sentences because they already know the context. Non-expert reviewers depend on them to understand the technical content between them.

This layered structure — context, technical claim, consequence — adds words to a document and therefore runs against the compression instinct of researchers accustomed to page-limited applications. The investment is worthwhile for any application where non-expert evaluation is part of the review process, because the consequence sentences that an expert reviewer skips are the sentences that a non-expert reviewer scores on.

The Project Abstract as Primary Document for Non-Experts

For most non-scientist funders — and for the non-expert members of mixed review panels — the project abstract or summary is the primary document they use to understand and evaluate the application. The full Research Strategy or Project Description may be read by expert reviewers but is often not fully read by non-expert evaluators, who rely on the abstract to understand what the project proposes to do and why.

This means the abstract is doing double duty in applications reviewed by mixed audiences: it must be precise enough to give expert reviewers an accurate preview of what they will evaluate in the full application, and it must be self-contained enough for non-expert reviewers to understand the project's significance and expected outcomes without reading further. Most abstracts accomplish one of these goals at the expense of the other — they are either too technical to be accessible to non-experts, or too simplified to give experts an accurate preview of the scientific content.

The resolution is to write the abstract in two layers. The first layer — typically the first half — establishes the problem, the gap, and the expected outcomes in accessible language that does not require specialist knowledge. The second layer — typically the second half — provides the specific scientific approach and the primary research questions or hypotheses with enough technical detail to give expert reviewers an accurate sense of the methodology. Experts read the whole abstract. Non-experts read the first layer and use it to decide how to engage with the rest of the material they are evaluating.

Lay Summaries: What They Actually Require

Many foundation funders and some federal programs require a separate lay summary in addition to the technical abstract — a shorter, more accessible description written explicitly for a general audience. Lay summaries are commonly mishandled in two opposite directions: some are written so simply that they lose the specific claim of the research (reducing to "we will study X"), while others are so compressed from the full abstract that they are still inaccessible to a general reader.

An effective lay summary answers three questions in three to five sentences: What problem does this research address, in terms that anyone can understand? What will the researchers actually do? And what will success look like — what will be different in the world if this research succeeds? These three questions correspond to the problem-method-outcome structure of any scientific abstract, but framed at a reading level and vocabulary level appropriate for someone without scientific training. Research on lay summary effectiveness published in PLOS ONE and in science communication journals consistently shows that lay summaries that answer these three questions explicitly are rated as more compelling and informative by non-scientist readers than summaries that describe the research process without connecting it to a concrete problem and outcome.