Do you know the feeling of staring at an empty page? Cursor blinking. Coffee getting cold (but still drinking it). You kind of know what you want to write: you have an experiment, some data, and you have a fairly good idea of how your data relates to previous studies. But somehow the words don't appear.

It feels awful, that's how. And if you're a researcher struggling with a manuscript, you probably know the feeling.

Below are a few tips to just write. There's no magic bullet; but a few common-sense tips can help you to get over the worst of your writer's block. And for what it's worth: they work for me.

Here's a truth: everyone who writes occasionally gets stuck. As I was about to delete the previous sentence because it's ambiguous, I realized that it's doubly true. Because [ everyone who writes | occasionally gets stuck ]; that is, even writers who normally write easily sometimes have trouble getting their words on paper.

But [ everyone who writes occasionally | gets stuck ]; that is, if you don't write often, you'll get stuck for sure, and get stuck properly. A writer is like an engine that needs to keep running; otherwise she gets cold and rusty. Many researchers write only a few short papers a year, and that's not enough to keep the engine warm. Clearly, the best remedy is therefore to write more: more papers, blog posts, love letters—anything to keep the engine running. (WhatsApp, Facebook, and Twitter don't count though.) But this is hardly valuable advice, because not everyone has the time, or the desire, to write more.

More realistically, you can preheat your engine by starting with the easy bits of your manuscript; if you're writing a research paper, these are generally the methods and results. These sections are formulaic, and therefore won't give you writer's block—or less so than the more difficult parts that do not have a fixed structure (but see below for my advice about templates). And by the time you've written the easy bits, your engine has defrosted, and you can start to take on the difficult bits.

Once you start writing the difficult parts of your manuscript, usually the introduction and discussion, you may get lost in a maze of your own words. You may find yourself asking: What exactly am I saying here? Does this point really follow from that point? And how relevant is this paragraph really? It's hard to keep writing if you don't understand what you've written so far.

If that happens, you may find it helpful to simplify your writing: shorten your sentences; replace vague words by concrete ones; use the active voice; and just say what you want to say in simple words. Once you've removed the fluff, you will find it easier to discern the logical flow of your own writing, and to recognize the parts and connections that don't make sense and therefore need to be revised. (Writing plain English is also a courtesy to your readers, because no-one enjoys dense academic gobbledygook.)

If you have trouble finding the right structure, it may be helpful to start from a template: a simple recipe for a decent text. And once you have something decent, you can polish it into something more than decent, maybe even good. Personally, when I write the introduction of a paper, I usually have the following template in mind:

1. First, in a short opening paragraph, I introduce the subject in a way that normal people can understand, but without simplifying it so much that my colleagues get annoyed: Visual attention and visual working memory are linked; that is, if you keep something in working memory (say a red apple) then things that resemble what you're keeping in memory (such as a red pepper) capture your attention. […]
2. Next, I describe one study that supports what I just said. I want the reader to understand the study, so I take my time to explain it; and I try not to overwhelm the reader by describing a lot of studies that all make similar points: To investigate the link between attention and working memory, Famous-Professor and colleagues conducted an experiment in which they […]. They found that […]. Based on this they concluded that […].
3. I then describe another study in the same way. I feel that describing two relevant studies is often the sweet spot. But only if they are complementary, and not just different ways to make the same point.
4. Then it's time to introduce what we don't know: Although previous studies have shown that […], we still don't know whether […].
5. Because that's what we're going to study: Therefore, to investigate whether […], we will conduct an experiment in which […]. We predict that […].

A template such as this is not an infallible recipe for a perfect paper. It's just a way to get something decent on paper that doesn't require too much creative thought. But when you have writer's block, you don't need perfection—anything will do.

Speaking of perfection: don't. Don't be a perfectionist. If you have writer's block (or in general, really), the best way is to start typing without thinking too much; and no going back and deleting a sentence that you just wrote either! Of course, you should revise your writing often, but you first need something to revise. And perfectionism stands in the way of getting that first something.

So those are my suggestions to overcome writer's block: preheat your writing engine by starting with the easy bits, use plain English so that you can understand your own writing, use a mental template that gives you a predefined structure to work with, and don't strive for perfection. Of course, overcoming writer's block is only the start—but it's a good start.