The Frontal Cortex

Scientific Prose

Posted by Jonah Lehrer on June 26, 2008
(9)
More »

There’s an interesting review on prediction errors and temporal difference learning theory in the latest Trends in Cognitive Sciences. (Really, it’s fascinating stuff.) But I don’t want to talk today about the content of the article. Instead, I want to discuss its form.

The vast, vast majority of science articles follow the same basic pattern: abstract, introduction, methods, results, discussion. (Update: As bsci points out in the comments, review articles obey a slightly less strict pattern, but they’re still pretty predictable.) There are no stories, no narrative, no amusing anecdotes. (Occasionally, there’s a pretentious epigraph – papers on LTP and memory always quote Proust – but that’s about it.) Rather, there’s just line after line of jargon leaden prose in the passive tense. If you’re lucky, you might get some passive-aggressive criticism of a competing hypothesis.

This is why the prediction error article got my attention. The authors, Yael Niv and Geoffrey Schoenbaum, didn’t obey the typical journal format. Instead, their paper consists of ten fundamental questions (i.e., “What is a prediction error and what is it good for?”) followed by a series of straightforward answers. The prose is elegant and clear, sprinkled with helpful metaphors. (Their opening metaphor about fancy Bordeaux wine was particularly helpful.) The images are relevant Dilbert cartoons, which say more than most bar graphs:

i-7bebbbe9ad1376932d8fba65c980058d-Dilbert_prediction error.jpg

Why can’t more science papers break the mold, even a little? I’m not saying scientists should start writing like Robbe-Grillet, but why not experiment a little with the writing process? Niv and Schoenbaum confess that the form of their article happened by accident:

This manuscript grew out of a set of e-mail questions Geoff sent Yael after they met at a reward circuits meeting held at Lake Arrowhead in California in 2006. The answers to the questions were so useful that they were often printed and passed around the laboratory. This generated further discussion and ideas and, ultimately, more questions. Because these dialogues were extraordinarily useful to the authors, they thought they also might be helpful to others.

It’s too bad scientists can only come up with new ways of writing science papers when they think they’re not actually writing a science paper. I understand, of course, the need for standardization and clarity, but I still think it’s time to update and liberate the format of most peer-review articles. Consider the results section. Instead of wasting space on PCR primers or the details of fMRI experiments, why not put all that stuff online, in a supplemental file? What I’d find much more useful is a youtube video of the experimental process, a video demonstration of how you sorted the fruit flies, or measured the mice in the water maze, or tested subjects in the brain scanner.

And why is there a prohibition against good science stories in science journals? Why not take a few sentences and describe where the idea for the experiment came from, or some of the experimental failures along the way, or the most surprising results. I’m not asking for a novel: all I want is an anecdote or two, a few honest insights into the messy reality of the scientific process.

(9)
More »

Comments

  1. #1 bsci
    June 26, 2008

    This is a review article! I just looked at the past three issues of TICS and NONE of the review articles have the intro,methods,results,discussion format. In fact, they seem to allow a lot of flexibility in formatting and some don’t even have an introduction. Sure this article is interesting and the journal probably had fun getting Dilbert licensing for republication, but please do some cursory research before proclaiming something unique.

  2. #2 PhysioProf
    June 26, 2008

    There are no stories, no narrative, no amusing anecdotes.

    I agree with the absence of amusing anecdotes, but the standard form of a scientific research article is exactly a narrative or story.

    We understand this, that and the other thing about blah, blah, blah. Based on this understanding, we pose the following hypothesis. In order to test this hypothesis, we did x, y, and z, and observed a, b, c. On the basis of these observations, we conclude that blah, blah, blah, thus supporting/rejecting this hypothesis. This implies blah, thus suggesting the importance of blah, blah, blah, and leading to the following future studies.

    What is this but a narrative or story? And the funny thing is that it is almost always a completed fabricated story. Most of the time, and experiment is performed after saying “let’s see what happens if we do x to y”. After you see the results, you go back and make up a hypothesis that the experiment tested, and which the results either support or exclude.

    All this “in order to test this hypothesis, we did this experiment” is almost always a total lie.

  3. #3 Dan
    June 26, 2008

    As a non-academic, I agree with the “readability” issue with most journal articles. On the other hand, it has sparked the movement of “blogging on peer-reviewed research”, ala Dave Munger and his researchblogging.org, and the need to interpret and explain in a more reader-friendly format. The other fun part is reading the press release write-ups on the research put out by the sponsoring university. “Of Two Minds” had a great example of that this week.

  4. #4 jonah
    June 26, 2008

    Thanks for your thoughtful comments. And yes, review articles in general (and Trends papers in particular) tend to be a little less predictable when it comes to format. But given the fact that review articles are supposed to represent a particular viewpoint, it’s still pretty surprising that virtually all review articles follow the same tedious pattern.

    As for Physioprof…That’s a great point about the inherent dishonesty of the scientific paper format. It’s funny how science papers still subscribe to a fictional version of the scientific process taught to fifth graders. (First, generate a hypothesis. Then, design an experiment to test the hypothesis…) So yes, I guess you’re right: science papers do present a narrative, they just tend to be incredibly dry, tedious and false. Not a great combination.

  5. #5 PhysioProf
    June 26, 2008

    So yes, I guess you’re right: science papers do present a narrative, they just tend to be incredibly dry, tedious and false.

    Maybe this is an issue for non-scientists reading scientific research, but it is a non-issue for the scientists working in a field themselves. The structure is very formal and rigid, so that the maximum amount of information can be accessed by the reader as efficiently and predictably as possible.

    Scientific research papers are not meant to be read through like a magazine article. They are meant to provide a highly structured format for presentation of a ton of detailed information, so that a user of that information can home in on exactly what they want as quickly and easily as possible.

    And the fact that the narrative is a lie is completely irrelevant. Everyone knows it’s a lie. It’s just a rhetorical device that allows for the experimental results to be placed in an appropriate conceptual context and related to existing work in a field.

    Personally, I would be very annoyed with an author who deviated from this very strict ritualized structure for a scientific paper, because it would make my work more difficult. Don’t mess with this!

    Where the writing of scientific papers could be dramatically improved is not in their top-level rhetorical and literary structure, but in the construction of decent clear English sentences.

  6. #6 Rman
    June 26, 2008

    I think the real issue here is that most scientific publications are incomprehensible by the general public. The purpose of scientific publications is to spread scientific knowledge to the world, however scientific publications are filled with so much jargon that only those in a small field can interpret the information. I believe that if scientific publications were written so that they are easier to understand to the general populous, that this would increase scientific literacy in the country and improve science education. Here in the United States there is a major concern about the future of science due to the lack of students entering science fields. Improving the literatre to allow more individuals to understand science might lead to greater interest in scientific subjects.

  7. #7 Mr. Gunn
    June 26, 2008

    Agreeing with Physioprof. It’s a structured format because it aids rapid interpretation. We’re not looking to be entertained.

  8. #8 Kjerstin
    June 27, 2008

    What impresses me most, is the way scientific writing always succeeds in concealing the fact that human minds were involved in the research. “Things were done.” “These results were produced.” It’s like someone just fed the previous studies cited in the introduction into the science machine and pressed the “perform research” button. And they still wonder why mainstream media can never get beyond the “Scientists have discovered…” level.

  9. #9 Janne
    June 27, 2008

    “The purpose of scientific publications is to spread scientific knowledge to the world, [...]”

    No. The purpose of scientific publications is to act as a record of note for accomplished work, and spread new scientific knowledge to other workers in the same field, in that order.

    The specialist non-scientist – the science reporter, the popular science writer, the fact checker, the textbook author – who has the domain knowledge to understand individual papers and be able to put them into a larger framework is the one whose purpose is to spread that scientific knowledge to the world at large.

    One paper by and large contains one idea. One result, one nugget of new information. No matter how well written it would be, it is normally simply at a too detailed a level to be of any interest or even be legible all by itself. It is only of any value when read in the context of many dozens or hundreds of other, often similarly detailed papers, and no matter what the structure or language, no layman is going to do that – or remain a layman for long in the rare case they persist.

    PhysioProf nails it above. Papers need a specific, common, well-understood structure. Since you need to look over dozens at a time that is the only way you can directly relate and compare them. That the structure may be an uncomfortable fit for a specific paper is irrelevant since the paper has little value by itself; neither is it relevant if a non-specialist has difficulty with the structure since they are not the target audience.

    The paper format for a specific domain is like having a specific, common file format for spreadsheets, or a common XML-format for, oh, business inventory. You don’t argue that each spreadsheet should have its own file format that fits it perfectly, and not argue that all inventory posts should have their own incompatible set of tags. A paper format is no different.