2456: Types of Scientific Paper
Explanation
In this comic, Randall satirizes scientific papers by claiming they all (or at least the large majority) fall into only a small number of categories, which he describes with somewhat humorous generalized titles. This comic may be a jab at mainstream news and their handling of scientific announcements; journalists are eager to report on what could turn out to be a scientific breakthrough even if it's very similar to stories they've already published about similar papers that turned out to be somewhat mundane. It may also relate to how some scientific findings follow certain patterns.
Breakdown of papers
Paper Title | Explanation | Visual Description |
---|---|---|
We put a camera somewhere new | This may involve miniaturisation or other improvements of imaging sensors, power supply, transmission or retention of data, environmental hardening and (possibly) recovery afterwards. Photographs and videos can be especially helpful in understanding what is or was going on, especially for the layman, than more limited signal traces.
Cameras have been inserted into every obvious bodily orifice (including swallowed, to be later excreted), placed in habitats to monitor wildlife, attached to wildlife to monitor habitats, sent into volcanic craters/ocean trenches/high altitudes/nuclear reactors, launched into space and sent past/round/onto several of the solar-system's more interesting bodies. This makes the "somewhere new" claim intriguing, possibly even comparable to 'clickbait'. This could also be generalized even more by replacing "camera" with "sensor", and then going to debate the newly derived sensor data. |
Includes a large figure, likely an image captured with the camera. |
Hey, I found a trove of old records! They don't turn out to be particularly useful, but still, cool! | Rather than starting with the aim of investigating some question, and finding some way of answering it by uncovering evidence, sometimes a writer may have stumbled upon a cache of historic documents that they then feel compelled to justify the resulting 'WikiWalk' they may have found themselves sucked into. The author may be far more excited about this than any future reader. This could also be a paper by a historian who found out ancient records which could be useful. A similar sentiment appears in 1979: History. | Small figure may show the most interesting fragment of the records. |
My colleague is wrong and I can finally prove it | This title refers to the occasional rivalries between scientists within a field, which can push them to seek proof that they, and not their colleague, are correct. This search for vindication, rather than truth, can lead the researcher to overlook significant flaws in their research, and can also make for tedious reading. It reflects a tone of smug self-satisfaction. | Like the two previous types, this one is laid out in two-column format with occasional section headings. Unlike those types, it has no figures. |
The immune system is at it again | The human immune system is notoriously complex, and there are countless papers in medical fields just describing its strangeness. While it is best known for preventing and battling infections, it can also overreact or even turn against the body that it is supposed to protect. Moreover it can overreact, for instance in allergic reactions or in a potentially lethal cytokine storm known to occur in certain viral infections, including influenza and COVID-19. The title may convey exasperation with the amorphous nature of this study subject. | Two-column format, no figures. |
We figured out how to make this exotic material, so email us if you need some | Researchers often attempt to create materials despite there not being any demand, predicting that in the future their material will be game-changing without any actual applications. These researchers have created such a material, and are offering to produce it for anyone who needs it. It is couched in terms of having created an answer for which there was not yet any proper question.
Sometimes the only reason for trying to create the materials is that they are peculiar or novel in some way, with no consideration of whether they might have any use apart from demonstrating the novel property. However, some other people might find that property very useful... and may lack the ability to make the substance for themselves. This may be also referring to the discovering/creating of elements and subatomic particles. The statement if you wish to buy it is humorous in these cases because they will decay too quickly to be purchased. |
Two-column format, no figures. |
What are fish even doing down there | Deep sea marine biology regularly discovers strange lifeforms in unexpected places, and theories explaining deep sea ecosystems are regularly confounded by new data.
Scientists may also bump into marine organisms when looking for something else. For example, one planned underwater neutrino detector picked up bioluminescence instead. Whichever way, the title probably reflects a totally unexpected result that is possibly too cross-disciplinary to be properly comprehended as an actual scientific advance by the authors. However, a proper study of the species could very well be an important paper. |
This one-column paper does not appear to have any headers, implying a longer, free-flowing format. |
This task I had to do anyway turned out to be hard enough for its own paper | There is a huge variety in the complexity and importance of subjects studied in scientific papers, and often some supposedly easy task will be sufficiently complicated as to merit its own paper. For example, a scientist may have discovered a better way of finding out if a substance is X or Y while studying something else.
The author may be glad to have been able to turn mundane 'housekeeping' activities, that don't normally do much to enhance academic reputations, into an actual opportunity to be cite-worthy. |
This paper is also one-column but it contains section headers. |
Hey, at least we showed that this method can produce results! That's not nothing, right? | One of the struggles of the scientific method is that many experiments will not produce the results scientists desired or expected. Negative or conflicting results of well-conducted research are as important as positive or dramatic ones, but are often ignored in favor of more novel findings. As a result, some journals are established specifically for negative results, reducing the bias towards only positive claims that may actually be outliers or anomalies.
In this case, the authors may otherwise have worked on their problem and been left with no citable proof of their efforts. The title perhaps reflects an attempt to present this as 'success' of a different kind, rather than a submission to such a null/negative-results platform. This may be similar to the above type of paper too. |
Two-column format, no figures. |
Check out this weird thing one of us saw while out for a walk | This paper may be imagined as an opportunistic publication. A department or team has seen itself low down on the local 'league table' for academic output. A brainstorming session for a way of rectifying this led to desperately seizing upon the first idle comment made (in lieu of any better sounding ideas) that can somehow be shoehorned into their respective subject area, and is now being presented similar to "this one weird thing" clickbait titles that almost always oversell their content.
This also works in the context of entomology. Insects have the most species of any class of animals by a wide margin, but due to their small size, they're not easily seen. As a result, new species are constantly being discovered in places as innocuous as someone's backyard. It also works in botany, especially floristics. Papers of first records of alien plants refer to weird things botanists saw on walks. Vagrant birds, unusual animal behaviour, and strange meteorological phenomena are other subcategories. |
Two-column layout. Includes several large figures, likely close-up photographs of the weird thing. There are no headers, as the paper may have little background or methodology, just observations. |
We are 500 scientists and here's what we've been up to for the last 10 years | Some papers summarize the work of big research teams, like those working on the Higgs Boson (list of authors starts at page 17 and goes to page 26 with foot notes about authors to page 29, and a dedication in the header would suggest that more than one other contributor died over the course of the research, which would be rather unusual for a smaller project) or LIGO. Since the discoveries which are made are a team effort, probably outlasting many of the individual tenures involved, the papers have many authors listed.
A credit for participation may not mean any particularly great contribution by each individual, but being left out (even for one summer's secondment, seven years before any results could be recorded) would be taken as a slight, and an opportunity missed to be 'citable' in the future. |
A huge portion of the page is taken up by the presumably 500 authors' names, above the main horizontal bar. |
Some thoughts on how everyone else is bad at research | Similar to the "my colleague is wrong" paper, but in this case applied to far greater swathes of the community by the author(s) of this (possibly unfocused) tract. Usually a "systematic review", the words 'some thoughts' might indicate a meta-approach with no original research - and possibly a passive-aggressive style of assessment. | No header sections, possibly because these particular thoughts are in the form of an essay or letter without an accompanying investigation. Formatting this article as a single column with large blocks of text could also be indicating a slightly unhinged rant by someone who—wrongly—perceives themselves as unjustly marginalized. |
We scanned some undergraduates | Initial research is often done at universities, so when human subjects are required, recruiting undergraduate students is a common, easy, and inexpensive way to gather enough people to conduct studies or experiments. This is extremely common in psychological or sociological studies, but can involve more medical (but non-invasive) 'scans', from simple eyeball-tracking to full-body MRI. This practice is often criticized, as it introduces a selection bias, which makes the results difficult to generalize to the entire population, as university students in a given country are not necessarily a representative sample of human beings as a whole. Nonetheless, easy accessibility makes these students a source of data for many academic papers. The low-key approach to the title (concentrating blandly upon the method with no references to results) may indicate that the results obtained are very trivial and no great developments were even made in implementation. Alternately, this is a truly ground-breaking paper obscured entirely by the lead author's over-narrow professional focus and avoidance of any hype. | Two-column format with figures. |
We've incrementally improved the estimate of this coefficient | Often scientific research, e.g. in cosmology or physics, will work with an assumed constant value that is known to be only an 'educated guess' of the actual definite value, or an inclusive range. However accurate/certain this is, further experimentation or observation may further narrow down the uncertainty involved to a statistically significant degree. An improvement to one of these constants also improves the accuracy of every single calculation that uses it.
Even if these improvements may seem trivial to those outside the discipline (e.g. narrowing down a seemingly esoteric value from 99.99% certainty to 99.995% certainty), they are probably understood as significant achievements by those aware of the effort needed to obtain such diminishing returns, and the authors are probably very excited to have done what they did. Another possible interpretation of this title is that it refers not to cosmological constants but to an exponent in algorithmic complexity, for example the 2014 paper that proved that the complexity of matrix multiplication is at most n^2.3728639 in place of the previous upper bound n^2.3729. |
(Only referenced in Title Text) |
Maybe all these categories are wrong | In some field that relies heavily upon classification (e.g. phylogenetic biology, or the Standard Model in physics) sometimes observations arise that cast doubt on the previously established ideas. It seems that this may have happened here, hopefully with a suggestion of how to reimagine the situation.
The article may have been written with with a sense of euphoria (the chance to present a paradigm shift in thinking, to rewrite the textbooks) or pessimism (it demonstrates only the failings in current thinking, without any obvious solution). Alternatively, it may be a reference to the categories of papers that this comic proposes. | |
We found a way to make student volunteers worse at tasks | Possibly a psychology experiment, and maybe not even the result expected. In general, the repetition of an activity will induce greater skill/capacity in a tested individual. By accident or design, the study group in this instance has induced the opposite correlation. (There are, however, some studies that explicitly look at how e.g. lack of sleep reduces productivity.)
Exactly what emotion the title reflects might depend upon whether the worsening was an intended result, or even how the team were able to refocus and seize upon the adverse outcomes. The tone of the title could even express surprise: some senior academics may already have a generally low opinion of student volunteers (as both naïve students and ones that have no better things to do than volunteer), but then managed to inadvertently engineer a situation in which particularly inept subjects sail well under the already very low bar of expectations for the study aims. |
Transcript
- [Heading:]
- Types of Scientific Paper
- [An array of 4 rows with 3 scientific papers each, is shown. We see the first page of each paper, but only its title is legible. Headings are shown as black lines, paragraphs of text are shown as several squiggly lines and figures are shown as empty white rectangles. Titles are as follows:]
- We put a camera somewhere new
- Hey, I found a trove of old records! They don't turn out to be particularly useful, but still, cool!
- My colleague is wrong and I can finally prove it
- The immune system is at it again
- We figured out how to make this exotic material, so email us if you need some
- What are fish even doing down there
- This task I had to do anyway turned out to be hard enough for its own paper
- Hey, at least we showed that this method can produce results! That's not nothing, right?
- Check out this weird thing one of us saw while out for a walk
- We are 500 scientists and here's what we've been up to for the last 10 years
- Some thoughts on how everyone else is bad at research
- We scanned some undergraduates
Trivia
- Originally, this comic's title text misspelled "volunteers" as "volunters".
- Another comic, 2012: Thorough Analysis, similarly categorizes or mocks research papers.
Derivatives
The comic inspired many derivatives, changing the paper titles to be more relevant to specific fields.
The hashtag #TypesOfScientificPapers on Twitter includes many of these.
There is a generator.
There is a moodboard compiling hundreds of them.
Some examples include:
Discussion
I've a feeling we could find actual papers that paraphrase down to those in the comic. Also, lol at the 500 scientists' "citation" section. 162.158.159.104 20:36, 28 April 2021 (UTC)
As we edit this we should probably pay attention to the content / layout of the article images: The number of lines beneath the title and layout of each "paper" he's drawn could be relevant to the joke. For example, the "500 scientists" presumably have a massive authors list, and the one on how "everyone else is doing it wrong" has a single author and a particularly "article-esque" layout.
172.68.132.45 21:04, 28 April 2021 (UTC)
- True. Do you think we should add another column describing the pictured paper to the explanation chart?
Reywas (talk) 21:06, 28 April 2021 (UTC)
For the "student volunteers" paper, many experiments involve adding hurdles for the participants to deal with. Like interrupting them, depriving them of sleep, adding distracting information, etc. It's not uncommon that these make them worse at the tasks. So this is just another research paper like that. Barmar (talk) 01:24, 29 April 2021 (UTC)
- Isn't the 'hurdles' style of paper WAY more common than the proposed psychological experiment? It's describing a simple tasks cognitive function test. They run those all the time to prove the effects of oxygen, caffeine, sleep, sugar, music, trauma, comfortable chairs, the color yellow, etc. I can't recall seeing ANY paper like the suggested psychology experiment to make people unlearn skills, let alone enough for that to be a whole category.
Nobody has pointed out that the "Maybe all these categories are wrong" title directly pertains to this very comic... John.Adriaan (talk) 02:17, 29 April 2021 (UTC)
- I think that’s the more likely ‘correct’ interpretation, honestly. Maybe we should add more detail now that it’s mentioned? Tague (talk) 12:56, 29 April 2021 (UTC)
- At the least the list of categories is incomplete. The broader question is whether there are concentrations of papers in some areas of feature space (a subjectively plausible conclusion - I could offer some examples from botany such as "we compared the performance of several cultivars of a crop species under specific conditions") or do papers fall into a relatively flat continuum in that space. 141.101.107.78 11:35, 30 April 2021 (UTC)
Should we add another colum to include the corresponding LaTeX template? Some of them seem like that could easily be found.
These are titles of papers, so shouldn't each word be capitalized? For example: "We Put a Camera Somewhere New". I realize that the original is in all caps, but that's because that's the usual format for comics . . . . 172.68.57.179 10:45, 29 April 2021 (UTC)
- Paper titles are usually not capitalized, contrary to journal names. You can see an example at the Higgs Boson paper cited in the comic description (there are, of course, others.) 141.101.105.134 11:32, 29 April 2021 (UTC)
Why does the explanation say "There are no headers" for the "We put a camera somewhere new" paper? I assume "headers" refers to "section headers", of which I see more in the camera-paper than in e.g. the immune system-paper (or the old records-paper). 141.101.96.28 11:25, 29 April 2021 (UTC)
- Probably referring to the lack of actual legible text for us to comment on the content of the paper. Tague (talk) 12:49, 29 April 2021 (UTC)
- I feel that the assumption made in a lot of the table that the text in each Paper is meant to be their literal title, is wrong. It strikes me more as an humorous explanation of "what sort of paper this is" for instance the first paper would indicate that a relevant category of scientific papers are about a camera being put someplace new and the data/photo's gathered from that, rather than an example of "clickbait". The actual papers in that category would presumably have an actual name relating to where the camera was actually put. 14:29, 29 April 2021 (UTC)~
- I interpret them as a mix of "over-generalized" headlines and less-than-literal summaries of that general sort of paper's content. Tague (talk) 14:41, 29 April 2021 (UTC)
- I read (e.g., for starters) "We put a camera somewhere new" both as very true to the emotional spirit and a paraphrasing of the true archetype membership being referenced - such as something like "Rat-mounted cameras for remote surveying of sewer pipes" (if that's not already been done, which I suspect it has!), etc. I suspect there's a few "one weird thing"-inspired titles out there, influenced by modern 'headline' links (with or without self-awareness), and know there's a whole history of "my colleague is wrong!" papers, even if not in exactly that wording, pushing the author's own biases in a self-important ranting style, or a rambling one that's an unstructured manifesto of 'thoughts' about all prior 'experts' on a pet issue. There's some deconstruction involved, but with easy reconstruction back to reality. 162.158.158.113 19:54, 29 April 2021 (UTC)
Derivatives
Should this (and any others, which I think likely exist or are about to) go in the main article? https://twitter.com/GreenBankObserv/status/1388148786707406854
"With apologies to Mr. Munroe, may we present: Types of Radio Astronomy Papers"
— JohnHawkinson (talk) 18:40, 30 April 2021 (UTC)
- These are good fun, but I don't think we are gonna put it up in image form in the main article. Maybe a section of external links will do. --162.158.165.235 00:56, 1 May 2021 (UTC)
- I don't think that links are a very effective way to show these. I feel like they need to be a gallery. Perhaps it should be a separate wiki page that is linked from the main article?
- Here are some more:
- — JohnHawkinson (talk) 13:45, 1 May 2021 (UTC)
The meme is now so popular there is an article in the atlantic about it; maybe that should be included: https://www.theatlantic.com/science/archive/2021/05/xkcd-science-paper-meme-nails-academic-publishing/618810/
- --162.158.89.113 14:28, 7 May 2021 (UTC)