1588: Hardware Reductionism

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
Hardware Reductionism
My MRI research shows a clear correlation between the size of the parietal lobe--the part of the brain that handles spatial reasoning--and enjoyment of 3D Doritos®.
Title text: My MRI research shows a clear correlation between the size of the parietal lobe--the part of the brain that handles spatial reasoning--and enjoyment of 3D Doritos®.

Explanation[edit]

Reductionism is the belief that things can be explained by their smaller parts. It can be abused when complex phenomena with multiple causes are attributed to a single, simple cause.

Neurological reductionism is the attempt to explain people's behavior and personality by physical features of their brain. With advances in neuroscience, and especially in brain imaging, there's a fad to claim that brain types determine what the mind is. Examples of this kind of bad reductionism would be:

  • Male brains have more grey matter than females. Therefore, males are smarter. For an example of criticism of this kind of reasoning, see Male and female brains: the REAL differences (4 December 2013).
  • Brains of gay males are slightly more symmetrical, as are female brains, when compared to straight males. Therefore, gay men are fated to be more effeminate. See Gay Men, Straight Women Have Similar Brains (16 June 2008).
  • The left side of the brain is associated with logical thinking, and the right, with visual and artistic thinking. Therefore, people divide into "left-brain" or "right-brain" types, depending on how good they are at using each side. See lateralization of brain function on Wikipedia.

There are several problems with this kind of reasoning. First, most studies identify correlation, not causation (see correlation does not imply causation). Brains are plastic; they can be shaped by experience. For example, if, in a given society, the females are taught to mind their appearance, and the males are taught that aesthetic considerations are unmanly, then of course the female brains will end up with more developed aesthetic centers. In other words, behavior and capabilities aren't always determined by the brain. Sometimes it's the behavior that shapes the brain; sometimes a third factor (e.g., malnutrition) shapes both.

Second, even when the brain is actually a cause of the behavior, it's far from the only piece in the puzzle. Many studies on brain differences are correlation studies, often about very small effect sizes. Unfortunately, the popular science media tends to gloss over the statistical concept of "effect size". For example, imagine a study that says that males' brains are 0.1% more likely than females' brains to exhibit attention deficit hyperactivity disorder (ADHD). Journalists are prone to report it simply as Study Shows that Males Have More ADHD, and this becomes a conversation sound-bite that neglects other factors, like genetics or pregnancy smoking. See also 882: Significant, which does not discuss effect size, but does raise other objections to writing soundbites based on a single study.

Another kind of excessive neuronal reductionism is the overemphasis on brain modules ("scientists identify brain area responsible for religious faith", and the like). Though it's true that the brain has specialized areas, it's also true that the processing is very complex, messy, and distributed all over. Some varieties of brain damage can often be overcome by learning to use undamaged areas of the brain.

The comic illustrates the problem by analogy to some better-understood general-purpose computing hardware: the CPU in a smartphone. Cueball and Megan have used their smartphones to take pictures of the same event: a triathlon, that is, an athletic competition comprising three modalities (e.g., swimming, cycling, and running). Cueball wonders why is it that Megan's photos are more popular, and Megan gives a reductionist explanation: She tells that her phone is quad-core (four cores) whereas Cueball's phone only has two cores (here she even throws in the typical sentence "research shows that" to make her claim sound more valid). A core is a part of a CPU that executes programming instructions, often described metaphorically as the "brain" of a computer or smartphone. Megan thinks that this means Cueball's smartphone can only capture two events at the same time; to anyone who understands how computers work this conclusion is absurd. She misunderstands how the specialized modules work (for that matter, how a digital camera works at all) and fails to realize that the number of cores is unrelated to how many events can be captured. Her claim is like saying that male brains are better at spatial reasoning, and therefore males are better triathlon photographers, or that females are better at multitasking, and therefore females are better triathlon photographers.

A CPU with more cores could process pictures faster, speeding up facial recognition or color filters. So it's true that Megan's CPU makes it slightly easier for her to take pictures. However, this has, at best, an extremely small effect on the number of "likes". There's a lot more going on with photography than the CPU of the phone: Megan's photographing skills, her luck in capturing interesting scenes, the number of online friends she has, etc.

So Megan misunderstands many things: the modularity of CPUs, the small effect of the CPU on the quality of her photography, and the actual causes of her success, much like people who reduce ability to structural features of the brain.

The title text is mocking reductionist explanations based on Randall's MRI (magnetic resonance imaging) research. One of the most famous (and disputed) claims about gendered brains is that women's brains are (slightly) worse at spatial reasoning. Doritos is a popular junk-food brand of tortilla chips that are typically so flat that they can be called a 2D snack. In the 1990s Frito-Lay (PepsiCo) introduced a special 3D version, the 3D Doritos. (These bloated snacks took up more surface area in one's mouth, and had a hollow center filled with cheese-flavored air). So the title text associates a larger spatial reasoning brain area with enjoyment of this three-dimensional variation of the popular junk-food snack; the conclusion could be that men like these 3D snacks more than women because of their better spatial reasoning, although there could obviously be several other reasons for such gender specific choice of junk-food. 3D Doritos were discontinued, but reintroduced in 2015, the year of this comic's release.

Transcript[edit]

[Cueball holds his smartphone looking at it while talking to Megan who is holding her smartphone in her hand.]
Cueball: Your photos from the triathlon got so many more likes than mine.
Megan: Yeah - My phone is quad-core. Research shows that iPhones like yours have just two cores, so they have a hard time capturing scenes with three different events in them.
[Caption below the frame:]
If we talked about phone hardware the way we talk about brain hardware

Trivia[edit]

  • When this comic was originally published there was a misspelling: triatholon instead of triathlon.


comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!

Discussion

Is "TRIATHOLON" just a typo, or does it have a special comic value? 141.101.106.77 (talk) (please sign your comments with ~~~~)

At this moment of creation it exists in a limbo in which it is both a typo and a joke, but now that it is has been released for viewers to take measures, the function will soon collapse into just one of the possible states. 162.158.34.197 13:03, 9 October 2015 (UTC)
It's clearly a spelling mistake (not a typo). See http://www.thefreedictionary.com/Triathalon, http://www.thefreedictionary.com/athelete. 173.245.50.149 13:19, 9 October 2015 (UTC)
I'm going with the theory it's a joke around the philosophical Holon. Elvenivle (talk) 17:53, 9 October 2015 (UTC)

I don't think this is just parodying left-brain right-brain myths. Rather, it's parodying neural reductionism of all kinds—the currently widespread myth that our selves are determined genetically by brain structure alone, minimizing the role of culture and the way experience rewires the brain. In particular, the part about "phones like yours" makes me think of "women are from venus"–style myths (where, say, a slight correlation is found between gender and size of spacial processing module, etc, and pop-sci media reports it as WOMEN ARE INHERENTLY BAD AT SPATIAL REASONING). Leoboiko (talk) 13:25, 9 October 2015 (UTC)

Seems plausible. Care to add? 108.162.216.104 13:39, 9 October 2015 (UTC

This is a real good, thorough and solid explanation. I'm impressed, well done! Flekkie (talk) 01:35, 12 October 2015 (UTC)

Or it could be a riff on the current "Thinking Fast and Thinking Slow" model, which considers the brain as having two distinct (Type 1 and Type 2) types of thinking, often described as two separate actors in the brain, in spite of the fact that they probably overlap a lot in the sections of the brain used. Blackbearnh (talk) 14:16, 9 October 2015 (UTC)

How about the typo of "coment" in the comment about the typo in the comic... 173.245.54.91 (talk) (please sign your comments with ~~~~)

"Coment" may be a typo (error when typing) or a misspelling (when you don't know the correct spelling). "Triatholon" can only be a misspelling, because the comic is not typed. 173.245.50.149 18:12, 9 October 2015 (UTC)
“ "Triatholon" can only be a misspelling, because the comic is not typed.”
You can make a mistake when handwriting. It's not a typo, but neither is it a misspelling.
173.245.49.78 21:55, 9 October 2015 (UTC)
I'm 100% sure Randall doesn't know the spelling of "triathlon". You simply don't add an extra letter when writing by hand "by mistake". BTW, it is corrected right now. 173.245.50.149 20:52, 14 October 2015 (UTC)

Might the two cores' difficulty handling three events be meant as a parallel to the functional brain study result showing humans multitask only two things, with one frontal lobe handling each task (and the introduction of a third task results in timesharing rather than parallel processing of all three)? 173.245.50.139 22:15, 9 October 2015 (UTC)

It could, but the three events in a triathlon don't involve multitasking - the events run in serial, not in parallel. I'm thinking about noting this in the text somewhere, but haven't though what to say about it yet - it seems peripheral. 108.162.250.161 00:25, 12 October 2015 (UTC)
In that case, you should put it in an extremely offset side-bar. Like, you know, in everyone's peripheral vision.... (...I'll get my coat.) 141.101.98.188 08:43, 13 October 2015 (UTC)

I wouldn't be surprised if someone actually believed the phone explanation. It sounds more plausible that several marketing claims. -- Hkmaly (talk) 18:29, 10 October 2015 (UTC)

How about the idea that people overgeneralize the function of objects (cores, regions of the brain) based on their names. If a dual-core phone can only process 2 things at a time then it can't process a triathlon as well as a quad-core phone. If a brain region is dedicated to 3D objects then there must be some connection to enjoying (processing) 3D doritos.173.245.54.14 13:10, 11 November 2015 (UTC)

Shouldn't the image here be changed to reflect the current version on the website, with the corrected spelling? 108.162.246.203 10:25, 18 April 2017 (UTC)

Yes. So I did it. Sorry you had to wait five years .... Nitpicking (talk) 20:17, 14 January 2022 (UTC)

I just realized that the "three events" thing is a pun based off the three events (biking, swimming, running) of a triathlon. Amazing stuff! 162.158.255.142 23:54, 10 December 2017 (UTC)