Snail shells are almost identical in shape to cinnamon rolls. As I tweaked away at the vector art, I could almost smell those puffy confections that beckon across malls and airports with their buttery, come-hitherish aroma. I colored these snails in orange, which reminded me of the kind with orange-flavored icing. Now if only I could cause a bakery fragrance to emit from my blog, readers might associate my writing with delicious baked goods.
Post hoc, ergo propter hoc, like pareidolia, deals with illusory connection. The Latin translates to "after this, therefore because of this," a phrase that nicely sums up the logical snafu. Post hoc reasoning occurs when I perceive that because one event follows another, the first event must cause the second. My favorite example of this fallacy is the restaurant bathroom visit; the waiter always arrives just after I've excused myself to the ladies room. So can I command food to arrive by visiting the ladies room? More importantly, does this work with cinnamon rolls?
My restaurant fallacy is actually a more complex form of post hoc fallacy called "correlation without causation," which occurs when I assume a cause-effect relationship between events that appear related in several instances. Maybe my knee aches before it rains or the day goes badly if I have 1980s hair. Considering recent fashion, maybe my day goes badly if I don't have '80s hair.
The causal connection can seem especially probable when the two events happen together fairly regularly, and for this reason, we still hear about scientific findings based upon purely correlative evidence. The strongest studies, however, have control groups, blinds, and double blinds to rule out external causes. Often two strongly-correlated events are not actually causally linked; a third or indirect outside factor causes both of them. My dashing to the ladies' room doesn't cause the waiter to arrive, but the passage of time encourages me to consume beverages, and also increases the chances that the food will be ready.
I've noticed that correlative statements appear in ladies' health magazines. Magazines often include phrases such as: "a recent study found that women who eat a cup of blueberries per week are 30% less likely to to be diagnosed with heart disease, so mix blueberries into your cereal every day." It may well be that some ingredient in blueberries helps to prevent heart disease, but it may also be that women who pay for blueberries in January also eat produce more often and exercise regularly--the magazine doesn't tell us. Maybe the study behind the factoid used blinds and controls, but these details weren't snappy enough to make the magazine. The the omission suggests that a correlation should be proof enough to convince the average reader to spoon blueberries into her branflakes anyways.
Similarly, my illustration asks you to assume that giant snails are dropping from planes. The numbers in health food statistics inspired the parachutes in my drawing. I know that they're supporting something, but from where I'm sitting, I can't see what.
It would be nice, I think, if all false correlations appeared as ridiculous as the idea of monster snail paratroopers; then they would be easier to spot. Come to think of it, I find the idea of precipitating snails pretty fun, and it's easier to trust a post hoc statement when I want it to be true.
I wonder whether any health benefits correlate with eating orange icing right out of the Pillsbury tube.
Post hoc, ergo propter hoc, like pareidolia, deals with illusory connection. The Latin translates to "after this, therefore because of this," a phrase that nicely sums up the logical snafu. Post hoc reasoning occurs when I perceive that because one event follows another, the first event must cause the second. My favorite example of this fallacy is the restaurant bathroom visit; the waiter always arrives just after I've excused myself to the ladies room. So can I command food to arrive by visiting the ladies room? More importantly, does this work with cinnamon rolls?
My restaurant fallacy is actually a more complex form of post hoc fallacy called "correlation without causation," which occurs when I assume a cause-effect relationship between events that appear related in several instances. Maybe my knee aches before it rains or the day goes badly if I have 1980s hair. Considering recent fashion, maybe my day goes badly if I don't have '80s hair.
The causal connection can seem especially probable when the two events happen together fairly regularly, and for this reason, we still hear about scientific findings based upon purely correlative evidence. The strongest studies, however, have control groups, blinds, and double blinds to rule out external causes. Often two strongly-correlated events are not actually causally linked; a third or indirect outside factor causes both of them. My dashing to the ladies' room doesn't cause the waiter to arrive, but the passage of time encourages me to consume beverages, and also increases the chances that the food will be ready.
I've noticed that correlative statements appear in ladies' health magazines. Magazines often include phrases such as: "a recent study found that women who eat a cup of blueberries per week are 30% less likely to to be diagnosed with heart disease, so mix blueberries into your cereal every day." It may well be that some ingredient in blueberries helps to prevent heart disease, but it may also be that women who pay for blueberries in January also eat produce more often and exercise regularly--the magazine doesn't tell us. Maybe the study behind the factoid used blinds and controls, but these details weren't snappy enough to make the magazine. The the omission suggests that a correlation should be proof enough to convince the average reader to spoon blueberries into her branflakes anyways.
Similarly, my illustration asks you to assume that giant snails are dropping from planes. The numbers in health food statistics inspired the parachutes in my drawing. I know that they're supporting something, but from where I'm sitting, I can't see what.
It would be nice, I think, if all false correlations appeared as ridiculous as the idea of monster snail paratroopers; then they would be easier to spot. Come to think of it, I find the idea of precipitating snails pretty fun, and it's easier to trust a post hoc statement when I want it to be true.
I wonder whether any health benefits correlate with eating orange icing right out of the Pillsbury tube.
I love this image — the colours are really well chosen and it has just enough detail. Your work is quite original in that it looks hand-made but is very clean and controlled at the same time.
ReplyDelete