Detailed guides to painful problems, treatments & more

Actually, correlation kinda does imply causation

 •  • by Paul Ingraham
Get posts in your inbox:
Weekly nuggets of pain science news and insight, usually 100-300 words, with the occasional longer post. The blog is the “director’s commentary” on the core content of PainScience.com: a library of major articles and books about common painful problems and popular treatments. See the blog archives or updates for the whole site.

There’s a famous rule: “correlation does not imply causation.” Unfortunately, it’s wrong, *as normally stated.

It’s missing an important word. It really should be: “correlation does not necessarily imply causation.” Because correlation actually does “imply” causation, and many (if not most) events that occur in sequence that appear to be causally related are in fact causally related. Human brains are dazzlingly good at correctly inferring causal relationships from observed correlations: clapping makes noise, braking stops cars, hot coals burn fingers. This mental super power served us well as we grew up as a species.

The problem is that we’re so good at it, and it’s such an essential mental skill, that we tend to overdo it and perceive causation in all kinds of situations where causality detection is much harder … like evaluating the results of medical treatments.

Complex causal relationships are as tricky to infer from simple observations as simple ones are easy. And we are just pathetically bad at figuring out exactly how events are causally connected — “mechanism of action.” Because of all the unknown variables. What’s really going on in a casual relationship almost always turns out to be different and waaaaay more complicated than we thought.

Nature: defying “common sense” since the dawn of intelligence.

But humans are causality bloodhounds: we smell it everwhere, even when we don’t understand what’s really going on (which we usually don’t). For instance, if someone who’s been limping and grimacing for days walks out of a massage appointment with a grin and a light step, then, yeah, massage probably did cause that result, one way or another.

[Update: see several important clarifications in the next post. Also, this has now been integrated into the bizarre, deep, dorky article The Power of Barking: Correlation, causation, and how we decide what treatments work.]

PainSci Member Login » Submit your email to unlock member content. If you can’t remember/access your registration email, please contact me. ~ Paul Ingraham, PainSci Publisher