If there is no plausible mechanism by which a scientific hypothesis could be true, then it’s almost certainly false.

But if there is a plausible mechanism for a hypothesis, then that only provides weak evidence that it’s true.

An example of the former:

Astrology teaches that the positions of planets in the sky when you’re born can affect your life trajectory. If that were true, it would contradict well-established facts in physics and astronomy. Nobody has ever observed a physical mechanism by which astrology could be true.

An example of the latter:

A 2023 study found an association between autism and diet soda consumption during pregnancy. The authors’ proposed mechanism is that aspartame (an artificial sweetener found in diet soda) metabolizes into aspartic acid, which has been shown to cause neurological problems in mice. Nonetheless, even though there is a proposed mechanism, I don’t really care and I’m pretty sure diet soda doesn’t cause autism. (For a more thorough take on the diet soda <> autism thing, I will refer you to Grug, who is much smarter than me.)

Why?

A lack of mechanism strongly rules out a hypothesis. If astrology were true, that would overturn some extremely well-established findings in physics. How could astrology possibly be true, given what we know about the laws of gravity?

Perhaps scientists have overlooked something. Perhaps the planets affect humans not via gravity but via some fifth as-yet-discovered fundamental force. But if astrologers can detect the fifth force, why haven’t physicists noticed it with all their careful experimentation?

On the other hand, the existence of a mechanism doesn’t count for much. I often see this in biology, where someone proposes a contrarian hypothesis with a possible biological mechanism but no supporting evidence from randomized experiments. I don’t take that sort of evidence very seriously. Biology is complicated, and chemicals have all sorts of effects on bodies, and it’s very hard to predict whether those effects are net good or bad just by looking at mechanisms.

For example, did you know that exercise increases inflammation? And inflammation is bad for you? And yet, exercise is good for you, because the acute inflammation caused by exercise is strongly outweighed by the long-term beneficial effects.

However, when a hypothesis has supporting evidence from experiments but a lack of plausible mechanism, I disbelieve the research. Experiments have demonstrated that people have psychic abilities. But I’m quite confident that people don’t have psychic abilities because there is no mechanism by which that could be true.

In the hierarchy of evidence, experiment beats mechanism, but lack of mechanism beats experiment.

This asymmetry is consistent with the law of Conservation of Expected Evidence. There are many plausible mechanisms out there in the world. A hypothesis must have a mechanism for it to be true, but the existence of a mechanism does not come anywhere close to proving a hypothesis correct.

Some more examples

Here are some more hypotheses that are strongly ruled out by a lack of plausible mechanism:

  • Some houses are haunted by ghosts.
  • Dowsing rods can detect underground water.
  • Fortune-tellers can predict the future.
  • Homeopathy can cure diseases.

Some hypotheses with plausible mechanisms that I nonetheless believe are false:

  • Seed oils are bad for you because they contain linoleic acid, which causes inflammation. This mechanism is true (as far as I know), but experiments comparing unsaturated fats (mainly seed oils) to saturated fats find that people who eat more of the former end up healthier; see Hooper et al. (2020)1 and WHO (2016)2. Experimental evidence indicates that seed oils have overall positive health effects.
  • Eating excess protein causes osteoporosis. The proposed mechanism is that proteins increase blood acidity which causes the body to extract calcium from bones to balance out this acidity. And indeed, people on high-protein diets excrete more calcium in their urine. But randomized controlled trials have found that adding protein to the diet reduces the risk of bone fracture (Koutsofta et al. (2018)3).4
    • Relatedly, you may hear some people say you should eat more alkaline foods to fix your body’s pH balance. It would indeed be bad if your body’s pH became too low, but the empirical evidence shows that dietary pH does not affect your body’s pH in that way (see Wikipedia).5
  • Sugar causes hyperactivity in children because it provides a short-term burst of energy. This mechanism is intuitive even if you don’t know much biology. But it’s not true—RCTs have consistently found no connection between hyperactivity and sugar consumption (Vreeman & Carroll (2008)6).
  • Eating cholesterol raises your blood cholesterol. The mechanism in this case is obvious: you eat food that contains cholesterol, and the cholesterol goes into your body. But your body regulates its own cholesterol production, and your blood cholesterol levels don’t have much to do with how much cholesterol you eat.7

Posted on Oct 08, 2025

Notes

  1. Hooper, L., Martin, N., Jimoh, O. F., Kirk, C., Foster, E., & Abdelhamid, A. S. (2020). Reduction in saturated fat intake for cardiovascular disease. doi: 10.1002/14651858.cd011737.pub3 ↩

  2. Mensink, R. P., & World Health Organization (2016). Effects of saturated fatty acids on serum lipids and lipoproteins: a systematic review and regression analysis. ↩

  3. Koutsofta, I., Mamais, I., & Chrysostomou, S. (2018). The effect of protein diets in postmenopausal women with osteoporosis: Systematic review of randomized controlled trials. ↩

  4. I heard about this research on the Iron Culture podcast, in which they went on to complain about how people care too much about mechanisms and ignore experimental evidence. It got me thinking about an apparent contradiction in my beliefs where I care a lot about mechanisms for ruling out astrology and ESP, but I don’t really care about mechanisms in nutrition or exercise science. After thinking about it, I realized that my position is perfectly sensible—it’s about using mechanisms to rule hypotheses out vs. in—and that’s how I came up with the idea to write this post. ↩

  5. I want to be careful not to say that an alkaline diet is unhealthy. Alkaline foods do tend to be particularly healthy—they’re mostly fruits and vegetables—but that’s coincidental, not because they’re alkaline per se. ↩

  6. Vreeman, R. C., & Carroll, A. E. (2008). Festive medical myths. ↩

  7. https://nutritionsource.hsph.harvard.edu/what-should-you-eat/fats-and-cholesterol/cholesterol/ ↩