The Law of Parsimony, or, Novacula Occami

• Aristotle, 300s BC: "We may assume the superiority ceteris paribus [other things being equal] of the demonstration which derives from fewer postulates or hypotheses."
• Ptolemy, 100s AD: "We consider it a good principle to explain the phenomena by the simplest hypothesis possible."
• William of Ockham, 1285-1347/49: “Pluralitas non est ponenda sine necessitate” (Plurality should not be posited without necessity.)
• John Punch, 1639: “Non sunt multiplicanda entia sine necessitate.” (Entities must not be multiplied beyond necessity.)

What is the Occam’s Razor Bias?

It is a type of an inductive bias, which I personally call non sequitur fallacies.

Karl Popper and Simplicity

Karl Popper in Simplicity, Chapter 7 of his his Logic of Scientific Discovery quotes Schlick on the nebulous definition of simplicity:

Simplicity is... a concept indicative of preferences which are partly practical, partly aesthetic in character... Even if we are unable to explain what is really meant "simplicity" here, we must yet recognize the fact that any scientist who has succeeded in representing a series of observations by means of a very simple formula (e.g., by a linear, quadratic, or exponential function) is immediately convinced that he has discovered a law.

I find the last example particularly humorous because I have generated formulas based on my own observations and have been inordinately proud of myself!

Popper goes on to say that simplicity, then, is viewed as the idea of what provides law-like regularity:

Believers in inductive logic assume that we arrive at natural laws by generalization from particular observations. (emphasis mine)

I'm sure you can see the inherent flaw in the inductive bias apparent here.

Neurobiology of Occam’s Razor

Neuroscience is still such an infant branch of science that I couldn’t find any studies done on this specific bias. However, I found a good imaging study on deductive v. inductive reasoning.

The seats of reason? by Goel et al., 1997

Goel et al. succinctly define and give examples for the two types of reason philosophers have delineated: DEDUCTIVE v. INDUCTIVE:

Valid deductive arguments involve the claim that their premises provide absolute grounds for accepting the conclusion. For example: (A) All men are mortal; Socrates is a man; therefore, Socrates is mortal. Validity is a function of the logical structure as opposed to sentence content. Arguments where the premises provide only limited grounds for accepting the conclusion are broadly called inductive arguments. For example: (B) Socrates is a cat; Socrates has 32 teeth; therefore, all cats have 32 teeth. (C) Socrates is a cat; Socrates has a broken tooth; therefore, all cats have a broken tooth.

DEDUCTIVE v. INDUCTIVE reason examples

Goel et al., studying 10 university-educated, right-handed males, demonstrated that inductive reasoning (under which the Occam’s Razor Bias falls) occurs in the

The deduction condition (Fig. 1, deduction–baseline) resulted in activation of the left inferior frontal gyrus (Brodmann areas 45, 47) and a region of the left superior occipital gyrus (Brodmann area 19). The induction condition (Fig. 1, induction–baseline) resulted in activation of a large area comprised of the left medial frontal gyrus, the left cingulate gyrus, and the left superior frontal gyrus (Brodmann areas 8, 9, 24, 32)

Figure 1 of Goel et al., 1997, demonstrating which areas of the brain inductive and deductive reasoning activate.

Further studies, then, need to be performed to see if more specific areas, even neurons, can be isolated for specific biases that fall under the inductive umbrella.

Figure 1 of Goel et al., 1997

---

Conclusions about Occam’s Razor

Why do we default to inductive reasoning when deduction is so much “better” (by better I mean more accurate, usually)? Well, Kahneman would probably agree, because it’s easier. Our System 1 is inherently lazy, and it controls almost everything.
Goel et al. (1997) concluded:

By every measure (reaction times, performance scores, and subject feedback) deduction was the more complex task.

Humans want ease and simplicity, and induction gives us what we crave.

---

Previously in the Cognitive Biases Series:

Background and Outline
Less is Better Effect
Next up: Conjunction Fallacy

---

References:

1. Buster Benson’s Cognitive Bias Cheat Sheet. Shave those biases!
2. Goel et al. 1997. The seats of reason? Don’t shave the neurons!
3. Karl Popper: The Logic of Scientific Discovery PDF

---

All images are original unless credited here
Image Credits:

1. John Manoogian III and Buster Benson’s Cognitive Bias Codex. Shave those biases!
2. Deductive v. Inductive Reason Examples: Goel et al. 1997
3. Penguins and TVs
4. Figure 1 of Goel el al. 1997