223 | Tania Lombrozo on What Explanations Are
What this covers
Sean Carroll hosts psychologist Tania Lombrozo in conversation about why humans seek explanations and what makes one satisfying. Rather than treating explanation as a fixed category or a mirror of objective truth, Lombrozo argues that explanation is a functional practice shaped by our need to learn and predict. The discussion weaves together philosophy—what an explanation is, whether it must bottom out—with experimental psychology: what people actually prefer, what makes them curious, and where their judgments diverge from scientific ideals. The conversation tracks explanations across domains, from physics to psychology to religion, revealing how the same cognitive machinery produces quite different demands depending on whether we are reasoning scientifically or spiritually.
The heart of the episode concerns Simplicity Preference and mechanism. Lombrozo shows that people prefer simpler explanations not because nature is simple, but because simplicity is a methodological strategy: it prods us to look for generalizable structure that transfers to new cases. The Mechanism-Generalization Link runs through much of the empirical work—people weight mechanistic information heavily precisely because it supports prediction beyond the observed case. But there are frictions. Experiments reveal that people often reject valid causal claims made via statistics alone, that they resist making untested predictions even when probability demands it, and that they prefer explanations matching their prior beliefs. Epistemic vs Non-Epistemic Functions emerge sharply in the religious-versus-scientific divide: people rate scientific explanations on evidence and logical force while rating religious ones on comfort, moral guidance, and social cohesion—and this split holds across all levels of belief strength. The episode also addresses learning: the Self-Explanation Effect, where merely rearranging information through self-explanation improves pattern recognition, suggests that explanation is as much about refining our grasp of what we already hold as about acquiring new facts.
Lombrozo argues that the human drive to explain is not primarily about accurately mirroring a simple reality but is a functional, learning-oriented practice: seeking explanations (especially simple, mechanistic ones) makes us look for generalizable structure that supports future prediction, while also serving non-epistemic social and emotional goals.
- Preferences for simpler explanations are rational as a methodological strategy that drives discovery of real structure, not because nature is simple.
- Mechanism information is valued because it supports generalization to new cases, unlike bare correlation.
- Explanations serve non-epistemic functions (comfort, morality, social regulation), especially evident in religious vs scientific explanations.
The human desire for satisfying explanations is psychologically crucial but lacks foundational justification.
- The desire to be satisfied with an explanation is crucially important psychologically but hard to justify on foundational grounds, because no one ever promised that the correct scientific explanation would be satisfying—the demand for satisfaction can persist even when the best available explanation is correct.
“no one ever promised you you would be satisfied with the correct scientific explanation.”
The question why there is something rather than nothing may not apply because the world lacks necessary explanation.
- The question 'why is there something rather than nothing' may be a grammatical construction that looks well-formed but does not actually apply, because the world is not the kind of thing that necessarily has an explanation for why it exists.
“that's the kind of grammatical construction that looks like it's perfectly well-formed, but it doesn't actually apply. The world is not the kind of thing that necessarily has an explanation for why it exists.”
Causal understanding develops through learning how actions produce observable reactions in the world.
- Babies spend their time making causal maps of the world, learning relationships of the form 'if I poke this, it reacts in this other way.'
“He claimed that babies spend their time making causal maps of the world, saying, "If I poke this, it reacts in this other way."”