credence
A subjective degree of belief, often treated as a probability; rational credence is the degree of belief one ought to have given evidence.
As defined in ms_215
Where it comes up · 34
- 262 | Eric Schwitzgebel on the Weirdness of the World9 claims
- AMA | November 20218 claims
- AMA | August 20214 claims
- AMA | June 20214 claims
- AMA | March 20224 claims
- AMA | September 20224 claims
- 253 | David Deutsch on Science, Complexity, and Explanation3 claims
- AMA | February 20223 claims
and 26 more
Most important claims · 10 of 80Search all claims for “credence” →
- Theoretical physicists propose a theory not because they think it is true but because they think it is possibly true and worth investigating, assigning it some credence; Carroll's most-cited paper proposes modified gravity to explain cosmic acceleration, yet he thinks the cosmological constant is the more likely explanation, illustrating that proposing a model and believing it are distinct.AMA | November 2021
- One should never assign exactly zero credence to any logically consistent possibility, because in Bayes's formula multiplying a zero prior by any finite likelihood always yields zero, so no future evidence could ever change your mind—every possibility should retain some small nonzero credence to remain open to updating.AMA | November 2021
- Scientific thinking is fallibilistic and ultimately empirical, but theoretical insight alone—without any new experiment—can dramatically boost or squelch credence in a theory, as demonstrated when 't Hooft proved spontaneously broken gauge theories are renormalizable in the early 1970s, causing interest in Weinberg's previously ignored 1967 electroweak model to surge from near-zero to becoming the most-cited particle physics paper.AMA | November 2021
- The simulation argument's probabilistic step misuses anthropic reasoning by assuming we are typical observers; this only seems humble but is actually maximally arrogant because you do not know what a typical observer looks like, so assuming typical observers are like you infers facts about all observers from your parochial experience — an unwarranted move.AMA | March 2022
- Cosmological scenarios dominated by random fluctuations should be assigned very low credence not because we are 'ruled out' as Boltzmann brains by data, but because such scenarios imply that essentially all of one's beliefs about the external world—including the reasoning that led to the scenario itself—are randomly fluctuated and false, making the theory self-undermining; yet one should not assign exactly zero credence, since zero credence is impervious to all future evidence.AMA | November 2021
- Black hole evaporation is widely accepted not because it has been observed but because it is a prediction of theories (classical general relativity and quantum field theory) that are well confirmed by experiment, and the correct epistemic attitude is to place high credence on the predictions of good theories applied to new situations until you have evidence otherwise.AMA | February 2021
- In a scenario where you are told a transporter is random but you always appear on pad 1, the correct approach is Bayesian: you assign prior credences to hypotheses (random, always pad 1, etc.) and update via Bayes' formula with each observation; no matter how high your prior in randomness, enough pad-1 observations shift you toward 'always pad 1,' but your credence should never reach 0 or 1 since future data could always change your mind.AMA | May 2021
- The word 'multiverse' refers to several distinct ideas deserving different levels of belief: the quantum mechanical multiverse (taken very seriously), the cosmological multiverse of distant regions with different physics (indifferent), and philosophical versions like David Lewis's modal realism in which every possible world is equally real.193 | Daniels on Everything, Everywhere, All at Once
- When evaluating a surprising claim like rats worldwide learning tasks faster, the hypothesis that would require changing the laws of physics (e.g., rat telepathy or collective consciousness) should be assigned by far the lowest credence, because our knowledge of physics is so secure that even strained mundane explanations (statistical fluctuation, measurement error, exaggeration in retelling) remain vastly more plausible.AMA | November 2021
- Priors are not developed with no data whatsoever; our biology, psychology, education, upbringing, and evolutionary history all count as data feeding intuitive priors. Different people starting with different priors is a feature not a bug, and rationality lies in the reasoning process from those priors; with enough data good Bayesians converge on the same answers.AMA | October 2022