Suppose I feel that event A is more plausible than event B. How can I test, verify, or falsify this?

For example, suppose I have a belief that my partner is cheating on me. Suppose I have another belief that I have a water bottle in my fridge.

Now of course, I can simply find out in time whether my partner is cheating or open the fridge to see if my water bottle is there. This will falsify or confirm each belief.

But how do I ensure that my credences prior to this verification are accurate? As in, how do I know if it is rational for me to have a higher credence in my partner cheating on me vs. there being a water bottle in my fridge.

Now I understand that there are many factors that might go in this decision, but is it possible for there to be technically better evidence for believing in A over B, but have your conscious experience still have a credence of belief that is higher in B over A? If so, how does one find out?

  • 3
    One doesn't. Credences are used to decide how to act. One tests not individual credences but their system of assigning them and acting on them as a whole, and it passes if their actions meet their expectations reasonably well. Making sure that credences are coherent and updating them based on incoming information helps.
    – Conifold
    Mar 6 at 1:41
  • This is the thing where my college prof said, "You can't". So, believe what you wish.
    – Scott Rowe
    Mar 6 at 1:44
  • 3
    You don't do anything a priori. Now - maybe you have other information that's on the periphery of your consciousness, or that has become unconscious and that may make you feel one way or another. For example, maybe you usually have water in your fridge - you are not consciously using that in a rational reasoning, but it's there and working its way to making you believe there is probably water in your fridge rather than not. Sometimes we are not fully aware of the prior information we have assimilated.
    – Frank
    Mar 6 at 2:02
  • So maybe you could try to scan what you have seen before, or see if there are clues in your past experience that could sway the odds one way or another. You have to play detective and pay attention to less obvious hints.
    – Frank
    Mar 6 at 2:12
  • We could quite obviously judge which among two stones is heavier, but is that necessary or even sufficient? Mar 6 at 8:02

4 Answers 4


A common test is "what odds would you bet on it?". This, roughly speaking, is part of the Bayesian Epistemology package; you're supposed to follow probability norms because if you bet in non-probabilistic ways, you'd be extremely exploitable.


But how do I ensure that my credences prior to this verification are accurate? As in, how do I know if it is rational for me to have a higher credence in my partner cheating on me vs. there being a water bottle in my fridge.

Let's suss out an important distinction between empirical methods and rational methods. Hiring a PI to follow your spouse and checking the contents of the fridge are empirical methods because they rely on empirical evidence:

Empirical evidence for a proposition is evidence, i.e. what supports or counters this proposition, that is constituted by or accessible to sense experience or experimental procedure. Empirical evidence is of central importance to the sciences and plays a role in various other fields, like epistemology and law.

Fundamentally, you know that your beliefs can be evaluated to a truth-condition by using your senses, relying on facts, third-party testimony, etc. This might be considered a posteriori. On the other hand, what you are asking, when you use the phrase "rational for me" is not a question of the empirical, the sensory, or a posteriori, so much as the using reason before testing your hypothesis or determining the truth condition as a function of the state of affairs in the world. One can use reason, but do so poorly and so if one does so well, it might be considered being rational. Reason and rationality are terms that have a number of sense, so it's important to see that there is the use of reason, but there is also the reasonable use of reason. Rationality tends to capture the essence of the latter. To test your credences before determing their truth-condition, consider whether or not they are rational. An example might help to make this distinction concrete.

I have encountered what I call fallacy mongers. People in arguments who know the name of fallacies, and attack your position by declaring every other claim fallacious. For instance, one can reasonable cite Einstein as an authority in physics and matters of physics given that he helped to craft a historically and philosophically important theory in regards to space-time in distinction to Aristotle or Newton before him. Yet, a fallacy monger might accuse you of appeal to authority to undermine your point not understanding that there are criteria that allow you to differentiate between appropriate invocation of authority and inappropriate conditions. The fallacy occurs ONLY in the latter condition. No matter to the fallacy monger! Any mention of authority is irrational (more so when it undermines their point when they claim they have found an exception to the conservation of momentum in their garage which is a true anecdote of mine). Thus, they are being reasonable using the language of reason, but doing it poorly or unreasonably.

  1. The first thing to learn about when being rational is what philosophers have to say about rationality itself. Consider reading the Oxford Handbook on Rationality (GB). It might not instantly make you more rational, but at least you'll have some ideas about what epistemologists and professional philosophers believe and articulate rationality to be. People tend to believe they are more rational then they actual are.
  2. Read up on fallacies, cognitive biases, and the philosophy of logic. There are books on all of these topics. Again, it is part of the hubris of the human mind to overestimate it's capacities in these topics once a little information is acquired. Dunning and Kruger are famous for showing that there's a difference between having ability and having estimates of ability, and that without lots of talent and/or hard work, the latter tends to be optimistic. There are people on this site who have obviously spent their lives mastering logic, and there are people on this site who argue with them obviously armed with a flimsy understanding of the basics. Don't be the latter; exercise humility.
  3. Read philosophy. Start off with encyclopedia entries and primary sources. Have you read Kant's Introduction to Logic? Descartes Meditations? Good. Now read professional philosophers interpretations of them in encyclopedia entries and articles and books. Do you understand the historical movements of empiricism and rationalism? Descartes, Hume, and those who follow often have profound insights into reasoning.
  4. Read psychology. Cognitive science, behavioral economics, and psychological linguistics have profound ideas, hypothesis, and ontologies. Make use of what science says. This is the notion of a naturalistic epistemology. Many "philosophers" like to believe their brain is an logical engine on par with a supercomputer, and that their introspection is comprehensive, rational self-knowledge. Nonsense. Our brains and bodies are highly imperfect, and our knowledge largely is fallible (IEP) and intuitive (SEP).

And prepare for the long-haul. Being rational and reasonable, particularly in the face of the tempestuous nature of life might call for epoche and ataraxia, and other approaches to life such as acknowledging that work is opportunity. In life, you'll have a window of opportunity from the time your brain fully develops in your early twenties to a possible cognitive decline in old age. Being rational isn't a course or book, it's a lifestyle choice, one that most philosophers, professional or otherwise aspire to.

  • One can use mathematical formalisms, but whether or not it is rational to use such and such a formalism is a question of rationality, and that is more of a qualitative question.
    – J D
    Apr 12 at 17:53

It is usually said that credence should be proportionate to the evidence. I've never understood what this means. But it could mean that a test of your credence would be statistical information about the probability of your belief being true given the evidence. Sometimes this may be available. It couldn't give you certainty, but it could help you deciding what to do. Apart from that, I can't think of anything apart from collecting more evidence about your partner and/or the contents of your fridge. But that would alter your credence in one direction or the other; it would not confirm that your credence is proportionate to the probability of your belief being true.


Revisiting this, I notice that I didn't explicitly answer the question in the last paragraph. It is perfectly possible

"for there to be technically better evidence for believing in A > over B, but have your conscious experience still have a > credence of belief that is higher in B over A?"

There are two things you can do to find out. You can try to acquire further evidence, as everybody is suggesting. This will change your credence, of course, but you will benefit from a more accurate credence. But you may also be in the grip of confirmation bias. There's no sure fire way to guard against this. The only ways that I can think to identify this are 1) conversations with other people - it would be best to try to find people who are less biased than you are - or 2) critical self-examination.

  • +1 "Credence should be proportionate to the evidence". This means, use intuition and common sense to determine if domain-specific expertise applies and if so in what way, which is equally and delightfully non-specific since intuition, common sense, and domain-specific expertise are extraordinarily broad themselves!
    – J D
    Apr 12 at 17:55
  • If you know in general terms what proportion of partners are unfaithful and what proportion of fridges contain water bottles then you would have some indirect information to go on. You could enhance this with more specific information, for example if you put a water bottle in your fridge then there might be a higher-than-average probability that it's still there.
    – Frog
    Apr 12 at 23:48
  • 1
    @Frog I'm interpreting the project as confirming one's credence but NOT altering it. One could certainly try to enhance one's information. But that would alter the level of your credence rather than merely confirming it.
    – Ludwig V
    Apr 22 at 15:05
  • @Ludwig V I’m not sure you can have the one without the other
    – Frog
    Apr 24 at 10:56
  • @Frog I agree. That's what I intended to say in the last sentence of my answer.
    – Ludwig V
    Apr 25 at 5:26

What you feel about A and B is known as intuition and it cannot be rationally explained for two reasons:

  1. It is the result of statistical analysis that takes into account all your past experiences. It is unexplainable for the same reason AI is fundamentally unexplainable -- too many variables, too many computations.
  2. The above analysis is performed by Kahneman's System 1 -- a neural network supercomputer in your subconsciousness. You conscious mind has no visibility into that process (nor would it help if it had). Only the bottom line is communicated to you in the form of feelings.

That's why whatever pops out from our subconsciousness feels like magic. And that's why it's hard for us to buy into computational theory of mind -- because of the sheer amount of information processing that happens in there. It's unbelievable.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .