Between uncertainty and certainty

In this study, 10 hematologists and 10 lung oncologists were interviewed regarding the information they provide to patients in four situations of uncertainty: determining the treatment that is in the patient's best interest; recurrence or progression of the patient's disease; determining when to withdraw life-prolonging treatment; discussing death, addressing questions such as whether the patient will die from the disease, and when. The primary finding is that delivery of information to patients with low survival rates can be improved by more and better disclosure by physicians at an earlier stage. The crucial point for physicians is to ascertain the wishes of patients, to learn what to reveal about what patients should expect, short term and long term, as death approaches.     

Probability,certainty and illusions

Some philosopheis (e.g. Ayer, Reichenbach, Lewis) use a version of the argument from illusion to prove that empirical statements are never certain. But this argument, unwittingly, also calls into doubt the certainty of calculations in logic and mathematics. The argument seems to call into question the application of any rule on the grounds that one might at some future time find out that one had misapplied it. But the argument from illusion is only the illusion of an argument.     

Ordinary language and absolute certainty

Philosophical Studies -     

Knowledge,certainty and probability

In this essay, I discuss some of the important logical principles governing the concepts of knowledge, certainty and probability. In the first section, I suggest a series of definitions of epistemic terms, employing as primitive the locution ‘p is epistemi‐cally possible to S’ In the second section, I develop an epistemic concept of probability and compare it to the concepts of certainty and knowledge. In the third section, I relate the epistemic concepts of certainty and probability to the quantifiers of traditional logic and to a non‐episteznic concept of probability. I conclude by noting similarities and differences between the two concepts of probability.     

Knowledge,certainty, and assertion

Researchers have debated whether knowledge or certainty is a better candidate for the norm of assertion. Should you make an assertion only if you know it's true? Or should you make an assertion only if you're certain it's true? If either knowledge or certainty is a better candidate, then this will likely have detectable behavioral consequences. I report an experiment that tests for relevant behavioral consequences. The results support the view that assertability is more closely linked to knowledge than to certainty. In multiple scenarios, people were much more willing to allow assertability and certainty to come apart than to allow assertability and knowledge to come apart.     

Modeling certainty equivalents for imprecise gambles

We present results from two experiments on the relative importance of, and subjects' differential sensitivity to, vagueness on both probabilities and outcomes. Subjects in these studies made certainty equivalent (CE) judgments for precise and vague gambles. In the first study subjects responded to gain gambles only; in the second they judged both gain and loss gambles. Model-free analyses of the results indicate (a) a higher concern for the precision of the outcomes than that of the probabilities, (b) vagueness seeking for positive outcomes and (c) vagueness avoidance for negative outcomes and (d) no strong modal attitude toward vagueness on the probability dimension. The greater salience of the outcomes can be explained by the nature of the response mode (CEs). The reflection of attitudes towards outcome vagueness in the two domains can be explained by the distinct goals of the DMs in the two cases, which cause them to focus on the highest (most desirable) possible gain or the largest (most dreaded) conceivable loss. We propose and fit a new model of decision making with vaguely specified attributes that generalize the Prospect Theory model for the precise case. The new generalized model combines the two submodels (preference among precise lotteries and effects of vagueness) and allows estimation of the vagueness parameters. These estimated parameters are consistent with, and confirm, the patterns uncovered by the qualitative analysis.