The issue is orthogonal to the provenance of conditional probability in the mind, that is, to the way such probabili- ties are mentally computed. exper 2 Also, (1) is a consequence of compelling axioms on belief change (Gärdenfors, 1988, Section 5.2), and its violation 0010-0277/$ - see front matter � 2012 Elsevier B. It is easy to check that Pr2 as deﬁned by (1) is a genuine probability distribution and that Pr (B) = 1 (as expected). For example, is the judged probability of a Dem- ocratic victory in 2012 supposing that Hilary Clinton is the vice presidential candidate the same as the judged proba- bility of a Democratic victory in 2012 after learning that Clinton, as a matter of fact, is the vice presidential candidate?
For each deck, the experimenter ﬁrst showed the cards to the participant, with the animals and colors in plain view.Branden Fitelson d, Daniel Osherson a ts a tight relationship between conditional probability and updat- ility of an event A after learning B should equal the conditional rior to learning B. In three experiments we found substantial differences probability of an event A supposing an event B compared to the it is still possible for (1) to hold.2 All that matters is whether the same degree of conﬁdence in event A is reached when supposing event B compared to learning it.This question seems not to have been addressed by pre- vious experiments on updating.Cards were presented brieﬂy (around 0.5 s apiece) to pre- vent counting.After all cards in the deck were presented, the participant shufﬂed the deck, drew one card at random, and put it on the table blank side up.