It turns out that this maximization leads to the following answers.
For Alice: 1. Credence p1=33% in Geneva, but answers q1=100%. 2. Credence p2=33% in Lausanne, but answers q2=0%. 3. Credence p3=33% in Zurich, but answers q3=0%. 4. Credence p4=33% in Lugano, but answers q4=0%.
I am surprised by these numbers:
i) I assume that p4=33% and not p4=1% is a typo?
ii) Also, when reading that q1=100%, while q2,q3= 0%, I was surprised. As p1,p2 and p3 are the same, (if I am not mistaken) Alice should be free to arbitrarily divide her probability mass between these three? Given that, I expected her to choose q1=q2=q3. In case others were confused by this detail too, it might be worth it to slightly complicate the example (along the lines of ‘Alice remembers an ambitious athlete friend being invited to Geneva once’ and using this as tie breaker for the honest probabilities)
I am surprised by these numbers:
i) I assume that p4=33% and not p4=1% is a typo?
ii) Also, when reading that q1=100%, while q2,q3= 0%, I was surprised. As p1,p2 and p3 are the same, (if I am not mistaken) Alice should be free to arbitrarily divide her probability mass between these three? Given that, I expected her to choose q1=q2=q3. In case others were confused by this detail too, it might be worth it to slightly complicate the example (along the lines of ‘Alice remembers an ambitious athlete friend being invited to Geneva once’ and using this as tie breaker for the honest probabilities)