Suppose an agent has to choose between two main options. He can choose neither or either, but not both. His preference for each of the options is unknown, probably even by the agent himself (behaviour during the experiment signals indecision). Picture the experiment akin to a subject getting to make a pick between two useable objects such as toys, cars, gadgets etc. He is allowed to play/experiment with both of them during the experiment. Throughout the experiment, he exhibits a moderate preference for one of them, and spends more time using it.

Then, the experimenter removes the other choice from the setup, and subsequently interviews the subject on his preferences for both options, while asking him to disregard the fact that his preferences do not matter anymore in the outcome. So, he's only left with one, presumably the favourite so far, but has to express opinions on both, after the fact.

1. Based on your model of human choice, how likely is he to irrationally exhibit a higher preference for the removed option and a decreased preference for the remaining one, on the grounds that the choice was no longer "his" to make?

2. How does this compare to the somewhat opposite "sour grapes" irrational tendency?

3. Has such an experiment been made, to confirm or disprove the existence of this bias?

(P.S. I'm aware there's a design flaw in that the first preference is observed and the other is self-reported. If that's a no-no, let's introduce a step where the subject is interviewed with both options still available, and measure increases or decreases relative to that.)