Anyone have a link to the actual study? The analogy with the streets makes it sound like subjects are following correct bandit-algorithm strategy because they d

Anyone have a link to the actual study? The analogy with the streets makes it sound like subjects are following correct bandit-algorithm strategy because they don't know (and have no way of knowing) the variability in outcomes is truly random