In January 2010, the LISS panel completed a questionnaire about the choices one makes when confronted with a known and unknown probability distribution. The respondents were presented with three games. Each game consisted of six rounds at most. The chance of winning with the known probability distribution varied from the second round on, depending on the respondent's choice in the previous round (‘known’ or ‘unknown’). After three games, two control questions were asked. The chance of winning with these questions depended on the chance of winning in the last round of game 1. Finally, two questions were posed about risk aversion. These questions also involved a game consisting of six rounds at most. Respondents either chose the option that guaranteed winning a certain sum, or the option that offered a 50% chance of a higher amount and a 50% chance of 0 euro. The amount for the guarantee option varied from the second round on, depending on the choice made in the previous round. Respondents were assigned one of two conditions beforehand. Respondents had 50% chance of being assigned condition 1, and 50% chance of condition 2. Within a household, panel members were assigned the same condition. Respondents under condition 1 knew that this involved a fictitious game and that no extra reward would be paid out. Respondents under condition 2 were informed that a real game would be played at the end, in which they could earn an extra reward of 15 euros.