We examined whether the classical framing effect observedwith the Asian Disease problem could be reversed when peoplemake decisions from experience. Ninety-five universitystudents were randomly allocated to one of three conditions:Description, Sampling (where the participants were allowed tosample through the outcomes presented as a pack of cards) andInteractive (where the participants were invited to spread outall possible outcomes in a sample) and made three gain-framedchoices and three loss-framed choices, with two filler tasksafter the first three choices. The results revealed a significantinteraction effect between framing and choice condition. In theDescription choice condition, participants were more risk-seeking with loss-framed problems. This pattern was reversedin the Sampling choice condition where participants were morerisk-seeking with gain frames. Finally, the Interactive choicecondition resulted in a classic pattern of framing effect,whereby people were more risk averse in the domain of gains.