Abstract
In this article the discrimination model of probabilistic choice proposed by Lindenberg is tested in a telephone survey. This model is based on the assumption of bounded rationality in the sense of restricted information processing capacity. Unlike subjective expected utility (SEU) theory, which is also tested, the discrimination model assumes that the cognitive limitations are so severe that actors will only be able to focus on one main situational goal, or `frame', at a time. Problems of this model are discussed, and measurement instruments and the results of a telephone survey with road users are presented. In an `experimentum crucis' the discrimination model is tested against the SEU model. (A)