Within the scope of this study, the effectiveness of two kinds of instructional support was evaluated with regard to the learner's interests. Two versions of a simulation program about the respiratory chain were developed, differing only in the kind of tasks provided for instructional support: One version contained problem-solving tasks, the other one contained worked-out examples. The focus was on the learner's interest in the subject and in computers. The first goal of the study was to find to what extent computer simulations incorporating the different kinds of instructional support have positive effects on situational subject-interest. The second goal was to evaluate the interactions between the learner's interests and the instructional support with regard to the learning results (subdivided into factual knowledge and understanding). Simulations with worked-out examples were shown to have positive effects on the learner's situational interest in the subject. This was not found to be the case in simulations with problem-solving tasks. Regardless of the kind of instructional support, learners with little interest in the subject were able to achieve significant gains in factual knowledge. However, improvement in understanding was dependent on the kind of instructional support. (C) 2008 Elsevier Ltd. All rights reserved.