This paper explores the influence of symbolic or representational learning materials on pupil engagement or learning outcomes, when 14-16 year old pupils use common types of science simulations. The project pilot phase involved three (15-16 year old) male pupils and a main phase involved twenty-one (14-15 year old) pupils. A retrospective accounts methodology (Clarke, 1998) presented pupils with a digital record of their "think aloud" (Ericsson & Simon, 1984) behaviour with the simulations, and they were asked for retrospective comment. Pre- and post-surveys were also used. This interaction record for two boys is used to illustrate the findings. This record was chosen, because the boys spoke aloud throughout the period of engagement, which generated useful data for microanalysis. Findings suggest that pupils working with science simulations face a trans-disciplinary demand (computer competence, information processing skills, traditional language proficiency, and science understanding.) In terms of traditional and subject literacy, pupils have to make sense of linguistically complex information when using common science simulations. They need to understand their subject, be confident and competent with the available technology, and possess language skills that enable them to establish links between the microscopic, macroscopic and symbolic components of science.