Within a virtual physics laboratory on buoyancy in fluids N=189 eighth and ninth grade secondary school students acquired knowledge during scientific discovery learning. The aim of the study was to investigate the effectiveness of a short introductory modeling of metacognitive help and additionally given learning goals for both eighth and ninth grade students. We used a mixed design with the experimental factors metacognitive help (with vs. without) and goal specificity (specific vs. nonspecific learning goals), the quasi-experimental factor grade of school (8 vs. 9), the time-factor conceptual knowledge (three assessments), and intelligence and motivation as control variables. Results indicate metacognitive help as significantly promoting the learning outcome of the ninth grade students. However, despite their significant overall increase of knowledge the eighth grade students could not benefit from the metacognitive help. The interaction of the grade of school, the metacognitive help, and the conceptual knowledge as time-factor showed to be significant.
