This study investigates the impact of utilizing dynamically linked concrete and abstract representations of objects in modern, virtual electric circuit laboratories on the cognitive evolution and the representational fluency of high school students. The students (N = 27, aged 16-17) were randomly divided into two classes: the first class used a realistic virtual laboratory environment, dynamically linked to a representation of the schematics of each circuit constructed, while the other class used the same virtual environment without the circuit schematics representation affordance. Both classes were given the same inquiry-based instruction and the same instructional material. A pre-/post-instructional assessment scheme was used to evaluate students' knowledge evolution and skills in reconstructing electric circuits from an either realistic virtual or abstract schematic form to the other. The statistical analysis of the results shows that, after instruction, all students in both groups are capable of understanding thoroughly the operation and transforming simple electric circuits from any given form to the other successfully. Our findings also indicate that the students who used the dynamically linked concrete and abstract representations of electric circuits during teaching seem to outperform the students of the other group, regarding their understanding of the operation and the transformation of more complex electric circuits. Furthermore, the ability to transform electric circuits from one form to another is found to correlate well with students' conceptual evolution. Potential implications of these findings are being discussed, while a model of the mental model construction process is being proposed.