In the physical and natural sciences, the complexity of natural systems and their interactions is becoming better understood. With increased emphasis on learning about complex systems, students will be encountering concepts that are dynamic, ill-structured and interconnected. Concept-mapping is a method considered particularly valuable for enabling learning in subject areas that are complex or ill-structured. Evaluations of concept-mapping tend to reflect their many applications. Many evaluations that try to measure enhancements in learning use test scores or grade-point averages as performance indicators, which provide little information on how cognitive processes have developed. In this study, a modification of the Biggs and Collis' Structure of the Observed Learning Outcomes (SOLO) taxonomy is applied to measure differences in the cognitive and structural complexity of learning outcomes between groups of students who participated in a concept-mapping activity with those from a cohort that did not. The evaluation demonstrated that the intervention was effective in enabling the great majority of students to achieve better connectivity in thinking, though improvements in overall performance were less significant. Almost all students perceived the intervention to be of value to their learning. (Contains 4 tables and 3 figures.)