Components of complex systems apply across multiple subject areas, and teaching these components may help students build unifying conceptual links. Students, however, often have difficulty learning these components, and limited reasearch exists to understand what types of interventions may best help improve understanding. We investigated 32 high school students' understandings of complex systems components and whether an agent-based simulation could improve their understandings. Pretest and posttest essays were coded for changes in six components to determine whether students showed more expert thinking about the complex system of the Chesapeake Bay watershed. Results showed significant improvement for the components Emergence (r = .26, p = .03), Order (r = .37, p = .002), and Tradoffs (r = .44, p = .001). Implications include that the experiential nature of the simulation has the potential to support conceptual change for some complex systems components, presenting a promising option for complex systems instruction.