Model-Based Assessment of Conceptual Representations Aytac Gogus This research study uses a model-based assessment by eliciting, representing, and analyzing how individuals conceptualize complex problems. This study uses an au- tomated, online, model-based set of tools called HIMATT (Highly Interactive Model- based Assessment Tools and Technologies) to compare the conceptual representations of the students with the conceptual representations of the instructors. According to findings, the tools of HIMATT, which are experimentally used to assess model-based cognitive representations, can provide an informative feedback for the instructors and can be applied to the assessment of learning in complex areas of education. The future plan is to translate HIMATT to Turkish to allow educators to use during instruction and assessment of learning in Turkey. Keywords: assessing conceptual representations, problem based learning, assessment of learning in complex domains, model based assessment, HIMATT. Information and communication tech- nologies have important roles in learning and instruction. Computers can be used as critical thinking tools and tools for the processes of problem solving. Thus, computers can help acquisition of facts to cognitive processes that encourage curiosity and creativity, and help students become better problem solv- ers (Seel & Blumshein, 2009). According to Ingham and Gilbert (1991), a model is a simplified representation of a system, which concentrates attention on specific aspects of the system, i.e., objects, events, or ideas. In cognitive psychology, conceptions of model- ing are captured by the theoretical concept of mental models (Seel & Blumshein, 2009). Mental models are defined as either personal internal representations of the target system being modeled or external representations of the target generated from one’s mental models and expressed through action, speech, written description, and other material depictions (Johnson-Laird, 1983; Gobert & Buckley, 2000). Since the mental models which are described as theoretical constructs are not observable, researchers can only learn about mental models if individuals communicate their mental models to externalize their mental models (Seel, Ifenthaler, & Pirnay-Dummer, 2009). The creation of an external model (a conceptual representation) can deepen one’s understanding of a complex domain and the models can be used to assess progress of learning and the development of expertise (Spector, 2009). With the recognized importance of models in education comes the need for ap- plying a theory of model-based learning and teaching (Gobert & Buckley, 2000). Using models to support learning and assessment is matching with the constructivist view that knowledge must be formed within the indi- vidual mind (von Glasersfeld, 1987, 1995) as well as collectively in science and society (Vygotsky, 1962). Learning occurs when people actively construct meaningful repre- sentations from presented information, such as coherent mental models that represent and communicate subjective experiences, ideas, thoughts, and feelings (Mayer et al., 1999; Seel & Blumshein, 2009). Using conceptual representations (mental models) constructed in response to problem scenarios can be an ef- Aytac Gogus, Sabancı University, Istanbul, Turkey Correspondence concerning this article should be addressed to Dr. Aytac Gogus at agogus@sabanciuniv.edu. PLEASE INCLUDE DEPARTMENT AND POSITION. THANK YOU.