Learning Efficiencies Using Multi-Agent Based Game Simulations Gordon Barker a , Barry Lee Reynolds b , Andrew Chan-Chio Lao b , Frank Wu b a Athabasca University, Alberta, Canada b Graduate Institute of Network Learning Technology National Central University, Jhongli, Taiwan gordon@barker.name Abstract: Learning effectiveness is a function of effective pedagogical practices. Accordingly, the question for learning technology designers should be “What combination of instructional strategies and delivery media will best produce the desired learning outcome for the intended audience?" This paper reviews four successful products that incorporate solid instructional strategies with agent based game simulations to provide a positive learning environment for students. Keywords: game based learning, simulation, learning technology, end-user programming Introduction The first introduction to basic computer science concepts that students receive occurs in elementary school. The students learn how to use some Windows applications such as PowerPoint, Word and Paint. Computer programming courses however, occur less frequently and are taught with lesser quality than that of other software packages. Research has revealed that children can improve their logical thinking, organization, judgment and problem solving ability through computer programming. Sedighian [1] studied various design strategies of simulations on children’s learning of and attitudes towards complex mathematical concepts. He researched the effectiveness of teaching two-dimensional geometry using different strategies. He developed three approaches including “direct object manipulation” (DOM) in which students manipulate a geometric object directly; “direct concept manipulation” in which the students manipulate the mathematical representation of the shape and “reflective direct concept manipulation” (RDCM) in which visual feedback aids were gradually removed, requiring students to think carefully about the object [2]. The conclusions were that simple manipulation of an object in a game or simulation environment had a slight negative impact on learning. The greatest increase in understanding came from those students using the RDCM method accentuated with involvement by the instructor. The main reason for this result is that learning how to play a game does not necessarily imply learning the target instructional domain [3]. Learning happens only when students actively build the connections between game moves and underlying knowledge. In order to solve this problem, computer scientists and programmers have developed the concept of programming through demonstration. End-users, taking advantage of the 737