STRUCTURAL KNOWLEDGE AS A PRECONDITION OF CONTROLLING COMPLEX DYNAMIC-SYSTEMS

In two experiments it was tested whether structural knowledge facilita tes the control of complex dynamic systems at learning and transfer In a practice phase the acquisition of structural knowledge was supporte d by demonstrating the dependencies between the system variables and b y tasks demanding knowledge application. The central groups solved com parable problems hut the acquisition of structural knowledge was not s upported. During the learning phase ail subjects had to control a comp lex dynamic system in several trials. The structural knowledge was ass essed by a ''pair-task'' where two variable names each were shown and subjects had to decide whether there was a relation between the two va riables in the system or not. Performance in a transfer task was analy sed as an indirect indicator of structural knowledge. In both experime nts the experimental groups acquired more structural knowledge than th e control groups. As expected this knowledge facilitated performance i n controlling the system at learning as well as at transfer. The struc turcal knowledge was acquired in less time and used more efficiently t han specific procedural knowledge. It was concluded that explicit stru ctural knowledge supports learning to control a dynamic system if that knowledge is acquired in an application-oriented context. Under these conditions explicit learning plays an important role in controlling c omplex dynamic systems.