Feedback in complex learning: Considering the relationship between utility and processing demands

While feedback has been a popular topic for research during the past century, there are still many unresolved issues. Two variables of interest, feedback probability acid feedback type (positive verses negative) have been shown to influence the effectiveness of instruction. I introduce a third variable of interest, scope.{09}The scope of feedback determines whether feedback is given for individual actions (local feedback) or higher-level problem solving goals (global feedback). Two studies were completed to better determine the interaction between these variables. Some forms of feedback place a high load on working memory; a key question was whether these forms would be ineffective. In Study 1, where negative feedback was given, subjects given high probability local feedback performed best during training, because they could more easily identify and correct errors. Subjects given global feedback performed best on two tasks designed to test transfer and memory. In Study 2, where positive feedback was given, global feedback was more effective during training. However, local and global subjects performed equally well on the tasks designed to test transfer and memory. Together these results suggest an interesting contrast. Negative feedback is more sensitive to changes in scope because when subjects are given information about errors, they are more likely to attempt correction. However correction requires intensive processing and may interfere with lasting learning. Because positive feedback helps a learner consider actions that have already been completed correctly, it can more easily be processed in relation to what the learner. For this reason, it is equally effective regardless of scope.