A computer model of the role of text integration in the solution of arithmetic word problems

Understanding arithmetic word problems involves a complex interaction of text comprehension and mathematical processes. This work presents a computer model of the hypothesized processes that are required of a young student solving arithmetic word problems, including the processes of sentence-level leading and text integration. Unlike previous computer simulations of word problem solving, which neglect the early stages of text processing, this model forces a detailed consideration of the linguistic process, which is being increasingly recognized as a primary source of difficulty. Three experiments were conducted to isolate critical test comprehension processes. Children\u27s probability of solution was analyzed in regression analyses as a function of the model\u27s text comprehension processes. A variable measuring the combined effects of the load on working memory and text integration inferences accounted for a significant amount of variance across four grade levels (K-3). The results suggest new process-oriented measures of determining why a particular word problem may be difficult, especially for young students. An implication for education is the potential for a difficulty-differentiated network of problems that includes a multiple number of rewordings for each traditional problem wording as an aid for classroom assessment, curriculum development and future computer-based learning environments