An investigation of procedural memory as an ingredient of task performance in accounting.

摘要

The development of expertise has attracted recent attention in the field of accounting. Cognitive psychology offers several interrelated theories to explain how humans acquire, store, and utilize knowledge to solve problems, to make decisions, and to perform judgments. They revolve around the idea that a working memory which is severely limited in capacity, is the weak link in human cognition. Automatization is an adaptive measure the mind has taken to conserve working memory space. As a result, all mental activity is theorized as the interworkings of conscious, effortful activity controlled by automatic productions that are often unconscious and unavailable to consciousness. Conscious knowledge is referred to as declarative knowledge; unconscious, automatic productions are referred to as procedural knowledge. Cognitive psychologists theorize that declarative knowledge when automatized into productions becomes increasingly unavailable to consciousness. The general research question addressed in this study was as follows: Is there a learned procedural component of memory that contributes to the performance of accounting tasks along with, but in a different way than, declarative knowledge?; A laboratory experiment was conducted in which students were trained to perform a representative accounting task. Cognitive skill acquisition and skill performance were measured with computers during interactive practice and postpractice tests. Consciousness of the manipulation of data possibilities and recall of the rules to perform the task also were measured.; First, the data collected support the notion that procedural knowledge contributes to performance in a different way than declarative knowledge. The results offer strong support that (a) subjects were unaware of the manipulation of data possibilities during the experiment and (b) subjects performed more accurately on data possibilities that were more practiced. Second, prepractice and postpractice declarative knowledge testing demonstrated an improvement in the recognition of the steps and rules used to perform the task, just the opposite of what was predicted by cognitive theory.; Future research should utilize a longer time frame to explore how much practice must take place before access to declarative knowledge of a process begins to decline. In addition, future research should explore further the possibility that more refined automatic productions are formed around separate data possibilities.