Knowledge Acquisition And The Conscious State Of Knowledge In Implicit Learning

As is well known, it is the fascinating talent of human being to be able to acquire implicitly complex structures and rules in the surroundings. The ability stimulates so many researchers to seek after the essence of the implicit learning. The common methods in implicit learning research consist mainly of complex system control paradigm, artificial grammar paradigm and serial reaction time paradigm; and the artificial grammar is preferred by the many researchers since it is closer to the natural language. Since its appearance (Reber,1967), the artificial grammar paradigm has been applied widely in solving the following two questions in implicit learning:(1) What kind of knowledge is acquired after the learning?(2) Is the knowledge acquired implicitly or explicitly? To answer these two key questions in the present dissertation we present a report about four experiments. These experiments were conducted by applying the classical artificial grammar paradigm, by introducing as well the subjective measurement method of structural knowledge invented by Dienes and Scott (2005).In Chapter1, the definition of implicit learning is introduced, and an overview is given for the background of emergence and development of the research. The following three issues are emphasized and expounded thoroughly:(1) the knowledge representation of implicit learning;(2) the instruction effect;(3) the influence factor of implicit learning;(4) the measurement of unconsciousness of implicit learning.Chapter2is the empirical part of the current study. The main purpose of Experiment1is to ascertain, on the basis of Guo et al.(2011), whether the subject's acquisition are rules or chunks under the two instructions of memory and rule search. It turns out that the subject under memory instruction can distinguish successfully legal strings from illegal ones, that is, there appears effect of grammar learning, while the subject under rule search instruction shows no effect of grammar learning or chunk learning; the results of subjective measurement of the structural knowledge show that the superiorities of implicit learning mainly come from the contribution of unconscious structural knowledge. All these results confirm that the unconscious structural knowledge plays an important role in artificial grammar learning. In Experiment2transfer tasks are introduced and it turns out that both incidental learning group and intentional learning group display well the grammar learning effect, and it shows that the grammar rules acquired in the artificial grammar learning are abstract and transferable. In Experiment3and4we controlled balanced the intensities of total chunks and anchor chunks as well adjacent repetitive structures and total repetitive structures, and besides, we manipulated two new variables of attention condition and learning mode. The results show that in contrast to chunk knowledge, the rule knowledge is robust and not prone to the interference of external factors such as interference tasks or presentation rate. On the basis of the results of these four experiments we claim that in the artificial grammar learning a learner can acquire both abstract rule knowledge and chunk knowledge. The rules acquired by implicit learning could not be influenced by attention resource and learning speed, and they are abstract and transferable; while chunk knowledge, acquired partly by implicit learning, in some extent could be influenced by attention resource, and it is not transferable.In Chapter3and4, summary and reflection are presented, the theoretical significance and possible application of the current study are expounded. In addition, the shortcomings of current study are explained and the new problems for further research are suggested.