| Implicit learning is defined as the nonintentional, automatic acquisition of knowledge about structural relations between objects or events. A new theoretical framework of implicit learning mechanisms is constructed in this dissertation, named Multi-Level Development View. Also, some empirical researches are done to support this theory. These theoretical and empirical issues propose that we should reconsider the continuous relations between implicit and explicit learning, but not analyze statically. It will create a new research area on implicit learning.In Chapter 1, the definition of implicit learning is discussed, the research backgrounds are reviewed, and the basic principles and the defects of the main experimental procedures are concluded. It is pointed out that the developed methodological approaches on implicit memory can be exploited by the researches on implicit learning.In Chapter 2, the neural mechanisms of sequence learning are concerned on neuroimaging and neuropsychological researches with human subjects. Behavioral researches on mental representation and processing automatization are reviewed. It especially mentions the value of secondary tasks to researches on implicit learning.In Chapter 3, inspired by the developmental mechanism of knowledge representation proposed by Karmiloff-Smith and the evolutionary considerations of implicit learning claimed by Arthur S. Reber, we state a Multi-Level Development View of implicit learning mechanisms, with multi-level awareness, representations and resources requirement thresholds. Each two levels are mediated by Representational Redescription (RR). Exogenous RR is distinguished from the endogenous one, and it follows the principle of Optimal Coding. Both of these two processes are motivated intrinsically. It needs resources to achieve RR, that is themain reason why there exist multi-level resources requirement thresholds. Furthermore, some assumptions on individual differences of implicit learning are carried out in term of the new theoretical framework. In the final section of Chapter 3, it involves a new methodological issue: detection of aging effects of implicit learning may contribute to find out the role of some proved processing mechanisms with age differences. It develops the main research logic in the following experiments.In Chapter 4, empirical studies are conducted. Sequence Learning procedure is applied in the first three experiments, which involve the issue of representation mechanisms and processing automatization. The results indicate that there exist .multi-representation and confirm the role of attention in implicit learning. The following views are falsified respectively: the secondary task (1) suppresses expression of learning; (2) disrupts grouping in learning; (3) reduces short-time memory capacity in learning. Especially, the principle of Optimal Coding is exploited to explain the results in Experiment 1, in which a sequence learning material with underlying structure is constructed; the influence of the secondary tasks presented through different sense modalities is explored firstly in Experiment 3, which results in the valuable conclusion. Converging operations are designed hierarchically to detect aging effects of implicit learning, which is the cue in these three experiments. The results show the possibility of multi-level representations and attention requirement thresholds. Hidden Covariation Detection procedure is used in the latter two experiments. The effects of some external factors, such as stimulus features, learning length and achievement motivation, are explored. To some degree, the results are consistent with the conclusions on the implicit learning mechanisms drown from the previous three experiments. Individual differences in implicit learning are another topic of these five experiments.In Chapter 5, summary and reflection are carried out. It emphasizes on the theory implication and the application values of this research. The inadequacy and the research points in future are also lightened. |
PDF Full Text Request