Mechanisms underlying the emergence of consciously aware knowledge in an implicit learning task

在内隐学习研究领域，如何在内隐学习情境中产生对序列的有意识意识是一个争论不休的问题。在文献中可以将两类理论进行对比：根据第一类理论，显性表征是由于内隐表征的加强而产生的。因此，这一类假定隐式表示和显式表示之间的数量差异。相反，第二类假设了内隐知识和外显知识之间的质的差异。例如，意外事件假说(UEH；Haider&Frensch，2005)就属于这第二类理论。它假设，隐性知识会导致行为变化，这可能会违反预期。这种违规行为随后会触发显式搜索过程，进而导致有意识地获取知识。实验的目的是相互检验这两类理论。在第一系列实验中，我们通过例如序列在特定背景下的重复率来操纵内隐获得的知识的强度。根据第一类理论，这应该增加有意识的可及序列知识。然而，如果UEH适用，我们应该能够证明，为了增加有意识的知识，还需要额外的违反预期。第二系列实验的重点是反应-效果学习(R-E学习)，这可能会增加经历违反预期的可能性。例如，Hoffmann和他的同事(2001)的发现表明，对反应效应的关注增加了显性知识的数量。此外，Keele et al.(2003)提出R-E学习可能不同于S-S和R-R学习的基本表征。R-E学习被认为是基于维度间的学习过程，因为反应与远端环境中的事件有关。相比之下，S-S和R-R学习的结果是在一个单一维度内的表征。因此，第二系列实验的目标是在S-S和R-R偶然性保持不变的内隐学习情境下操纵R-E偶然性。我们推测，如果一个人自己的行为偶然导致了远端环境中的某些事件，那么关于S-S-和R-R序列的有意识的知识应该会增加。总体而言，计划中的实验有助于我们理解意识，以及在内隐学习环境中有意识地获得知识的潜在机制。

In the field of implicit learning research, it is a highly debated question how conscious awareness about a sequence arises in an implicit learning situation. Two classes of theories can be contrasted in the literature: According to the first class, explicit representations result from strengthening of implicit representations. Thus, this class presumes quantitative differences between implicit and explicit representations. On the contrary, the second class postulates qualitative differences between implicit and explicit knowledge. As an example, the Unexpected-Event Hypothesis (UEH; Haider & Frensch, 2005) belongs to this second class of theories. It presumes that implicit knowledge leads to behavioral changes which might violate the expectations. This violation then triggers explicit search processes which in turn lead to consciously accessible knowledge. The goal of the proposed experiments is to test these two classes of theories against each other.In the first series of experiments, we manipulate the strength of implicitly acquired knowledge by, for instance, the repetition rate of the sequence within a certain context. According to the first class of theories, this should increase consciously accessible sequence knowledge. However, should the UEH apply, we should be able to show that an additional violation of expectations is needed in order to increase conscious knowledge. The second series of experiments then focuses on response-effect learning (R-E learning) which might increase the probability of experiencing a violation of expectancies. For instance, the findings of Hoffmann and colleagues (2001) suggest that increasing attention to response effects increases the amount of explicit knowledge. Furthermore, Keele et al. (2003) suggest that R-E learning might differ from S-S and R-R learning in terms of the underlying representation. R-E learning is assumed to be based on inter-dimensional learning processes because reactions are associated with events in the distal environment. By contrast, S-S and R-R learning results from representations within one single dimension. Therefore, the goal of the second series of experiments is to manipulate the R-E contingencies in an implicit learning situation while S-S and R-R contingencies are kept constant. We presume that conscious knowledge about S-S- and R-R sequences should increase if ones own actions contingently lead to certain events in the distal environment. Overall, the planned experiments contribute to our understanding of consciousness and the mechanisms underlying the emergence of consciously accessible knowledge in an implicit learning situation.