1. Summary The distinction between implicit and explicit learning remains a controversial issue in cognitive psychology. It has been claimed that implicit learning, unlike explicit learning, is automatic in nature, i.e. requiring a minimum of attentional resources. Previous research has not yet settled this debate, often because the concept of attention was ill-defined. Therefore, I want to determine the influence of different attentional components on implicit learning, in particular visuospatial selective attention and attentional control. For this purpose, I will use behavioral measurements and Event-Related Potentials (ERP's) to assess whether perceptual and cognitive load (a) affect implicit learning of attended information and (b) allow implicit learning of unattended information. Implicit learning will be derived from incidental learning of task-relevant and task-irrelevant sequence structure in visual search tasks. If implicit learning is truly automatic in nature, learning should not be affected by increased perceptual and cognitive load. 2. Is implicit learning automatic? Traditionally, a distinction is made between conscious, hypothesis-driven explicit learning and unconscious, incidental implicit learning. Implicit learning is usually defined as the process through which we become sensitive to certain regularities in the environment (1) in the absence of intention to learn about those regularities (2) in the absence of awareness that one is learning, and (3) in such a way that the resulting knowledge is difficult to express (Cleeremans et al., 1998). Implicit learning is considered to be a primitive but powerful form of adaptation that is involved in the acquisition of simple as well as complex behavior, like for instance the acquisition of motor skill, language and social skills. Much evidence supports the distinction between implicit and explicit learning, and amnesic patients often show intact implicit learning but impaired explicit learning. Nevertheless, the existence of implicit learning and especially its unconscious nature remains subject to debate. Instead of attempting to define and measure the ambiguous concept of consciousness, several authors propose to refer to implicit learning as learning that occurs automatically, i.e. learning that requires a minimum of attentional resources. 2.1. Attentional capacity Numerous studies have demonstrated that people become incidentally sensitive to regularity in their environment while being engaged in a concurrent mental activity. Most of this research has been conducted using the serial reaction time task (SRT task, Nissen & Bullemer, 1987). In the SRT task, participants have to react as fast as possible to a target appearing in one of four horizontal locations by pressing a spatially corresponding response key. Although they are not informed that the target is presented following a regular sequence (for instance, 121342314324, with the numbers 1 to 4 corresponding to the locations from left to right), reaction times (RTs) progressively decrease with training and increase abruptly when the regular sequence is replaced by a random sequence (sequence learning effect). This pattern of results is observed in the absence of learning instructions and often without awareness of the sequential nature of the task. The automatic nature of implicit sequence learning in the SRT task is typically established in a dual task paradigm, in which cognitive load (e.g. working memory load), usually a tone-counting task, is added to the SRT task in order to reduce the amount of available attentional resources. The observation that sequence learning in several studies is intact under dual task conditions (e.g. Frensch, et al. 1994) supports that learning runs independently of attentional resources. However, other authors, reporting impaired learning in combination with tone- counting, claim that detrimental effects in previous studies might have been masked because of the increasing automatization of learning (e.g. Rowland & Shanks, 2006a). To conclude, despite many studies, the question regarding the need for attentional capacity remains unanswered. 2.2. Selective attention A number of authors have tried to solve the issue of automatic learning by focusing on selective aspects of attention. In the SRT task of Jimenez and Mendez (1999) the relevant location of the stimulus followed a structured sequence, but at the same time the irrelevant shape dimension of the stimulus predicted the subsequent stimulus location in 80% of the trials. This predictive relationship between location and shape was only learned by dual task participants who had to keep count of some of the shapes. Single task participants, who only performed the SRT task and thus did not need to attend to the shapes, did not demonstrate any shape learning. At the same time, sequence learning of the stimulus location was unaffected by the dual task manipulation. According to the authors, this suggests that implicit learning of regularity occurs automatically, but that the predictive information needs to be selectively attended to in order to be learned. However, it remains to be determined whether it is theoretically defendable to make a distinction between selective and capacity aspects of attention, since both are intrinsically linked. 3. Current research project 3.1. Another approach: multiple components of attention The studies mentioned above illustrate the need for developing alternative means to investigate the role of attention in implicit learning. A possible solution is to investigate attention in its various aspects. Research has demonstrated that attention is not a unitary concept. Different attentional networks have been distinguished at both cognitive and neuroanatomical levels (e.g. Posner & Petersen, 1990; Posner & Rothbart, 2007). A posterior attentional system, related to visual perception, has been proposed to be responsible for visuo-spatial selective attention (selective function), while an anterior attentional system operating at a later stage has been suggested to be responsible for control processes (executive function; e.g. Johnston et al., 1995; Lavie et al., 2004). As put forward by Rowland and Shanks (2006a), "Such componentional models of attention undoubtedly provide richer frameworks for studying the effects of attention on implicit learning".