Using a simple change detection task involving vertical and horizontal stimuli, I. M. Thornton and D. Fernandez-Duque (2000) showed that the implicit detection of a change in the orientation of an item influences performance in a subsequent orientation judgment task. However, S. R. Mitroff, D. J. Simons, and S. L. Franconeri (2002) were not able to replicate this finding after correcting for confounds and thus attributed Thornton and Fernandez-Duque’s results to methodological artifacts. Because Mitroff et al.’s failure to replicate might in turn have stemmed from several methodological differences between their study and those of Thornton and Fernandez-Duque (2000) and Fernandez-Duque and Thornton (2003), the current authors set out to conduct a further replication in which they corrected all known methodological biases identified so far. The results suggest that implicit change detection indeed occurs: People’s conscious decisions about the orientation of an item appear to be influenced by previous undetected changes in the orientation of other items in the display. Implications of this finding in light of current theories of visual awareness are discussed.

Artificial grammar learning (AGL) is one of the most commonly used paradigms for the study of implicit learning and the contrast between rules, similarity, and associative learning. Despite five decades of extensive research, however, a satisfactory theoretical consensus has not been forthcoming. Theoretical accounts of AGL are reviewed, together with relevant human experimental and neuroscience data. The author concludes that satisfactory understanding of AGL requires (a) an understanding of implicit knowledge as knowledge that is not consciously activated at the time of a cognitive operation; this could be because the corresponding representations are impoverished or they cannot be concurrently supported in working memory with other representations or operations, and (b) adopting a frequency-independent view of rule knowledge and contrasting rule knowledge with specific similarity and associative learning (co-occurrence) knowledge.

A machine with human like consciousness would be an extremely complex system. Prior work has demonstrated that the way in which information handling resources are organized (the resource architecture) in an extremely complex learning system is constrained within some specific bounds if the available resources are limited, and that there is evidence that the human brain has been constrained in this way. An architectural concept is developed for a conscious machine that is within the architectural bounds imposed by resource limitations. This architectural concept includes a resource driven architecture, a description of how conscious phenomena would be supported by information processes within that architecture, and a description of actual implementations of the key information processes. Other approaches to designing a conscious machine are reviewed. The conclusion is reached that although they could be capable of supporting human consciousness-like phenomena, they do not take into account the architectural bounds imposed by resource limitations. Systems implemented using these approaches to learn a full range of cognitive features including human like consciousness would therefore require more information handling resources, could have difficulty learning without severe interference with prior learning, and could require add-on subsystems to support some conscious phenomena that emerge naturally as consequences of a resource driven architecture. Key words: consciousness; information model; system resource architecture; system design

The radical plasticity thesis — 2 In this chapter, I sketch a conceptual framework which takes it as a starting point that conscious and unconscious cognition are rooted in the same set of interacting learning mechanisms and representational systems. On this view, the extent to which a representation is conscious depends in a graded manner on properties such as its stability in time or its strength. Crucially, these properties are accrued as a result of learning, which is in turn viewed as a mandatory process that always accompanies information processing. From this perspective, consciousness is best characterized as involving (1) a graded continuum defined over “quality of representation”, such that availability to consciousness and to cognitive control correlates with quality, and (2) the implication of systems of metarepresentations. A first implication of these ideas is that the main function of consciousness is to make flexible, adaptive control over behavior possible. A second, much more speculative implication, is that we learn to be conscious. This I call the “radical

When one is conscious of something, one is also conscious that one is conscious. Higher-Order Thought Theory [Rosenthal, D. (1997). A theory of consciousness. In N. Block, O. Flanagan, & G. Güzeldere (Eds.), The nature of consciousness: Philosophical debates. Cambridge, MA: MIT Press] takes it that it is in virtue of the fact that one is conscious of being conscious, that one is conscious. Here, we ask what the computational mechanisms may be that implement this intuition. Our starting point is Clark and Karmiloff-Smith’s [Clark, A., & Karmiloff-Smith, A. (1993). The cognizer’s innards: A psychological and philosophical perspective on the development of thought. Mind and Language, 8, 487–519] point that knowledge acquired by a connectionist network always remains “knowledge in the network rather than knowledge for the network”. That is, while connectionist networks may become exquisitely sensitive to regularities contained in their input–output environment, they never exhibit the ability to access and manipulate this knowledge as knowledge: The knowledge can only be expressed through performing the task upon which the network was trained; it remains forever embedded in the causal pathways that developed as a result of training. To address this issue, we present simulations in which two networks interact. The states of a first-order network trained to perform a simple categorization task become input to a second-order network trained either as an encoder or on another categorization task. Thus, the second-order network “observes ” the states of the first-order network and has, in the first case, to reproduce these states on its output units, and in the second case, to use the states as cues in order to solve the secondary task. This implements a limited form of metarepresentation, to the extent that the second-order network’s internal representations become re-representations of the first-order network’s internal states. We conclude that this mechanism provides the beginnings of a computational mechanism to account for mental attitudes, that is, an understanding by a cognitive system of the manner in which its firstorder

Abstract not found

Implicit learning – broadly construed as learning without awareness – is a complex, multifaceted phenomenon that defies easy

journal homepage: www.elsevier.com/locate/concog

Abstract: It is usually taken as given that consciousness involves superior or more elaborate forms of information processing. Contemporary models equate consciousness with global processing, system complexity, or depth or stability of computation. This is in stark contrast with the powerful philosophical intuition that being conscious is more than just having the ability to compute. I argue that it is also incompatible with current empirical findings. I present a model that is free from the strong assumption that consciousness predicts superior performance. The model is based on Bayesian decision theory, of which signal detection theory is a special case. It reflects the fact that the capacity for perceptual decisions is fundamentally limited by the presence and amount of noise in the system. To optimize performance, one therefore needs to set decision criteria that are based on the behaviour, i.e. the probability distributions, of the internal signals. One important realization is that the knowledge of how our internal signals behave statistically has to be learned over time. Essentially, we are doing statistics on our own brain. This ‘higherorder’ learning, however, may err, and this impairs our ability to set and maintain optimal criteria for perceptual decisions, which I argue is central to perception consciousness. I outline three possibilities of how conscious perception might be affected by failures of ‘higher-order ’ representation. These all imply that one can have a dissociation between consciousness and performance. This model readily explains blindsight and hallucinations in formal terms, and is beginning to receive direct empirical support. I end by discussing some philosophical implications of the model.

This paper explores the interplay between unconscious, conscious, and metaconscious processes in social cognition. We distinguish among mental states that are (i) genuinely unaware, (ii) aware, but lack meta-awareness, and (iii) meta-aware—internally articulated as states of the perceiver. We review key studies from our own and related research programmes to highlight this theoretical framework, and to illustrate access, translational, and temporary dissociations between levels of awareness. The discussed phenomena include unconscious affect, mind-wandering, verbal overshadowing, theory-based biases in reporting of experiences, and many others. We also show how our framework can offer new perspectives on some classic social psychology findings and inspire discovery of new findings. However, we also highlight challenges inherent in establishing whether a phenomenon is genuinely unconscious or experientially conscious but lacking in meta-awareness. In daily life, and in scientific studies, people are frequently asked questions like ‘‘How do you feel right now?’’, ‘‘Do you want to smoke?’’, ‘‘Do you find