What do people learn when they do not know that they are learning? Until recently, all of the work in the area of implicit learning focused on empirical questions and methods. In this book, Axel Cleeremans explores unintentional learning from an information-processing perspective. He introduces a theoretical framework that unifies existing data and models on implicit learning, along with a detailed computational model of human performance in sequence-learning situations.

69 Thompson-Schill, S.L. _et al. _(1997) Role of left inferior prefrontal cortex 59 Buckner, R.L. _et al. _(1996) Functional anatomic studies of memory in retrieval of semantic knowledge: a re-evaluation _Proc. Natl. Acad._ retrieval for auditory words and pictures _J. Neurosci. _16, 6219–6235 _Sci. U. S. A. _94, 14792–14797 60 Buckner, R.L. _et al. _(1995) Functional anatomical studies of explicit and 70 Baddeley, A. (1992) Working memory: the interface between memory implicit memory retrieval tasks _J. Neurosci. _15, 12–29 and cognition (...) _J. Cogn. Neurosci. _4, 281–288 61 Bäckman, L. _et al. _(1997) Brain activation in young and older adults 71 Petrides, M. (1994) Frontal lobes and behavior _Curr. Opin. Neurobiol._ during implicit and explicit retrieval _J. Cogn. Neurosci. _9, 378–391. (shrink)

Complete URL to this document: http://srsc.ulb.ac.be/axcWWW/93-Principles.html.

While the study of implicit learning is nothing new, the field as a whole has come to embody — over the last decade or so — ongoing questioning about three of the most fundamental debates in the cognitive sciences: The nature of consciousness, the nature of mental representation (in particular the difficult issue of abstraction), and the role of experience in shaping the cognitive system. Our main goal in this chapter is to offer a framework that attempts to integrate current (...) thinking about these three issues in a way that specifically links consciousness with adaptation and learning. Our assumptions about this relationship are rooted in further assumptions about the nature of processing and of representation in cognitive systems. When considered together, we believe that these assumptions offer a new perspective on the relationships between conscious and unconscious processing and on the function of consciousness in cognitive systems. (shrink)

Implicit learning is 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.

Consciousness has many elements, from sensory experiences such as vision and bodily sensation, to nonsensory aspects such as memory and thought. All are presented as experiences of a single subject, and all seem to be contained within a unified field of experience. This unity raises many questions: How do diverse systems in the brain co-operate to produce a unified experience? Are there conditions under which this unity breaks down? Is conscious experience really unified at all? Such questions are addressed in (...) this thought-provoking book. (shrink)

Over the past few years numerous proposals have appeared that attempt to characterize consciousness in terms of what could be called its computational correlates: Principles of information processing with which to characterize the differences between conscious and unconscious processing. Proposed computational correlates include architectural specialization (such as the involvement of specific regions of the brain in conscious processing), properties of representations (such as their stability in time or their strength), and properties of specific processes (such as resonance, synchrony, interactivity, or (...) information integration). In exactly the same way as one can engage in a search for the neural correlates of consciousness, one can thus search for the computational correlates of consciousness. The most direct way of doing is to contrast models of conscious versus unconscious information processing. In this paper, I review these developments and illustrate how computational modeling of specific cognitive processes can be useful in exploring and in formulating putative computational principles through which to capture the differences between conscious and unconscious cognition. What can be gained from such approaches to the problem of consciousness is an understanding of the function it plays in information processing and of the mechanisms that subtend it. Here, I suggest that the central function of consciousness is to make it possible for cognitive agents to exert ?exible, adaptive control over behavior. From this perspective, consciousness is best characterized as involving (1) a graded continuum de?ned over quality of representation, such that availability to consciousness and to cognitive control correlates with properties of representation, and (2) the implication of systems of meta-representations. (shrink)

We explore a network architecture introduced by Elman (1988) for predicting successive elements of a sequence. The network uses the pattern of activation over a set of hidden units from time-step 25-1, together with element t, to predict element t + 1. When the network is trained with strings from a particular finite-state grammar, it can learn to be a perfect finite-state recognizer for the grammar. When the network has a minimal number of hidden units, patterns on the hidden units (...) come to correspond to the nodes of the grammar, although this correspondence is not necessary for the network to act as a perfect finite-state recognizer. We explore the conditions under which the network can carry information about distant sequential contingencies across intervening elements. Such information is maintained with relative ease if it is.. (shrink)

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 plasticity thesis” — the hypothesis that consciousness emerges in systems capable not only of learning about their environment, but also about their own internal representations of it. (shrink)

Comparing the relative sensitivity of direct and indirect measures of learning is proposed as the best way to provide evidence for unconscious learning when both conceptual and operative definitions of awareness are lacking. This approach was first proposed by Reingold & Merikle (1988) in the context of subliminal perception. In this paper, we apply it to a choice reaction time task in which the material is generated based on a probabilistic finite-state grammar (Cleeremans, 1993). We show (1) that participants (...) progressively learn about the statistical structure of the stimulus material over training with the choice reaction time task, and (2) that they can use some of this knowledge to predict the location of the next stimulus in a subsequent “generation” task. However, detailed partial correlational analyses of the correspondence between performance during the reaction time task and the statistical structure of the training material showed that large effects remained even when controlling for explicit knowledge as assessed by the generation task. Hence we conclude (1) that at least some of the knowledge expressed through reaction time performance can not be characterized as conscious, and (2) that even when associations are found at a global level of analysis, dissociations can still be obtained when more detailed analyses are conducted. Finally, we also show that participants are limited in the depth of the contingencies they can learn about, and that these limitations are shared by the Simple Recurrent Network model of Cleeremans & McClelland (1991). (shrink)

Consciousness remains a mystery—“a phenomenon that people do not know how to think about—yet” (Dennett, , p. 21). Here, I consider how the connectionist perspective on information processing may help us progress toward the goal of understanding the computational principles through which conscious and unconscious processing differ. I begin by delineating the conceptual challenges associated with classical approaches to cognition insofar as understanding unconscious information processing is concerned, and to highlight several contrasting computational principles that are constitutive of the connectionist (...) approach. This leads me to suggest that conscious and unconscious processing are fundamentally connected, that is, rooted in the very same computational principles. I further develop a perspective according to which the brain continuously and unconsciously learns to redescribe its own activity itself based on constant interaction with itself, with the world, and with other minds. The outcome of such interactions is the emergence of internal models that are metacognitive in nature and that function so as to make it possible for an agent to develop a (limited, implicit, practical) understanding of itself. In this light, plasticity and learning are constitutive of what makes us conscious, for it is in virtue of our own experiences with ourselves and with other people that our mental life acquires its subjective character. The connectionist framework continues to be uniquely positioned in the Cognitive Sciences to address the challenge of identifying what one could call the “computational correlates of consciousness” (Mathis & Mozer, ) because it makes it possible to focus on the mechanisms through which information processing takes place. (shrink)

Running head: Implicit sequence learning ABSTRACT Can we learn without awareness? Although this issue has been extensively explored through studies of implicit learning, there is currently no agreement about the extent to which knowledge can be acquired and projected onto performance in an unconscious way. The controversy, like that surrounding implicit memory, seems to be at least in part attributable to unquestioned acceptance of the unrealistic assumption that tasks are process-pure, that is, that a given task exclusively involves either implicit (...) or explicit knowledge. (shrink)

Consider the following three situations: learning to perform a complex skill such as gymastics (a stunning demonstration of which participants to ICP 2004 experienced during the opening ceremony), learning a complex game such as the ancient Chinese game of Weichi (more widely known as Go), or learning natural language. What these situations have in common, beyond the sheer complexity of the required skills, is the fact that most of what we learn about each appears to proceed in a manner that (...) does not depend so much on the acquisition of explicit, declarative information or on the deployment of intentional strategies, but instead critically depends on repeated practice: Developing the skills needed to execute complex movements in gymnastics, to. (shrink)

Over the past decade, many findings in cognitive about the contents of consciousness: we will not address neuroscience have resulted in the view that selective what might be called the ‘enabling factors’ for conscious- attention, working memory and cognitive control ness (e.g. appropriate neuromodulation from the brain- stem, etc.). involve competition between widely distributed rep-.

referred to as implicit learning (Reber, 1989). Implicit learning contrasts with explicit learning (exhibited for.

Over the past decade, sequence learning has gradually become a central paradigm through which to study implicit learning. In this chapter, we start by briefly summarizing the results obtained with different variants of the sequence learning paradigm. We distinguish three subparadigms in terms of whether the stimulus material is generated either by following a fixed and repeating sequence (e.g., Nissen & Bullemer, 1987), by relying on a complex set of rules from which one can produce several alternative deterministic sequences (e.g., (...) Lewicki, Hill & Bizot, 1988; Stadler, 1989), or by following the output of a probabilistic set of rules such as instantiated by noisy finite-state grammars (Cleeremans & McClelland, 1991; Jiménez, Mendéz & Cleeremans, 1996). Next, we focus on the processes involved in sequence representation and acquisition. We suggest that the sensitivity to the sequential structure observed in the probabilistic subparadigm can only be a result of the acquisition of a representation of the statistical constraints of the material, and that this sensitivity emerges through the operation of mechanisms that are well instantiated by connectionist models such as the Simple Recurrent Network (Elman, 1990; Cleeremans, 1993b). We present new simulation work meant to explore to what extent the model can also account for specific data obtained in a paradigmatic instance of deterministic, rule-based sequence learning task: Lewicki et al. (1988)'s situation. Finally, we report on the results of an experiment that compares learning on otherwise similar deterministic and probabilistic structures, and we show that learning of both types of structures is equivalent only under conditions that maximally hinder explicit acquisition. Taken together, these simulation and experimental data lend support to the claim that implicit learning in all three sequence learning subparadigms can amount to a form of statistical sequence learning. They also suggest that distinguishing among several theories of sequence representation and acquisition may require us to analize the data in great detail. Hopefully, however, some truth can be found in such details.. (shrink)

According to Unconscious Thought Theory (Dijksterhuis & Nordgren, 2006), complex decisions are best made after a period of distraction assumed to elicit “unconscious thought”. Here, we suggest instead that the superiority of decisions made after distraction results from the fact that conscious deliberation can deteriorate impressions formed online during information acquisition. We found that participants instructed to form an impression made better decisions after distraction than after deliberation, thereby replicating earlier findings. However, decisions made immediately were just as good as (...) decisions made after distraction, which suggests (1) that people had already made their decision during information acquisition, (2) that deliberation-without-attention does not occur during distraction, and (3) that ruminating about one's first impression can deteriorate decision quality. Strikingly, in another condition that should have favored unconscious thought even more, deliberated decisions were better than immediate or distracted decisions. These findings were replicated in a field study. (shrink)

Time is a fundamental dimension of consciousness. Many studies of the “sense of agency” have investigated whether we attribute actions to ourselves based on a conscious experience of intention occurring prior to action, or based on a reconstruction after the action itself has occurred. Here, we ask the same question about a lower level aspect of action experience, namely awareness of the detailed spatial form of a simple movement. Subjects reached for a target, which unpredictably jumped to the side on (...) some trials. Participants expressed their expectancy of a target shift during the upcoming movement, pointed at the target as quickly and accurately as possible before returning to the start posiment to the target shift if required and reproduced the spatial path of the movement they had just made, as accurately as possible, to give an indication of their awareness of the pointing movement. We analysed the spatial disparity between the initial and the reproduced movements on those with a target shift. A negative disparity value, or undershoot, suggests that motor awareness merely reflects a sluggish record of coordinated motor performance, while a positive value, or overshoot, suggests that participants’ intention to point to the shifting target contributes more to their awareness of action than their actual pointing movement. Undershoot and overshoot thus measure the reconstructive and the preconstuctive aspects of action awareness, respectively. We found that trials on which subjects strongly expected a target shift showed greater overshoot and less undershoot than trials with lower expectancy. Conscious expectancy therefore strongly influences the experience of the detailed motor parameters of our actions. Further, a delay inserted either between the expectancy judgement and the pointing movement, or between the pointing movement and the reproduction of the movement, had no effect on visuomotor adjustment but strongly influenced action awareness. Delays during either interval boosted undershoots, suggesting increased reliance on a time-limited sensory memory for action. The experience of action is thus strongly influenced by prior thoughts and expectations, but only over a short time period. Thus, awareness of our actions is a dynamic and relatively flexible mixture of what we intend to do, and what our motor system actually does. (shrink)

Change blindness—our inability to detect changes in a stimulus—occurs even when the change takes place gradually, without any disruption [Simons, D. J., Franconeri, S. L., & Reimer, R. L. . Change blindness in the absence of a visual disruption. Perception, 29, 1143–1154]. Such gradual changes are more difficult to detect than changes that involve a disruption. Using this method, David et al. [David, E., Laloyaux, C., Devue, C., & Cleeremans, A. . Change blindness to gradual changes in facial expressions. (...) Psychologica Belgica] recently showed substantial blindness to changes that involve facial expressions of emotion. In this experiment, we show that people who failed to detect any change in the displays were nevertheless influenced by the changing information in subsequent recognition decisions about which facial expression they had seen, and that their confidence in their decisions was lower after exposure to changing vs. static displays. The findings therefore support the notion that undetected changes that occur in highly salient stimuli may be causally efficacious and influence subsequent behavior. Implications concerning the nature of the representations associated with undetected changes are discussed. (shrink)

Subliminal perception occurs when prime stimuli that participants claim not to be aware of nevertheless influence subsequent processing of a target. This claim, however, critically depends on correct methods to assess prime awareness. Typically, d9 (‘‘d prime’’) tasks administered after a priming task are used to establish that people are unable to discriminate between different primes. Here, we show that such d9 tasks are influenced by the nature of the target, by attentional factors, and by the delay between stimulus presentation (...) and response. Our results suggest that the standard d9 task is not a straightforward measure of prime visibility. We discuss the implications of our findings for subliminal perception research. (shrink)

Time plays a central role in consciousness, at different levels and in different aspects of information processing. Subliminal perception experiments demonstrate that stimuli presented too briefly to enter conscious awareness are nevertheless processed to some extent. Implicit learning, implicit memory, and conditioning studies suggest that the extent to which memory traces are available for verbal report and for cognitive control is likewise dependent on the time available for processing during acquisition. Differences in the time available for processing also determine not (...) only the extent to which one becomes conscious of action, but also provides the basis for making attributions of authorship to experienced acts. In this paper, we offer a brief overview of these different findings and suggest that they can all be understood based on the fact that consciousness takes time. From this perspective, the availability of representations to conscious awareness depends on the quality of these representations — the extent to which they are strong, stable in time, and distinctive. High-quality representations occur when processes of global competition have had sufficient time to operate so as to make the system settle into the best possible interpretation of the input. Such processes implement global constraint satisfaction and critically depend on reentrant processing, through which representations can be further enriched by high-level constraints. We discuss these ideas in light of current theories of consciousness, emphasizing the fact that consciousness should be viewed as a process rather than as a static property associated with some states and not with others. (shrink)

Change blindness—our inability to detect changes in a stimulus—occurs even when the change takes place gradually, without disruption (Simons et al., 2000). Such gradual changes are more difficult to detect than changes that involve a disruption. In this experiment, we extend previous findings to the domain of facial expressions of emotions occurring in the context of a realistic scene. Even with changes occurring in central, highly relevant stimuli such as faces, gradual changes still produced high levels of change blindness: Detection (...) rates were three times lower for gradual changes than for displays involving disruption, with only 15% of the observers perceiving the gradual change within a single trial. However, despite this high rate of change blindness, changes on faces were significantly better detected than color changes occurring on non facial objects in the same scene. (shrink)

In their comment on Sandberg, Timmermans, Overgaard, and Cleeremans , Dienes and Seth argue that increased sensitivity of the Perceptual Awareness Scale is a consequence of the scale being less exclusive rather than more exhaustive. According to Dienes and Seth, this is because PAS may measure some conscious content, though not necessarily relevant conscious content, “If one saw a square but was only aware of seeing a flash of something, then one has not consciously seen a square.” In this (...) reply, we claim that there is a difference between conscious visual experience, which may be partial, and the resulting conscious content, which is conceptual. Whereas PAS measures the first, confidence judgments and post-decision wagering measure the second. (shrink)

Research on consciousness is currently enjoying a spectacular revival of interest in the cognitive sciences. From an empirical point of view, the NCC program — the search for the “Neural Correlates of Consciousness” — holds the promise of establishing correlations between physiological and phenomenal states in a way that directly resembles G. T. Fechner´s (1860) so-called “inner psychophysics”. Should the NCC program be entirely successful, we would thus be able to predict phenomenal states based on physiological states. we would be (...) able to predict phenomenal states based on physiological states. In this paper, we explore some of the conceptual and methodological difficulties of this approach. In both neurobiology and psychology, there are serious measurement problems that stand in the way of correlation research, even after the “hard problem” has been set aside. Thus, even if one had identified certain internal functional states as indicators of phenomenal states, the empirical psychologist would still be confronted with fundamental problems, such as determining the absence or presence of these functional states. In this respect, philosophy of science may help and provide a metatheoretical framework for the current interdisciplinary project. (shrink)

Through the use of double task conditions, the sequence learning (SL) paradigm offers unique opportunities to study the relationships between learning and attention. In their original study, Nissen & Bullemer (1987) argued that a secondary tone-counting task prevents SL because it exhausts participants’ attentional resources. Other authors have instead suggested that the detrimental effects of tone-counting are due to scheduling conflicts between performing the main and secondary tasks rather than to attentional load. Frensch & Miner (1994), for instance, suggested that (...) the secondary task impairs sequence learning because it lengthens the response-to-stimulus interval (RSI) and hence makes it less likely for relevant contingencies to be represented together in short-term memory, — a condition for learning. Stadler (1995), on the other hand, argued that the secondary task introduces variability in the RSI and disrupts the organization of the sequence into chunks. Further, according to Willingham, Greenberg & Cannon Thomas (1997) manipulation of the RSI influences performance but not sequence learning.. (shrink)

Maud Boyer Arnaud Destrebecqz Axel Cleeremans.

68 words Main Text: 1256 words References: 192 words Total Text: 1516 words.

A significant part of everyday learning occurs incidentally — a process typically described as implicit learning. A central issue in this domain and others, such as language acquisition, is the extent to which performance depends on the acquisition and deployment of abstract rules. Shanks and colleagues [22], [11] have suggested (1) that discrimination between grammatical and ungrammatical instances of a biconditional grammar requires the acquisition and use of abstract rules, and (2) that training conditions — in particular whether instructions orient (...) participants to identify the relevant rules or not — strongly influence the extent to which such rules will be learned. In this paper, we show (1) that a Simple Recurrent Network can in fact, under some conditions, learn a biconditional grammar, (2) that training conditions indeed influence learning in simple auto-associators networks and (3) that such networks can likewise learn about biconditional grammars, albeit to a lesser extent than human participants. These findings suggest that mastering biconditional grammars does not require the acquisition of abstract rules to the extent implied by Shanks and colleagues, and that performance on such material may in fact be based, at least in part, on simple associative learning mechanisms. (shrink)

Axel Cleeremans Research Associate of the National Fund for Scientific Research (Belgium) Laboratoire de Psychologie Industrielle et Commerciale Université Libre de Bruxelles CP 122 Avenue F.-D. Roosevelt, 50 1050 Bruxelles..

Comparing the relative sensitivity of direct and indirect measures of learning is proposed as the best way to provide evidence for unconscious learning when both conceptual and operative definitions of awareness are lacking. This approach was first proposed by Reingold & Merikle (1988) in the context of subliminal perception. In this paper, we apply it to a choice reaction time task in which the material is generated based on a probabilistic finite-state grammar (Cleeremans, 1993). We show (1) that participants (...) progressively learn about the statistical structure of the stimulus material over training with the choice reaction time task, and (2) that they can use some of this knowledge to predict the location of the next stimulus in a subsequent “generation” task. However, detailed partial correlational analyses of the correspondence between performance during the reaction time task and the statistical structure of the training material showed that large effects remained even when controlling for explicit knowledge as assessed by the generation task. Hence we conclude (1) that at least some of the knowledge expressed through reaction time performance can not be characterized as conscious, and (2) that even when associations are found at a global level of analysis, dissociations can still be obtained when more detailed analyses are conducted. Finally, we also show that participants are limited in the depth of the contingencies they can learn about, and that these limitations are shared by the Simple Recurrent Network model of Cleeremans & McClelland (1991). (shrink)

The goal of this study is to characterize observers’ abilities to discriminate between endogenous and exogenous changes. To do so, we developed a new experimental paradigm. On each trial, participants were shown a dot pattern on the screen. Next, the pattern disappeared and participants were to reproduce it. Changes were surreptuously introduced in the stimulus, either by presenting participants anew with the dot pattern they had themselves produced on the previous trial or by presenting participants with a slightly different dot (...) pattern . We analyzed awareness of the changes and behavioral adaptation to them in a dynamical manner. We observe signal attenuation in the presence of endogenous change, dissociation between self-attribution reports and behavioral effect of agency. We discuss the source of this sensitive attenuation as well as the relation between a minimal or core self and an extended, narrative or autobiographical self. (shrink)

Consider Susan Hurley's depiction of mainstream views of the mind: "The mind is a kind of sandwich, and cognition is the filling" (p. 401). This particular sandwich (with perception as the bottom loaf and action as the top loaf) tastes foul to Hurley, who devotes most of "Consciousness in Action" to a systematic and sometimes extraordinarily detailed critique of what has otherwise been dubbed "classical" models of the mind. This critique then provides the basis for her alternative proposal, in which (...) perception, action and environment are deeply intertwined. (shrink)

As linguistic competence so clearly illustrates, processing sequences of events is a fundamental aspect of human cognition. For this reason perhaps, sequence learning behavior currently attracts considerable attention in both cognitive psychology and computational theory. In typical sequence learning situations, participants are asked to react to each element of sequentially structured visual sequences of events. An important issue in this context is to determine whether essentially associative processes are sufficient to understand human performance, or whether more powerful learning mechanisms are (...) necessary. To address this issue, we explore how well human participants and connectionist models are capable of learning sequential material that involves complex, disjoint, longdistance contingencies. We show that the popular Simple Recurrent Network model (Elman, 1990), which has otherwise been shown to account for a variety of empirical findings (Cleeremans, 1993), fails to account for human performance in several experimental situations meant to test the model’s specific predictions. In previous research (Cleeremans, 1993) briefly described in this paper, the structure of center-embedded sequential structures was manipulated to be strictly identical or probabilistically different as a function of the elements surrounding the embedding. While the SRN could only learn in the second case, human subjects were found to be insensitive to the manipulation. In the new experiment described in this paper, we tested the idea that performance benefits from “starting small effects” (Elman, 1993) by contrasting two conditions in which the training regimen was either incremental or not. Again, while the SRN is only capable of learning in the first case, human subjects were able to learn in both. We suggest an alternative model based on Maskara & Noetzel’s (1991) Auto-Associative Recurrent Network as a way to overcome the SRN model’s failure to account for the empirical findings.. (shrink)

The constructivist notion that features are purely functional is incompatible with the classical computational metaphor of mind. I suggest that the discontent expressed by Schyns, Goldstone and Thibaut about fixed-features theories of categorization reflects the growing impact of connectionism, and show how their perspective is similar to recent research on implicit learning, consciousness, and development. A hard problem remains, however: How to bridge the gap between subsymbolic and symbolic cognition.

The goal of computational cognitive neuroscience is to understand how the brain embodies the mind by using biologically based computational models comprised of networks of neuronlike units. This text, based on a course taught by Randall O'Reilly and Yuko Munakata over the past several years, provides an in-depth introduction to the main ideas in the field. The neural units in the simulations use equations based directly on the ion channels that govern the behavior of real neurons and the neural networks (...) incorporate anatomical and physiological properties of the neocortex. Thus the text provides the student with knowledge of the basic biology of the brain as well as the computational skills needed to simulate large-scale cognitive phenomena. (shrink)

While the study of implicit learning is nothing new, the field as a whole has come to embody — over the last decade or so — ongoing questioning about three of the most fundamental debates in the cognitive sciences: The nature of consciousness, the nature of mental representation (in particular the difficult issue of abstraction), and the role of experience in shaping the cognitive system. Our main goal in this chapter is to offer a framework that attempts to integrate current (...) thinking about these three issues in a way that specifically links consciousness with adaptation and learning. Our assumptions about this relationship are rooted in further assumptions about the nature of processing and of representation in cognitive systems. When considered together, we believe that these assumptions offer a new perspective on the relationships between conscious and unconscious processing and on the function of consciousness in cognitive systems. (shrink)