1 Cognitive Architectures 2 Multilayer Perceptrons 3 Relations between Variables 4 Structured Representations 5 Individuals 6 Where does the Machinery of Symbol Manipulation Come From? 7 Conclusions.

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.

Naturally occurring speech contains only a limited amount of complex recursive structure, and this is reflected in the empirically documented difficulties that people experience when processing such structures. We present a connectionist model of human performance in processing recursive language structures. The model is trained on simple artificial languages. We find that the qualitative performance profile of the model matches human behavior, both on the relative difficulty of center‐embedding and cross‐dependency, and between the processing of these complex recursive structures and (...) right‐branching recursive constructions. We analyze how these differences in performance are reflected in the internal representations of the model by performing discriminant analyses on these representations both before and after training. Furthermore, we show how a network trained to process recursive structures can also generate such structures in a probabilistic fashion. This work suggests a novel explanation of people's limited recursive performance, without assuming the existence of a mentally represented competence grammar allowing unbounded recursion. (shrink)

I examine the phenomenon of implicit learning, the process by which knowledge about the rule-governed complexities of the stimulus environment is acquired independently of conscious attempts to do so. Our research with the two seemingly disparate experimental paradigms of synthetic grammar learning and probability learning, is reviewed and integrated with other approaches to the general problem of unconscious cognition. The conclusions reached are as follows: Implicit learning produces a tacit knowledge base that is abstract and representative of the structure of (...) the environment; such knowledge is optimally acquired independently of conscious efforts to learn; and it can be used implicitly to solve problems and make accurate decisions about novel stimulus circumstances. Various epistemological issues and related problems such as intuition, neuroclinical disorders of learning and memory, and the relationship of evolutionary processes to cognitive science are also discussed. (shrink)

How people learn chunks or associations between adjacent items in sequences was modelled. Two previously successful models of how people learn artificial grammars were contrasted: the CCN, a network version of the competitive chunker of Servan‐Schreiber and Anderson [J. Exp. Psychol.: Learn. Mem. Cogn. 16 (1990) 592], which produces local and compositionally‐structured chunk representations acquired incrementally; and the simple recurrent network (SRN) of Elman [Cogn. Sci. 14 (1990) 179], which acquires distributed representations through error correction. The models' susceptibility to two (...) types of interference was determined: prediction conflicts, in which a given letter can predict two other letters that appear next with an unequal frequency; and retroactive interference, in which the prediction made by a letter changes in the second half of training. The predictions of the models were determined by exploring parameter space and seeing howdensely different regions of the space of possible experimental outcomes were populated by model outcomes. For both types of interference, human data fell squarely in regions characteristic of CCN performance but not characteristic of SRN performance. (shrink)

The dominant theory of what people can learn implicitly is that they learn chunks of adjacent elements in sequences. A type of musical grammar that goes beyond specifying allowable chunks is provided by serialist or 12‐tone music. The rules constitute operations over variables and could not be appreciated as such by a system that can only chunk elements together. A series of studies investigated the extent to which people could implicitly (or explicitly) learn the structures of serialist music. We found (...) that people who had no background in atonal music did not learn the structures, but highly selected participants with an interest in atonal music could implicitly learn to detect melodies instantiating the structures. The results have implications for both theorists of implicit learning and composers who may wish to know which structures they put into a piece of music can be appreciated. (shrink)

Page argues that localist models can be applied to a number of problems that are difficult for distributed models. However, it is easy to find examples where the opposite is true. This commentary illustrates the superiority of distributed models in the domain of artificial grammar learning, a paradigm widely used to investigate implicit learning.

Subjects exposed to strings of letters generated by a finite state grammar can later classify grammatical and nongrammatical test strings, even though they cannot adequately say what the rules of the grammar are (e.g., Reber, 1989). The MINERVA 2 (Hintzman, 1986) and Medin and Schaffer (1978) memory‐array models and a number of connectionist outoassociator models are tested against experimental data by deriving mainly parameter‐free predictions from the models of the rank order of classification difficulty of test strings. The importance of (...) different assumptions regarding the coding of features (How should the absence of a feature be coded? Should single letters or digrams be coded?), the learning rule used (Hebb rule vs. delta rule), and the connectivity (Should features be predicted only by previous features in the string, or by all features simultaneously?) is investigated by determining the performance of the models with and without each assumption. Only one class of connectionist model (the simultaneous delta rule) passes all the tests. It is shown that this class of model can be regarded by abstracting a set of representative but incomplete rules of the grammar. (shrink)

This paper shows how a neural network can model the way people who have acquired knowledge of an artificial grammar in one perceptual domain (e.g., sequences of tones differing in pitch) can apply the knowledge to a quite different perceptual domain (e.g., sequences of letters). It is shown that a version of the Simple Recurrent Network (SRN) can transfer its knowledge of artificial grammars across domains without feedback. The performance of the model is sensitive to at least some of the (...) same variables that affect subjects' performance—for example, the model is responsive to both the grammaticality of test sequences and their similarity to training sequences, to the cover task used during training, and to whether training is on bigrams or larger sequences. (shrink)

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)

A significant part of everyday learning occurs incidentally — a process typically described as implicit learning. A central issue in this and germane domains such as language acquisition is the extent to which performance depends on the acquisition and deployment of abstract rules. In an attempt to address this question, we show that the apparent use of such rules in a simple categorisation task of artificial grammar strings, as reported by Shanks, Johnstone, and Staggs (1997), can be simulated by means (...) of a simple recurrent network, and may thus turn out not be incompatible with the acquisition of statistical regularities rooted in the processing of exemplars of the presented material. (shrink)

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)

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)