This paper describes how agents can learn an internal model of the world structurally by focusing on the problem of behavior-based articulation. We develop an on-line learning scheme -- the so-called mixture of recurrent neural net (RNN) experts -- in which a set of RNN modules becomes self-organized as experts on multiple levels in order to account for the different categories of sensory-motor flow which the robot experiences. Autonomous switching of activated modules in the lower level actually represents the articulation of the sensory-motor flow. In the meanwhile, a set of RNNs in the higher level competes to learn the sequences of module switching in the lower level, by which articulation at a further more abstract level can be achieved. The proposed scheme was examined through simulation experiments involving the navigation learning problem. Our dynamical systems analysis clarified the mechanism of the articulation; the possible correspondence between the articulation...

Synchrony among multicortical EEGs 2 Freeman, Gaál & Jörnsten Information transfer and integration among functionally distinct areas of cerebral cortex of oscillatory activity requires some degree of phase synchrony of the trains of action potentials that carry the information prior to the integration. However, propagation delays are obligatory. Delays vary with the lengths and conduction velocities of the axons carrying the information, causing phase dispersion. In order to determine how synchrony is achieved despite dispersion, we recorded EEG signals from multiple electrode arrays on five cortical areas in cats and rabbits, that had been trained to discriminate visual or auditory conditioned stimuli. Analysis by time-lagged correlation, multiple correlation and PCA, showed that maximal correlation was at zero lag and averaged.7, indicating that 50 % of the power in the gamma range among the five areas was at zero lag irrespective of phase or frequency. There were no stimulus-related episodes of transiently increased phase locking among the areas, nor EEG "bursts " of transiently increased amplitude above the sustained level of synchrony. Three operations were identified to account for the sustained correlation. Cortices broadcast their outputs over divergent-convergent axonal

prediction is made for clinical testing that wave packets are precursor to states of awareness. They are not by themselves accessible to experience, as may be the macroscopic states initiated by global state transitions. Keywords: EEG; meaning; mesoscopic neurodynamics; phase cone; state transition; wave packet. 1. Introduction Brain systems operate on many levels of organization, each with its own scales of time and space. Dynamics applies to every level from the atomic to the molecular, and from macromolecular organelles to the neurons that incorporate them. In turn neurons form populations, these form the subassemblies in brains, and so on to embodied brains interacting intentionally with material, interpersonal, and social environments. Each level is macroscopic to that below it and microscopic to that above it. Among the most di#cult tasks scientists face are those of conceiving and describing the exchanges between levels, seeing that the measures of time 3 and distance ar

One held that psychological functions such as language or memory could never be traced to a particular region of brain. If one had to accept, reluctantly, that the brain did produce the mind, it did so as a whole and not as a collection of parts with special functions. The other camp held that, on the contrary, the brain did have specialized parts and those pars generate separate mind functions.

Abstract – This paper describes our efforts to develop a robot with a sense of self using a multiagent-based cognitive architecture and control with three distinctive memory systems, namely (1) spatio-temporal short-term memory, (2) procedural / declarative / episodic long-term memory and (3) a task-oriented adaptive working memory based on psychological and computational neuroscience models. Such a robot may be called a cognitive robot. Cognitive robots share a number of key features with conscious machines. We explore the interface between cognitive robots and conscious machines through an internal model called the Self Agent. Index terms – cognitive robot, machine consciousness, self agent, adaptive working memory, cognitive control I.

This paper considers how fluent language users are rational in their language processing, their unconscious language representation systems optimally prepared for comprehension and production, how language learners are intuitive statisticians, and how acquisition can be understood as contingency learning. But there are important aspects of second language acquisition that do not appear to be rational, where input fails to become intake. The paper describes the types of situation where cognition deviates from rationality and it introduces how the apparent irrationalities of L2 acquisition result from standard phenomena of associative learning as encapsulated in the models of Rescorla and Wagner (1972) and Cheng and Holyoak (1995), which describe how cue salience, outcome importance, and the history of learning from multiple probabilistic cues affect the development of ‘learned selective attention’ and transfer. This article considers how fluent language users are rational in their language processing, rational in the sense that their unconscious language representation

Abstract: Consciousness is only marginally relevant to artificial intelligence (AI), because to most researchers in the field other problems seem more pressing. However, there have been proposals for how consciousness would be accounted for in a complete computational theory of the mind, from theorists such as Dennett, Hofstadter, McCarthy, McDermott, Minsky, Perlis, Sloman, and Smith. One can extract from these speculations a sketch of a theoretical synthesis, according to which consciousness is the property a system has by virtue of modeling itself as having sensations and making free decisions. Critics such as Harnad and Searle have not succeeded in demolishing a priori this or any other computational theory, but no such theory can be verified or refuted until and unless AI is successful in finding computational solutions of difficult problems such as vision, language, and locomotion. 1

Abstract not found

The belief that conscious will is merely ‘‘an illusion created by the brain’ ’ appears to be gaining in popularity among cognitive neuroscientists. Its main adherents usually refer to the classic, but controversial ‘Libet-experiments’, as the empirical evidence that vindicates this illusion-claim. However, based on recent work that provides other interpretations of the Libet-experiments, we argue that the illusion-claim is not only empirically invalid, but also theoretically incoherent, as it is rooted in a category mistake; namely, the presupposition that neuronal activity causes conscious will. We show that the illusion-claim is based on the behaviorist ‘input-output ’ paradigm, and discuss the notions of ‘self-organization ’ and ‘self-steering ’ to provide an alternative perspective on the causal efficacy of conscious will. In the final sections, a tentative theoretical picture is sketched of conscious will as an instance of self-steered self-organization. We conclude that the subjective experience of conscious will is not a misguided one, but rather that the mechanisms supporting conscious will are considerably more complex than mainstream cognitive neuroscience currently acknowledges. Keywords: Conscious Will; Benjamin Libet; Self-Organization 1.

Schizophrenia is a complex and heterogeneous disease, incorporating at least three distinct subsyndromes: psycho-motor poverty (poverty of speech, lack of spontaneous movement, blunting of affect), disorganisation (inappropriate affect, disturbances of the form of thought), and reality distortion (Liddle 1987, Johnstone 1991). The reality distortion syndrome encompasses the so