Ichiro Tsuda

Original Contribution: Dynamic link of memory-Chaotic memory map in nonequilibrium neural networks

We present a model for a dynamic link of memory in terms of a self-organized chaotic transition in nonequilibrium neural networks. The chaotic transition blocks pinning on parasitic(false) memory and allows a successive retrieval of true memory. We introduced a new neurodynamics, which allows neural networks to be temporarily unstable, keeping stability due to convergent dynamics. The results obtained in this paper suggest that the cortical chaos may serve for dynamically linking true memory as well as a memory search.

Chaotic Itinerancy as a Dynamical Basis of Hermeneutics in Brain and Mind

Abstract We propose a new dynamical mechanism for information processing in mind and brain. We emphasize that a hermeneutic process is one of the key processes manifesting the functions of the brain and that it can be formulated as an itinerant motion in ultrahigh dimensional dynamical systems, which may give a new realm of the dynamic information processing. Our discussions are based on the notion of chaotic information processing and the observations of biological chaos.

Noise-induced order

A new noise effect on chaos in one-dimensional mappings is reported. The transition from chaotic behavior to ordered behavior induced by external noise is observed in a certain class of one-dimensional mappings. This transition is clearly shown in terms of the Lyapunov number, entropy, power spectrum, and the nature of orbits.

Memory Dynamics in Asynchronous Neural Networks

A model which can perform learning, formation of memory without teacher for successive memory recalls is presented. The philosophical background of the study is summarized. The investigated network consists of two sets both composed of asynchronously firing model neurons. One set of neurons is responsible for the field effect, and the other is introduced as an input/ output module. The field effect is given in the form of the systems self·response. It is shown that positive and negative global feedbacks by the field effect play an essential role in the successive recall of stored patterns. The possibility that these proposed mechanisms are implemented in the brain is discussed. We obtained a quasi-deterministic law on the level of a macrovariable concerning a random successive recall of memory representations by taking a Lorenz·plot of this macrovariable. We show that this macroscopic order is deterministic chaos steming from collapse of tori and this type of chaos can be an effective gadget for memory traces. § 1. General introduction -philosophical background of the study

Chaotic pulsation in human capillary vessels and its dependence on mental and physical conditions

We found a chaotic pulsation in a finger’s capillary vessels in both normal subjects and psychiatric patients, as well as cardiac chaos. A proof of chaos was made by the reconstruction of the dynam...

A Hermeneutic Process of the Brain

the brain. Concerning a pattern recognition, the difference between the functions of the brain and of the present artificial intelligence is discussed. The concept of metaconscious is introduced to explain the hermeneutic process. In this relation, it is suggested that the existence of a hidden dynamics is of importance for mans cognition.

Extended information in one-dimensional maps

Abstract An information theoretical structure of one-dimensional maps is investigated by regarding the maps as a “channel” rather than a “source”. Computing various mutual information numerically for one-dimensional dynamical systems, we can observe directly the movement of information as to initial conditions. One feature of information theoretical structure emerged from this computation is such that initially localized information spreads over whole system as time passes. Two groups of maps which are contrasting with each other concerning this feature exhibit a striking difference in their transmission ability of information when coupled to form a linear chain.

Reward prediction based on stimulus categorization in primate lateral prefrontal cortex

To adapt to changeable or unfamiliar environments, it is important that animals develop strategies for goal-directed behaviors that meet the new challenges. We used a sequential paired-association task with asymmetric reward schedule to investigate how prefrontal neurons integrate multiple already-acquired associations to predict reward. Two types of reward-related neurons were observed in the lateral prefrontal cortex: one type predicted reward independent of physical properties of visual stimuli and the other encoded the reward value specific to a category of stimuli defined by the task requirements. Neurons of the latter type were able to predict reward on the basis of stimuli that had not yet been associated with reward, provided that another stimulus from the same category was paired with reward. The results suggest that prefrontal neurons can represent reward information on the basis of category and propagate this information to category members that have not been linked directly with any experience of reward.

Can stochastic renewal of maps be a model for cerebral cortex

Abstract We introduce a new type of stochastic dynamics as stochastic renewal of maps, relating to the neurodynamics of cortical memory process. This stochastic dynamics can be reformulated by a skew product transformation of two kinds of variables, one of which describes an underlying dynamical system and the other describes chaotic dynamics, say, Bernoulli shift. The feature of orbits in phase space is investigated in the particular case of a neurodynamics model for cortical chaotic memories. A new computational result on the functional role of cortical chaos is obtained. We also present a neurobiological interpretation of psychological perception and memories by means of the notion of chaotic itinerancy.

A new type of self-organization associated with chaotic dynamics in neural networks.

A new type of self-organized dynamics is presented, in relation with chaos in neural networks. One is chaotic itinerancy and the other is chaos-driven contraction dynamics. The former is addressed as a universal behavior in high-dimensional dynamical systems. In particular, it can be viewed as one possible form of memory dynamics in brain. The latter gives rise to singular-continuous nowhere-differentiable attractors. These dynamics can be related to each other in the context of dimensionality and of chaotic information processings. Possible roles of these complex dynamics in brain are also discussed.