A possible neural mechanism underlying consciousness based on the pattern processing capabilities of pyramidal neurons in the cerebral cortex.

This paper examines a possible neural mechanism underlying the phenomenon of consciousness by exploring the ability of cortical pyramidal cells to process patterns of information. A numerical model of a section of a neuron is described that enables the transient membrane potential distribution to be predicted following inputs to the cell. The neuron model incorporates learning by modelling the ability of dendritic spine receptors to undergo changes in their sensitivity if they receive inputs when their local membrane is depolarized. Simulations show the cell to be capable of recognizing patterns amongst its inputs, and to be able to extend its repertoire of learnt patterns by associating one pattern of inputs with another. These pattern processing capabilities can take place within the apical dendritic tree, whilst the soma sees a much attenuated and slower response that reflects the general level of pattern recognition taking place in the dendrites. It is argued here that the distribution of patterns stored in the apical dendrites of all the pyramidal cells represents the cortical knowledge base. Incoming patterns of information are compared with those stored in the cortical knowledge base to pick out components that have been experienced before. The resulting distribution of soma responses represents the way the incoming patterns of information are perceived. Most of the patterns learnt arise from other pyramidals and represent information about cortical behaviour itself. The distribution of soma responses therefore represents a perception of the self as well as a perception of the environment. It is argued here that this intimate perception of the self could underlie the phenomenon of consciousness.

Computer simulation of neurone pattern processing.

Physiological simulation techniques can provide a very useful additional tool with which to explore the behaviour of neurones, especially for probing those areas that are not easily accessible to experimental monitoring techniques such as activity in the dendrites. This paper explores the information processing capabilities of a cerebral cortical pyramidal neurone using a computer simulation. The computer model simulated two important aspects of pyramidal cell behaviour: the transient distribution of membrane potential over a proportion of the apical dendritic tree of the cell, and the ability of the receptors in the neurones' dendritic spines to undergo plastic changes. The simulation of receptor behaviour modelled the ability of receptor regions to increase their sensitivity if a prior input had occurred when the spine head membrane was depolarized. The model was used to explore the response of the cell to different inputs. It underlined the need for large amounts of local activity before spinal receptors changed their sensitivity. It also demonstrated the ability of pyramidal cells to both recognize patterns of input and to associate one pattern of inputs with another. It is argued that this pattern association property could be a neuronal basis for association learning.

Clinical experience in rehabilitation robotics.

A robotic workstation system for the disabled, based on a commercially available arm, was tested with six patients at the Spinal Injuries Unit, Odstock Hospital, Salisbury. A questionnaire was administered to those who used the system. Users evaluated the usefulness and performance of the system and commented on their reactions to the use of robots in rehabilitation. The users were generally favourable as regards the ease of use of the system using a two-switch input, operating a scanning menu. All users wanted the robot to be able to replay previously created routines, and the majority also wanted to be able to directly control the robot as well. The users were unsure about the potential usefulness of the system. Because a robot is by definition a flexible device, the context in which it is introduced will effect the way it is received by potential users. Tests in a hospital environment are useful because there is a high concentration of users in their own home situations will give a better idea of the usefulness of such devices. The system was not ideal from the point of visibility and layout, and was too large for use in a domestic environment. The layout was largely dictated by the geometry of the manipulator. Therefore a new workstation system has been constructed using a purpose built manipulator. This new system particularly aims to overcome the poor layout of the earlier workstation and benefits from feedback from users.

Development of a robot arm and workstation for the disabled.

A robotic workstation for the severely physically disabled is being developed. The prototype system consists of a commercially available arm mounted in a workstation set up for various manipulative tasks. This system has been tested with eight disabled users in a hospital situation and is to undergo evaluation in the homes of disabled users. Based on experience with this system, a new arm has been designed and will be built into a redesigned workstation.

Mechanisms of association learning in the post-synaptic neurone.

A general mechanism for association learning in neurones is presented based on the biochemistry and electrophysiology of synaptic receptor regions. The mechanism involves the voltage sensitive generation of second messengers in the post-synaptic neurone, and their subsequent influence on the sensitivity of the local receptor region. The mechanism is examined in detail using a computer model of the reaction dynamics of two important synaptic receptors; the adrenergic receptor and the NMDA-type glutamate receptor. The computer model simulates concentration changes of several molecules in the post-synaptic neurone following an input to a receptor. It also models changes in membrane potential caused by the cell firing, and allows for the impact of these potential changes on second messenger generation within the cell. Finally the model incorporates two possible mechanisms whereby high concentrations of second messenger can lead to long-lasting changes in receptor sensitivity. The model demonstrates an enhancement of synaptic response when previous inputs to the same receptor region have occurred when the cell was firing. Using the model, long term potentiation (LTP) of the glutamate response is demonstrated following a high frequency input to the glutamate receptor. A neurone with ten glutamate receptors is simulated and demonstrates that individual neurones should be capable of recognising patterns of input activity when those patterns have repeatedly occurred at times of major cellular activity.

Design methodology for aids for the disabled.

Although many aids for the disabled appear to be quite simple, their design is subject to many difficulties. These problems arise because of the intimate relationship between the equipment and the human body. Equipment design is therefore affected by many biological variables that are difficult to define at the beginning of the design exercise and many will not even be obvious until a device is tried out with a user. Standard design methodologies rely on all the variables being defined in a thorough specification at the start of the design process and therefore often lead to ineffective devices or very long development programmes. An alternative methodology is presented which overcomes the problems by separating the user interface aspects of the design from the supporting features. The user interface aspects are then allowed to evolve in conjunction with tests with potential users and finally integrated with the supporting features once a satisfactory solution has been found. Much more effective devices result with much shorter development effort.

Basic module for an adaptive control system based on neurone information processing.

The design of such devices as robotic aids for handicapped people, powered prostheses and manipulative aids such as page turners would benefit from the use of an adaptive control system. Much recent work on adaptive networks has been based on simplified models of the information processing capabilities of neurones. Neurones are now known to be capable of association learning and memory and this study incorporates these features in a neurone model. A single neuronal input system, the NMDA-type glutamate receptor, is modelled by deriving finite difference equations from its reaction dynamics so that the concentration of several molecules in the receptor can be plotted as a function of time. The model shows association learning taking place at the glutamate receptor. A whole neurone with ten glutamate receptor regions is also modelled and shows that a neurone should be capable of recognizing patterns of inputs. As the neurone model is complicated and slow to run, a much simplified form of the model is described which embodies the basic features of neurone information processing in a simple algorithm.