The junctions shown in the INTRODUCTION TO THE STRUCTURES OF THE NEURAL SYSTEM all involve an Activa inserted between two electrolytic conduits and in contact with the surrounding external medium (possibly through a third conduit. It is the base terminal of the Activa that is always connected to the surrounding medium in these types of junction.

This basic circuit configuration can appear in at least six distinct operating modes. The mode depends on the specific values of the electrical circuit elements connected to the Activa.

The photoreceptor cell is not included in the following list because of its unique topology that includes multiple operating modes. It is discussed in detail in Section 4.7.

These modes are shown in the following figure.

It should be noted that only the synapse involves an active electrolytic semiconductor device, an Activa, that is external to the morphological boundary of a neuron. In all other cases, the Activa is found within the morphological envelope of the neuron. This fact has impeded the recognition of the operating mechanisms within the cells for a very long time. The actual identification of the location of the Activa requires the electon microscope operating at magnifications exceeding x125,000. At these levels the cytological details of the cell walls can be resolved and the precise spacing between the two conduit walls required for "transistor action" can be demonstrated.

The low impedance current connection--the Synapse

Frame A shows the basic connections to the Activa as found in a synapse. The base of the Activa is in direct contact with the ground potential formed by the surrounding electrolytic medium. The input current i_in is supplied by the plasma of the input conduit. The output current, i_out, flows into the plasma of the output conduit. By proper arrangement of the potentials on the three terminals of this configuration, the ratio of the output current to the input current is usually greater than 0.99 with a maximum of 1.000. There is virtually no signal current loss, or gain, at a synapse.

The synapse in a voltage divider circuit

Although there is a very high current transfer efficiency at a synapse, the actual current flow through the synapse is often determined by a more complex configuration. When connected to the next junction as shown in frame B, the two Activas form a voltage divider between V_in and the ground potential of the surrounding medium. As a result of this configuration, the voltage, V_e, is given by the ratio of the transfer impedance of the Activa within the synapse to the input impedance of the second Activa. These impedances are functions of the area of each active junction forming the individual Activa. As in the case of man-made active devices, the Activas can be binned into different classes according to their current carrying capacity. Although the synapse is not labeled in this frame, the second Activa is labeled a type AB. This type is typical of the active device in a bipolar cell.

The basic differencing circuit--the Horizontal and Amercine Cells

Frame C shows the impact of modifying the circuit of Frame A by adding an impedance in the lead from the base of the Activa to the surrounding medium and providing a second input connection to the base terminal. The effect of the impedance is to introduce a degree of negative internal feedback with respect to the signal introduced at the terminal marked V_in(1). The output voltage V_out remains of the same polarity as the input voltage but is of reduced amplitude. A signal introduced at the terminal V_in(2) is treated differently in this simple circuit. It can be reproduce as a higher voltage at the collector terminal but the polarity of the output voltage is reversed relative to the input signal. The result is a simple circuit with an output voltage proportional to the difference between the amplified values of the two input voltages. This is the key circuit found within the lateral cells (the Horizontal and Amercine Cells). A type AL Activa is found in these circuits. These devices are physically small and are unable to provide a large output current.

Each of the impedances, shown as a hollow rectangle in the figure contain a basic circuit element, usually a diode, in parallel with a current generator. The diode is formed by the asymmetrical electrical performance of the conduit lemma and the current generator is a result of the electrostenolytic process related to the Glutamate Cycle, the metabolic process providing power to each element of the neural system. These devices can also be represented by a diode in series with a voltage source in accordance with Thevinen's Theorem.

The driven pulse encoder--The Parasol Ganglion Cell

Frame D shows a larger capacity Activa of type AG in the same basic configuration as the bipolar cell of frame B but with two additional circuit elements. There is an impedance in the base to surrounding medium circuit, as in frame C, and there is an additional capacitance between the collector terminal and the medium.

The gray box in this and subsequent frames is used to represent a myelinated conduit. Such conduits are typical of an Axon used in a signal projection circuit. It is also typical of the input conduit of a Stellate Cell and the conduit found between Nodes of Ranvier. Such an axon is typically a cylindrical structure of high length to diameter ratio. Electrically, this structure exhibits a low capacitance and a low, but significant inductance. The real resistive components of such a structure are negligible at the frequencies of interest in the neural system. However, the combined capacitance and inductance create an equivalent input resistance and determine the transmission velocity of signals traveling along the conduit.

The negative internal feedback introduced by the base impedance and the reactance associated with the capacitance in the collector lead make the circuit conditionally unstable (the circuit is subject to oscillation). The conditional stability of the circuit is determined by the emitter to base voltage of the circuit.

If the emitter to base voltage is such as to restrict the flow of current from the emitter to the collector, the circuit will be stable. However, if the emitter to base voltage should change for an instant to support current flow from the emitter to the base, the circuit will attempt to amplify this current, will become unstable and produce a single very large pulse of current in the output circuit.

If the emitter to base voltage is maintained above the critical level following the first pulse, the circuit will proceed to oscillate at a frequency determined primarily by the circuit values of the capacitance and the base impedance and the voltage at the emitter terminal.

The above two paragraphs describe the operating mode of the Parasol Ganglion Cells. This mode is known as the driven pulse oscillator. It responds quickly to an initial stimulus and subsequently responds with a pulse train with pulse interval proportional to the amplitude of the stimulus. Cells of this type are used to transmit the luminance information of the visual system.

The pulses generated by the Parasol and Midget Ganglion Cells are one form of the pulses known as "Action Potentials" A slightly different form will be identified when discussing the Nodes of Ranvier.

The free running encoder--The Midget Ganglion Cell

Frame D can also be used to describe the operation of the Midget Ganglion Cell. This cell type is normally biased so that the emitter to base voltage always encourages current to flow from the emitter to the base. Under this condition, the circuit is unstable and will oscillate as described above. In this case, any external stimulus applied to the input terminal can have either of two effects. If the stimulus voltage causes an increase in the current through the emitter, the circuit will oscillate at a higher frequency (the interval between the pulses will be reduced). However, if the stimulus causes a reduction in the emitter to base voltage, the frequency of oscillation will be reduced (the interval between pulses will be extended).

The above mode of operation defines the Midget Ganglion Cell. It oscillates at a nominal frequency (nominal pulse spacing) under quiescent conditions and changes frequency of oscillation in response to a bipolar electrical input signal. Cells of this type are used to transmit the chrominance, polarization, and appearance information of the visual system.

The pulse regenerator circuit--the Node of Ranvier

If the circuit of frame D is modified only slightly, it becomes that used in the Node of Ranvier. These Nodes occur in both the extended axons of Ganglion Cells and in the axons of signal projection neurons. In this circuit, the capacitance associated with the emitter to surrounding medium and the capacitance associated with the collector to medium ar more nearly equal in size. This characteristic provides a more powerful and more symmetrical output pulse. The voltage between the emitter and base is always set to prevent oscillation of the circuit except when a stimulus is received at the input. If a brief voltage pulse is received, the circuit is driven into an oscillation that is limited to one pulse. Because of the size of the capacitances present, this pulse is large and powerful. However, the peak of this pulse occurs considerably later in time than the occurrance of the input pulse. This "regeneration delay" is the primary factor in determining the group velocity of the action potentials traveling along a nerve.

Because of the more nearly equal size capacitances in the Node of Ranvier, the output pulse waveform is usually more symmetrical than that of the initial Activa of a Ganglion Cell.

The pulse to analog signal recovery circuit--the Stelllate Cell

The same basic circuit configuration as shown in frame D can be used to recover the information associated with the pulse trains generated by the Ganglion Cells and regenerated by the Nodes of Ranvier. The Stellate Cell is shown in frame F. It most closely resembles the Ganglion Cell of frame D. It has a small capacitance associated with the emitter terminal and a large capacitance associated with the collector terminal.

The Stellate Cell, with its relatively large output capacitance, operates as a charge integrator. Each pulse received at the emitter results in a quantity of charge being placed on the capacitor. This charge can leak off of the capacitor with time through the impedance in the collector circuit. By establishing the proper time constant in the collector circuit, a signal very similar to that used to encode the action pulse stream at the Ganglion Cells can be recovered. By comparing the recovered potential with the ground potential of the medium, a monopolar signal is recovered representative of the original luminance signal. By comparing the recovered signal to a fixed reference, a bipolar signal is recovered that is representative of the original chrominance, polarization or appearance signal. This is the form described by the label G₂ in the figure. Both forms are fully described in the main text.