Neuronal models for conditioned reflexes

The classification of modifiable synapses and their use in models for conditioning

Modifiable synapses are analysed theoretically. It is proved that they fall into three classes (called A, B and C), such that any two members of the same class can, with the aid of non-remembering elements that perform simple logical operations, replace one another in any net; but a member of class A cannot replace one of class B or C. A member of class B can replace one of class A or C only if non-logical elements, for example noise generators, are included in the net. Models are designed that use modifiable synapses and show the principal features of classical and operant conditioning. If the models are to be given the property of extinction, classical conditioning requires modifiable synapses of classes A and B or of class C, and operant conditioning of class B or C. It is argued that if the conditioning and memory mechanisms of the nervous system store information by means of modifiable synapses, some of these must be of class B or C.

Neural dynamics of attentionally modulated Pavlovian conditioning: blocking, interstimulus interval, and secondary reinforcement

Selective information processing in neural networks is studied through computer simulations of Pavlovian conditioning data. The model reproduces properties of blocking, inverted-U in learning as a function of interstimulus interval, anticipatory conditioned responses, secondary reinforcement, attentional focusing by conditioned motivational feedback, and limited capacity short-term memory processing. Conditioning occurs from sensory to drive representations (conditioned reinforcer learning), from drive to sensory representations (incentive motivational learning), and from sensory to motor representations (habit learning).The conditionable pathwas contain long-term memory traces that obey a non-Hebbian associative law. The neural model embodies a solution to two key design problems of conditioning, the synchronization and persistence problems. This model of vertebrate learning is compared with data and models of invertebrate learning. Predictions derived from models of vertebrate learning are compared with data about invertebrate learning, including data from Aplysia about facilitator neurons and data from Hermissenda about voltage-dependent Ca(2+) currents. A prediction is stated about classical conditioning in all species, called the secondary conditioning alternative, and if confirmed would constitute an evolutionary invariant of learning.

Molecular biology of learning: modulation of transmitter release

Until recently, it has been impossible to approach learning with the techniques of cell biology. During the past several years, elementary forms of learning have been analyzed in higher invertebrates. Their nervous systems allow the experimental study of behavioral, neurophysiological, morphological, biochemical, and genetic components of the functional (plastic) changes underlying learning. In this review, we focus primarily on short-term sensitization of the gill and siphon reflex in the marine mollusk, Aplysia californica. Analyses of this form of learning provide direct evidence that protein phosphorylation dependent on cyclic adenosine monophosphate can modulate synaptic action. These studies also suggest how the molecular mechanisms for this short-term form of synaptic plasticity can be extended to explain both long-term memory and classical conditioning.

Simulation of anticipatory responses in classical conditioning by a neuron-like adaptive element

A neuron-like adaptive element is described that produces an important feature of the anticipatory nature of classical conditioning. The response that occurs after training (conditioned response) usually begins earlier than the reinforcing stimulus (unconditioned stimulus). The conditioned response therefore usually anticipates the unconditioned stimulus. This aspect of classical conditioning has been largely neglected by hypotheses that neurons provide single unit analogs of conditioning. This paper briefly presents the model and extends earlier results by computer simulation of conditioned inhibition and chaining of associations.

Classical conditioning of kindled seizures

As a model of reflex epilepsy, rats were classically conditioned to show paroxysmal spike activity to an auditory stimulus. A tone stimulus was paired with seizures induced by kindling. After repeated pairings, paroxysmal discharges were elicited at tone onset. The duration of the tone, proximity to the stimulation, and the number of pairings appeared to be important in the conditioning process.

Neural modeling of human behavior: a proposal for a new type of modifiable synapse

Presently published neural models for the stimulation of human behavior have reached a limit in terms of their ability to account for the complexity and magnitude of such behavior. In this paper, these limitations are explored, and a new model is presented which overcomes them. The model employs a new type of modifiable synapse which allows for the integration of three separate bits of information rather than the conventional two bits.

Neuronal circuits involved in learning: the use of modifiable synapses in the simulation of behavior

Our present knowledge of brain structure and function requires that any model of neuronal circuitry designed to account for learning must satisfy three conditions. It must (1) meet economy restrictions on the number of cells in the brain, (2) use the same set of cells to account for a number of different behaviors, and (3) not require a detailed embryological specification of its connections. Previously published models have failed to meet one or more of these conditions. In this paper, a model is presented which does satisfy them, and in doing so accounts in detail for classical conditioning, operant conditioning, and other learning tasks. The model employs the types of synapses proposed by Burke and Hebb in simple modular circuits as a means of providing independent storage of information at each modifiable synapse.

Nerve net models of plausible size that perform many simple learning tasks

Nerve net models are designed that can learn up to γ of α n possible sequences of stimuli, where γ is large but α n much larger still. The models proposed store information in modifiable synapses. Their connexions need to be specified only in a general way, a large part being random. They resist destruction of a good many cells. When built with Hebb synapses (or any other class B or C synapses whose modification depends on the conjunction of activities in two cells) they demand a number of inputs to each cell that agrees well with known anatomy. The number of cells required, for performing tasks of the kind considered as well as the human brain can perform them, is only a small fraction of the number of cells in the brain. It is suggested that the models proposed are likely to be the most economical possible for their tasks, components and constructional constraints, and that any others that approach them in economy must share with them certain observable features, in particular an abundance of cells with many independent inputs and low thresholds.

Disuse in the lateral geniculate nucleus of the cat

1. An attempt has been made to produce disuse in the lateral geniculate nucleus (LGN) of the cat, in the synapses between the optic nerve fibres and the principal cells of the nucleus. Evidence is produced that destruction of the visual receptor cells by iodoacetate or 1,5-di(p-aminophenoxy) pentane dihydrochloride effectively silences the optic nerve discharge and so achieves this result.2. Disuse of the LGN synapses does not cause any decrease in synaptic efficiency. The LGN response to a single stimulus to the optic nerve was not appreciably altered, and the depression which normally follows single or repeated stimuli was much reduced or absent. This increased responsiveness was unaffected by prolonged tetanization of the optic nerve during the experiment.3. Two possible explanations of the increased responsiveness are suggested: a post-synaptic ;decentralization'-hypersensitivity and an increased output of transmitter per impulse. The relevance of these results to theories of learning is discussed.4. The LGN response of adult cats kept in complete darkness for periods of up to 966 days was not appreciably different from that in the normal cat and there was little or no increased responsiveness. This suggests that many retinal ganglion cells continue to discharge in total darkness for long periods. There is a possibility that disuse may develop after a long time in the dark.