Warren S. McCulloch

What the Frog's Eye Tells the Frog's Brain

In this paper, we analyze the activity of single fibers in the optic nerve of a frog. Our method is to find what sort of stimulus causes the largest activity in one nerve fiber and then what is the exciting aspect of that stimulus such that variations in everything else cause little change in the response. It has been known for the past 20 years that each fiber is connected not to a few rods and cones in the retina but to very many over a fair area. Our results show that for the most part within that area, it is not the light intensity itself but rather the pattern of local variation of intensity that is the exciting factor. There are four types of fibers, each type concerned with a different sort of pattern. Each type is uniformly distributed over the whole retina of the frog. Thus, there are four distinct parallel distributed channels whereby the frogs eye informs his brain about the visual image in terms of local pattern independent of average illumination. We describe the patterns and show the functional and anatomical separation of the channels. This work has been done on the frog, and our interpretation applies only to the frog.

How we know universals: the perception of auditory and visual forms

Two neural mechanisms are described which exhibit recognition of forms. Both are independent of small perturbations at synapses of excitation, threshold, and synchrony, and are referred to partiular appropriate regions of the nervous system, thus suggesting experimental verification. The first mechanism averages an apparition over a group, and in the treatment of this mechanism it is suggested that scansion plays a significant part. The second mechanism reduces an apparition to a standard selected from among its many legitimate presentations. The former mechanism is exemplified by the recognition of chords regardless of pitch and shapes regardless of size. The latter is exemplified here only in the reflexive mechanism translating apparitions to the fovea. Both are extensions to contemporaneous functions of the knowing of universals heretofore treated by the authors only with respect to sequence in time.

The heterarchy of values determined by the topology of nervous nets

Because of the dromic character of purposive activities, the closed circuits sustaining them and their interaction can be treated topologically. It is found that to the value anomaly, whenA is preferred toB,B toC, butC toA, there corresponds a diadrome, or circularity in the net which is not the path of any drome and which cannot be mapped without a diallel on a surface sufficient to map the dromes. Thus the apparent inconsistency of preference is shown to indicate consistency of an order too high to permit construction of a scale of values, but submitting to finite topological analysis based on the finite number of nervous cells and their possible connections.

A model of the vertebrate central command system

This paper is based on the hypothesis that the reticular formation (RF) is the structure invertebrates that commits an animal to one mode of behavior or another. Examples of modes are sleep, eat, drink, fight, flee and mate. There are never more than about 25 such modes for any given animal, and if properly interpreted, they are mutually exclusive. The problem of the RF is how, in a fraction of a second, its million or more neurons are able to reach a workable consensus as to the proper mode of total commitment. The RF neurology, which is essentially constant from frog to man, is reviewed on the assumption that it provides the main clues. In particular, the RF Golgi anatomy of the Scheibels, which reveals RF circuit patterns, is caricatured to produce a computer simulation model, S-RETIC. The design of S-RETIC and its satisfactory simulation are described. The model consists of a dozen probabilistic hybrid computer modules linked together with jumpers of different lengths to form an anastomotic array. This array is neither serial nor parallel. An enhanced S-RETIC, STC-RETIC, has also been simulated, and in addition to rolling from mode to mode as a proper function of its 84 binary inputs, it is capable of habituation, conditioning, extinction, generalization, and limited trialand-error discrimination. An enriched version of STC-RETIC is discussed which is designed to operate asynchronously and show appropriate endogenously influenced behavior. The place of an RF model in the functional organization of a complete android robot is outlined.

Evidence That Cut Optic Nerve Fibers in a Frog Regenerate to Their Proper Places in the Tectum

The frogs retina projects into the superficial neuropil of the opposite tectum in four functionally different layers of terminals. Each layer displays a continuous map of the retina in terms of its particular function. The four maps are in register. The fourth-dimensional order is reconstituted after section and regeneration of the optic fibers.

A logical calculus of the ideas immanent in nervous activity

Because of the “all-or-none” character of nervous activity, neural events and the relations among them can be treated by means of propositional logic. It is found that the behavior of every net can be described in these terms, with the addition of more complicated logical means for nets containing circles; and that for any logical expression satisfying certain conditions, one can find a net behaving in the fashion it describes. It is shown that many particular choices among possible neurophysiological assumptions are equivalent, in the sense that for every net behaving under one assumption, there exists another net which behaves under the other and gives the same results, although perhaps not in the same time. Various applications of the calculus are discussed.

A Statistical Consequence of the Logical Calculus of Nervous Nets

A formal method is derived for converting logical relations among the actions of neurons in a net into statistical relations among the frequencies of their impulses.

The brain computing machine

The author, a medical doctor, employs electrical engineering terminology to show how the brain may be likened to a digital computing machine consisting of ten billion relays called neurons. To carry the analogy further, the performance of the brain is governed by inverse feedback, subsidiary networks secure invariants, or ideas, predictive filters enable us to move toward the place where the object will be when we get there, and complicated servo-mechanisms enable us to act with facility and precision. Disorders of function are explained in terms of damage to the structure, improper voltage of the relays, and parasitic oscillations.

Effects of Strychnine with Special Reference to Spinal Afferent Fibres

“Thou art man and canst abide a truth Tho bitter” Tcnnyson In 1809 Magendie said in a lecture to the Institute of France: “an entire family of vegetables (the bitter strychnos) has the singular property of exciting strongly the spinal marrow.” Since that time, the dramatic property of strychnine of inducing convulsions has enjoyed the attention of many research workers. This interest was exaggerated by the develop ment of the technique of “strychnine neuronography” by Dusser de Barcnne. The localized convulsions generated in the region of locally applied strychnine were used to follow first-order pathways. Attempts have been made to locate those structures within the central nervous system on which strychniue acts and to understand the way in which the convulsim activity is generated. Much of this work has been reviewed by two major contributors to our knowledge of strychnine: Dusser de Barenne (22, 23) and Bremer (12, 13).

Some Mechanisms for a Theory of the Reticular Formation

Throughout the life of the vertebrates, the core of the central nervous system, sometimes called the reticular formation, has retained the power to commit the whole animal to one mode of behavior rather than another. Its anatomy, or wiring diagram, is fairly well known, but to date no theory of its circuit action has been proposed that could possibly account for its known performance. Its basic structure is that of a string of similar modules, wide but shallow in computation everywhere, and connected not merely from module to adjacent module, but by long jumpers between distant modules. Analysis of its circuit actions heretofore proposed in terms of finite automata or coupled nonlinear oscillators has failed.