PubMed Health⌕ Search

Biomedical subjects

A M Uttley

Publications and source records attributed to A M Uttley.

10 recordsLinked to original sources
PubMed1976

A two-pathway informon theory of conditioning and adaptive pattern recognition.

A neural network theory is proposed which offers an explanation of many of the facts of classical and operant conditioning and adaptive pattern recognition. Interconnected networks of units have been studied and simulated which embody only two rules; firstly, units have inputs from pathways of variable and of fixed conductivity; secondly, the conductivity of a variable pathway is made proportional to the negative of the mutual information function between the signals at its input and output. The signal in a fixed pathway indicates whether the total input to the variable pathways is a member or not of some class. After a learning phase in which the unit, called an informon, receives such labelled inputs, it is able to predict the class of future unlabelled inputs. Such units are stable and their steady state can be calculated.

Conditioning, Psychological↗
PubMed1976

Simulation studies of learning in an informon network.

A single informon reproduces classical conditioning to compound stimuli under a wide range of conditions during reinforcement and non-reinforcement; it also reproduces simple operant conditioning. A network of 210 informons with 8400 simultaneouslly varying pathways has been simulated; it can learn to recognize handprinted numerals. The networks studied must be more richly interconnected before they can reproduce more detailed behaviour of animals.

Animals↗
PubMed1976

Neurophysiological predictions of a two-pathway informon theory of neural conditioning.

According to the informon theory there must be variable and fixed synapses in a neurone for conditioning to occur. For a variable synapse to behave like an informon pathway its conductivity needs to depend only on the average values of its presynaptic potential and of the internal state of the neurone. Eight predictions are made about the detailed functioning of such a synapse. In a minimal hypothesis all fixed synapses are inhibitory; but sign reversals are considered. Let one unit A in the receptive field of a neurone drive it through a fixed synapse, and all other units, e.g. B, drive variable synapses; then the theory predicts that the conductivity of the B synapse becomes proportional to the mutual information function between the signals at A and B; so inputs which tend to occur with the A signal become connected positively to the neurone. Applied to visual pathways this principle leads to the formation of edge and grating detectors. If X and Y cells excite variable and fixed synapses respectively, simple and complex cells should be driven by both X and Y cells, the latter being inhibitory. The two-pathway theory resolves two apparent conflicts between experimental facts.

Conditioning, Psychological↗
PubMed1968

Factors affecting the conductivity of pathways in the cerebral cortex.

1. We have investigated the conductivity of neural pathways in slabs of unanaesthetized, isolated, cerebral cortex, cut from the isolated forebrains of twenty-five cats.2. Neurones within the isolated area were indirectly excited, either by a small electrode thrust into the subcortical white matter, or by remote stimulation of the pial surface. Sometimes a small electrode was employed for intracortical stimulation.3. The response of single neurones to these stimuli was recorded with extracellular micropipettes. Submaximal stimuli produced a stochastic response which was measured from the post-stimulus histogram (PSH) and provided an estimate of the probability of discharge at various times after the stimulus.4. The PSH often displayed several discrete humps of different latencies, indicating several pathways between stimulated and recording point. Conductivity measurements were usually restricted to the pathway of shortest latency.5. The conductivity of a pathway was defined as C = xy/x, averaged over 1 or 2 min, wherex = frequency of afferent test volleys,xy = frequency of response: i.e. of those action potentials contained within a well-defined hump of the PSH,(y = frequency of all discharges of the recorded neurone).6. The progress of conductivity was tested with some constant form of cortical stimulation, repeated at regular intervals of 1-5 sec. Reliable results were obtained for twenty-six pathways subjected to thirty-eight experiments.7. Temporary alteration of x, y or xy (conditioning with Deltax, Deltay or Deltaxy), for a period of 6-25 min, often caused a subsequent change in the conductivity (DeltaC) of a pathway which sometimes attenuated with a time constant of about 10 min, but which could persist without detectable attenuation for 20-30 min.8. Conditioning periods less than 6 min rarely produced changes in conductivity; alterations of conductivity were more likely to be caused by conditioning periods longer than 17 min, than by periods of 6-16 min.9. Changes in conductivity were usually correlatednegatively with temporary changes in xy,negatively with temporary changes in x,positively with temporary changes in y.10. Nineteen of twenty-six pathways tested showed properties consistent with the formula [Formula: see text], whereG(1) lies between -1.0 and -0.16, with mean value -0.50,G(2) lies between 0 and 0.42, with mean value +0.12,G(3) lies between -0.61 and 0 with mean value -0.38,K is a coefficient which is usually different for each experiment. Four out of twenty-six pathways so tested provided results which did not fit this formula; three out of twenty-six pathways did not give adequate information.

Animals↗