Biological Cybernetics By: Jay Barra Sean Cain. Biological Cybernetics An interdisciplinary medium for experimental, theoretical and application- oriented.

Biological Cybernetics

By: Jay Barra

Sean Cain

Biological Cybernetics

• An interdisciplinary medium for experimental, theoretical and application-oriented aspects of information processing in organisms.

– Sensory, motor, cognitive, and ecological phenomena

– Experimental studies of biological systems

Biological Cybernetics

• Topics Covered:

– Experimental studies of biological systems

• Quantitative modeling

– Computational, technical, or theoretical studies

• Understanding biological information processing

– Artificial implementation of biological information processing and self-organizing principles

Biological Cybernetics

• Performance and Function of systems:

– Communication between life sciences and technical or theoretical disciplines

• Neural Interactions a specific interest

Coupled Van der Pol Oscillators - A model of excitatory and inhibitory neural interactions

• Relate basic neural activities on the cellular level to:– The various observed electroencephalograms (EEG)– Potential differences in the cortex

• Interactions between and within excitatory and inhibitory neurons– Modeled by Wilson and Cowan

• EEG phenomena studied in relation to such models

• Simple models explaining the rhythmicity of the EEG– Based on Wilson and Cowan model

• A feedback loop through a third set of neurons – Lopes da Silva et al.• Positive and negative feedback loops with two excitatory and one inhibitory

subsets of neurons – Zetterberg et al.

• Van der Pol Oscillators:– Mutually coupled relaxation oscillators

– Done as a mathematical model for the electrical activity of the human and animal gastro-intestinal tract

• Reasonable to model EEG phenomena by a number of coupled oscillators

• Distribution of entrained oscillator frequencies can show the peak-dip shape of the EEG

• Phenomenological mathematical models:– Explain EEG data– Offer a guide for experiments on the EEG

Coupled Van der Pol Oscillators

Biological Systems

• Biological clocks consist of:– A group of interacting oscillators.– Circadian rhythms, biochemical oscillators,

pacemaker neurons, etc. – May be related to biological oscillators

• Interesting subject in biological systems:– The entrainment of oscillators under periodic forces– The entrainment of a number of interacting oscillators

Biological Systems

• Apart from biological systems– Nonlinear circuit theory

• Can be combined with biological systems for a better understanding of EEG phenomena

• Relation of physiological parameters to the coefficients in the circuits

Study Conclusions

• Hope to further the understanding of very generalized cortex-like tissue in conjunction with the EEG phenomena

System-Theoretical Analysis of the Clare Bishop Area (CBA) in the Cat

• Clare Bishop Area:

– Retinotypically organized cortical area of the cat brain

– Connected to a great variety of visual areas in a very complex way

Experimental Analysis

• Difficult because:

– The greater the distance from the retina, the more specific the signal combinations necessary to analyze the system become

– Feedback loops cannot be opened, unequivocal identification of CBA cell properties is impossible

– Nonlinear character has a great influence on signal processing

Experimental Analysis

• Circumvent Difficulties:– Specific signal combinations have been developed,

being restricted to:

• Deterministic and stochastic signals

• Determination of a number of cell properties

• Hypotheses on the function of the CBA are tested

Conclusions

• Only moving stimuli produce responses

• Reaction of Cells:

– Direction specific– Prefer the extrafoveal direction

• Systems existent in the CBA:

– Have different behavior for:• Space • Time• Amplitude

Study Conclusions

• CBA primary task:

– Orientation

• Exact localization of objects independent of velocity

• Treatment of objects in relation to background stimuli

Cybernetics of Limb Movement

• Determine stresses on joints during motion• Neural planing involved in mapping spatial

planning to move joint to new position– Example, hand reaches for an apple

• Hand is the extension, upper arm and fore arm movements are automtically mapped

Computations in controlling movement

• Joint Torque

• Masses of individual segments

• Effects of gravity and momentum

Conclusions

• Normal straight line motion of joints involves– Compensation for dynamic interaction at each

joint segment

Information Content of Texture Gradients

• Information gradients determine responses in stimuli sensory systems– Hairs on the back of the hand

• Touching one hair may not be felt, touching three will

Visual gradients

• Non-linear distribution gives interpretation of horizons in a 2D image– This is why

perspective works in drawings

Visual Gradients

Clear Horizon Horizon diminishes

Thresholds for activation

• Affected by regularity of the information gradient

• Global Uniformity• Shape containing the pattern• Foreshortening (perspective)

Pattern types

• Locally regular and globally uniform– Field of flowers, cars in parking lot

• Locally regular but globally varying– Waves on a lake

• Locally irregular but globally uniform– Field of rocks, boulders and pebbles

• Locally irregular and globally varying

Works Cited• System-Theoretical Analysis of the Clare Bishop Area in the Cat. Jensen

H.J. Springer-Verlag Biological Cybernetics 39 53-66 (1980).

• Coupled Van der Pol Oscillators – A model of Excitatory and Inhibitory Neural Interactions. Kawahara, Takuji. Springer-Verlag. Biological Cybernetics 39, 37-43 (1980).

• Dynamic Interactions between Limb Segments During Planar Arm Movement, S. Kemmerling, D. Varju, Biological Cybernetics 44 67-77 (1982)

• The Information Content of Texture Gradients, Steven, Kent, Biological Cybernetics 42 95-105(1981)