Biomedical subjects

Randall D Beer

Publications and source records attributed to Randall D Beer.

7 recordsLinked to original sources

PubMed2006

Mechanical reconfiguration mediates swallowing and rejection in Aplysia californica.

Muscular hydrostats, such as tongues, trunks or tentacles, have fewer constraints on their degrees of freedom than musculoskeletal systems, so changes in a structure's shape may alter the positions and lengths of other components (i.e., induce mechanical reconfiguration). We studied mechanical reconfiguration during rejection and swallowing in the marine mollusk Aplysia californica. During rejection, inedible material is pushed out of an animal's buccal cavity. The grasper (radula/odontophore) closes on inedible material, and then a posterior muscle, I2, pushes the grasper toward the jaws (protracts it). After the material is released, an anterior muscle complex (the I1/I3/jaw complex) pushes the grasper toward the esophagus (retracts it). During swallowing, the grasper is protracted open, and then retracts closed, pulling in food. Grasper closure changes its shape. Magnetic resonance images show that grasper closure lengthens I2. A kinetic model quantified the changes in the ability of I2 and I1/I3 to exert force as grasper shape changed. Grasper closure increases I2's ability to protract during rejection, and increases I1/I3's ability to retract during swallowing. Motor neurons controlling radular closure may therefore affect the behavioral outputs of I2's and I1/I3's motor neurons. Thus, motor neurons may modulate the outputs of other motor neurons through mechanical reconfiguration.

Animals↗

PubMed2006

Design and implementation of multipattern generators in analog VLSI.

In recent years, computational biologists have shown through simulation that small neural networks with fixed connectivity are capable of producing multiple output rhythms in response to transient inputs. It is believed that such networks may play a key role in certain biological behaviors such as dynamic gait control. In this paper, we present a novel method for designing continuous-time recurrent neural networks (CTRNNs) that contain multiple embedded limit cycles, and we show that it is possible to switch the networks between these embedded limit cycles with simple transient inputs. We also describe the design and testing of a fully integrated four-neuron CTRNN chip that is used to implement the neural network pattern generators. We provide two example multipattern generators and show that the measured waveforms from the chip agree well with numerical simulations.

Computers, Analog↗

PubMed2006

Connection and coordination: the interplay between architecture and dynamics in evolved model pattern generators.

We undertake a systematic study of the role of neural architecture in shaping the dynamics of evolved model pattern generators for a walking task. First, we consider the minimum number of connections necessary to achieve high performance on this task. Next, we identify architectural motifs associated with high fitness. We then examine how high-fitness architectures differ in their ability to evolve. Finally, we demonstrate the existence of distinct parameter subgroups in some architectures and show that these subgroups are characterized by differences in neuron excitabilities and connection signs.

Animals↗

PubMed2006

Parameter space structure of continuous-time recurrent neural networks.

A fundamental challenge for any general theory of neural circuits is how to characterize the structure of the space of all possible circuits over a given model neuron. As a first step in this direction, this letter begins a systematic study of the global parameter space structure of continuous-time recurrent neural networks (CTRNNs), a class of neural models that is simple but dynamically universal. First, we explicitly compute the local bifurcation manifolds of CTRNNs. We then visualize the structure of these manifolds in net input space for small circuits. These visualizations reveal a set of extremal saddle node bifurcation manifolds that divide CTRNN parameter space into regions of dynamics with different effective dimensionality. Next, we completely characterize the combinatorics and geometry of an asymptotically exact approximation to these regions for circuits of arbitrary size. Finally, we show how these regions can be used to calculate estimates of the probability of encountering different kinds of dynamics in CTRNN parameter space.

Animals↗

PubMed2004

Neural control exploits changing mechanical advantage and context dependence to generate different feeding responses in Aplysia.

How does neural control reflect changes in mechanical advantage and muscle function? In the Aplysia feeding system a protractor muscle's mechanical advantage decreases as it moves the structure that grasps food (the radula/odontophore) in an anterior direction. In contrast, as the radula/odontophore is moved forward, the jaw musculature's mechanical advantage shifts so that it may act to assist forward movement of the radula/odontophore instead of pushing it posteriorly. To test whether the jaw musculature's context-dependent function can compensate for the falling mechanical advantage of the protractor muscle, we created a kinetic model of Aplysia's feeding apparatus. During biting, the model predicts that the reduction of the force in the protractor muscle I2 will prevent it from overcoming passive forces that resist the large anterior radula/odontophore displacements observed during biting. To produce protractions of the magnitude observed during biting behaviors, the nervous system could increase I2's contractile strength by neuromodulating I2, or it could recruit the I1/I3 jaw muscle complex. Driving the kinetic model with in vivo EMG and ENG predicts that, during biting, early activation of the context-dependent jaw muscle I1/I3 may assist in moving the radula/odontophore anteriorly during the final phase of protraction. In contrast, during swallowing, later activation of I1/I3 causes it to act purely as a retractor. Shifting the timing of onset of I1/I3 activation allows the nervous system to use a mechanical equilibrium point that allows I1/I3 to act as a protractor rather than an equilibrium point that allows I1/I3 to act as a retractor. This use of equilibrium points may be similar to that proposed for vertebrate control of movement.

Animals↗

PubMed2004

Autopoiesis and cognition in the game of life.

Maturana and Varela's notion of autopoiesis has the potential to transform the conceptual foundation of biology as well as the cognitive, behavioral, and brain sciences. In order to fully realize this potential, however, the concept of autopoiesis and its many consequences require significant further theoretical and empirical development. A crucial step in this direction is the formulation and analysis of models of autopoietic systems. This article sketches the beginnings of such a project by examining a glider from Conway's game of life in autopoietic terms. Such analyses can clarify some of the key ideas underlying autopoiesis and draw attention to some of the central open issues. This article also examines the relationship between an autopoietic perspective on cognition and recent work on dynamical approaches to the behavior and cognition of situated, embodied agents.

Artificial Intelligence↗

PubMed2002

Center-crossing recurrent neural networks for the evolution of rhythmic behavior.

A center-crossing recurrent neural network is one in which the null-(hyper)surfaces of each neuron intersect at their exact centers of symmetry, ensuring that each neuron's activation function is centered over the range of net inputs that it receives. We demonstrate that relative to a random initial population, seeding the initial population of an evolutionary search with center-crossing networks significantly improves both the frequency and the speed with which high-fitness oscillatory circuits evolve on a simple walking task. The improvement is especially striking at low mutation variances. Our results suggest that seeding with center-crossing networks may often be beneficial, since a wider range of dynamics is more likely to be easily accessible from a population of center-crossing networks than from a population of random networks.

Algorithms↗