Stable and Rapid Recurrent Processing in Realistic Autoassociative Memories

1998 ◽  
Vol 10 (2) ◽  
pp. 431-450 ◽  
Author(s):  
Francesco P. Battaglia ◽  
Alessandro Treves
Keyword(s):  
Memory Retrieval ◽  
Time Course ◽  
Storage Capacity ◽  
Time Constants ◽  
Exponential Time ◽  
Integrate And Fire ◽  
Recurrent Processing ◽  
Recurrent Collateral ◽  
Analytical Result ◽  
And Storage

It is shown that in those autoassociative memories that learn by storing multiple patterns of activity on their recurrent collateral connections, there is a fundamental conflict between dynamical stability and storage capacity. It is then found that the network can nevertheless retrieve many different memory patterns, as predicted by nondynamical analyses, if its firing is regulated by inhibition that is sufficiently multiplicative in nature. Simulations of a model network with integrate-and-fire units confirm that this is a realistic solution to the conflict. The simulations also confirm the earlier analytical result that cued-elicited memory retrieval, which follows an exponential time course, occurs in a time linearly related to the time constant for synaptic conductance inactivation and relatively independent of neuronal time constants and firing levels.

Voltage-clamp analysis of the ionic conductances in a leech neuron with a purely calcium-dependent action potential

1987 ◽  
Vol 58 (6) ◽  
pp. 1468-1484 ◽  
Author(s):  
J. Johansen ◽  
J. Yang ◽  
A. L. Kleinhaus
Keyword(s):  
Steady State ◽  
Action Potential ◽  
Time Course ◽  
Outward Current ◽  
Time Constants ◽  
Exponential Time ◽  
Calcium Dependent ◽  
Voltage Dependent ◽  
Ca Currents ◽  
Single Exponential

1. The purely calcium-dependent action potential of the anterior lateral giant (ALG) cell in the leech Haementeria was examined under voltage clamp. 2. Analysis with ion substitutions showed that the ALG cell action potential is generated by only two time- and voltage-dependent conductance systems, an inward Ca-dependent current (ICa) and an outward Ca-dependent K current IK(Ca). 3. The kinetic properties of the inward current were examined both in Cs-loaded neurons with Ca as the current carrier as well as in Ba-containing Ringer solutions with Ba as the current carrier, since Ba effectively blocked all time- and voltage-dependent outward current. 4. During a maintained depolarization, Ba and Ca currents activated with a time constant tau m, they then inactivated with the decay following a single exponential time course with a time constant tau h. The time constants for decay of both Ba and Ca currents were comparable, suggesting that the mechanism of inactivation of ICa in the ALG cell is largely voltage dependent. In the range of potentials from 5 to 45 mV, tau m varied from 8 to 2 ms and tau h varied from 250 to 125 ms. 5. The activation of currents carried by Ba, after correction for inactivation, could be described reasonably well by the expression I'Ba = I'Ba(infinity) [1--exp(-t/tau m)]. 6. The steady-state activation of the Ba-conductance mBa(infinity) increased sigmoidally with voltage and was approximated by the equation mBa(infinity) = (1 + exp[(Vh-6)/3])-1. The steady-state inactivation hBa(infinity) varied with holding potential and could be described by the equation hBa(infinity) = [1 + exp(Vh + 10/7)]-1. Recovery from inactivation of IBa was best described by the sum of two exponential time courses with time constants of 300 ms and 1.75 s, respectively. 7. The outward current IK(Ca) developed very slowly (0.5–1 s to half-maximal amplitude) and did not inactivate during a 20-s depolarizing command pulse. Tail current decay of IK(Ca) followed a single exponential time course with voltage-dependent time constants of between 360 and 960 ms. The steady-state activation n infinity of IK(Ca) increased sigmoidally with depolarization as described by the equation n infinity = [1 + exp(Vh-13.5)/-8)]-1. 8. The reversal potentials of IK(Ca) tail currents were close to the expected equilibrium potential for potassium and they varied linearly with log [K]o with a slope of 51 mV. These results suggest a high selectivity of the conductance for K ions.(ABSTRACT TRUNCATED AT 400 WORDS)

scholarly journals Time-Course of Vibratory Adaptation and Recovery in Cutaneous Mechanoreceptive Afferents

2005 ◽  
Vol 94 (5) ◽  
pp. 3037-3045 ◽  
Author(s):  
Y. Y. Leung ◽  
S. J. Bensmaïa ◽  
S. S. Hsiao ◽  
K. O. Johnson
Keyword(s):  
Time Course ◽  
Fiber Type ◽  
Companion Paper ◽  
Neural Mechanisms ◽  
Type I ◽  
Vibratory Stimulation ◽  
Time Constants ◽  
Fire Model ◽  
Integrate And Fire ◽  
Central Factors

Extended suprathreshold vibratory stimulation applied to the skin results in a desensitization of cutaneous mechanoreceptive afferents. In a companion paper, we describe the dependence of the threshold shift on the parameters of the adapting stimulus and discuss neural mechanisms underlying afferent adaptation. Here we describe the time-course of afferent adaptation and recovery. We found that absolute and entrainment thresholds rise and fall exponentially during adaptation and recovery with time constants that vary with fiber type. slowly adapting type I (SA1) afferents adapt most rapidly, and pacinian (PC) afferents adapt most slowly, whereas rapidly adapting (RA) afferents exhibit intermediate rates of adaptation; SA1 fibers also recover more rapidly from adaptation than RA and PC fibers. We also showed that threshold adaptation is accompanied by a shift in the timing of the spikes within individual cycles of the adapting stimulus (i.e., a shift in the impulse phase). We invoked an integrate-and-fire model to explore possible mechanisms underlying afferent adaptation. Finally, we found that the time-course of afferent adaptation is more rapid than that of its psychophysical counterpart, as is the time-course of recovery from adaptation, suggesting that central factors play a role in the psychophysical phenomenon.

scholarly journals Transparent CoAP Services to IoT Endpoints through ICN Operator Networks

Sensors ◽  
2019 ◽  
Vol 19 (6) ◽  
pp. 1339 ◽  
Author(s):  
Hasan Islam ◽  
Dmitrij Lagutin ◽  
Antti Ylä-Jääski ◽  
Nikos Fotiou ◽  
Andrei Gurtov
Keyword(s):  
Storage Capacity ◽  
Communication Overhead ◽  
Full Potential ◽  
Core Network ◽  
Computation Complexity ◽  
State Management ◽  
Iot Devices ◽  
Novel Applications ◽  
And Storage ◽  
The Internet Of Things

The Constrained Application Protocol (CoAP) is a specialized web transfer protocol which is intended to be used for constrained networks and devices. CoAP and its extensions (e.g., CoAP observe and group communication) provide the potential for developing novel applications in the Internet-of-Things (IoT). However, a full-fledged CoAP-based application may require significant computing capability, power, and storage capacity in IoT devices. To address these challenges, we present the design, implementation, and experimentation with the CoAP handler which provides transparent CoAP services through the ICN core network. In addition, we demonstrate how the CoAP traffic over an ICN network can unleash the full potential of the CoAP, shifting both overhead and complexity from the (constrained) endpoints to the ICN network. The experiments prove that the CoAP Handler helps to decrease the required computation complexity, communication overhead, and state management of the CoAP server.

scholarly journals Transport and storage of serotonin by thrombin-treated platelets.

1975 ◽  
Vol 65 (2) ◽  
pp. 359-372 ◽  
Author(s):  
H J Reimers ◽  
D J Allen ◽  
I A Feuerstein ◽  
J F Mustard
Keyword(s):  
Storage Capacity ◽  
Platelet Membrane ◽  
Transport Rate ◽  
Primary Reason ◽  
High Concentration ◽  
Serotonin Content ◽  
Amine Storage ◽  
Storage Granules ◽  
And Storage ◽  
Amine Storage Granules

Repeated thrombin treatment of washed platelets prepared from rabbits can decrease the serotonin content of the platelets by about 80%. When these platelets are deaggregated they reaccumulate serotonin but their storage capacity for serotonin is reduced by about 60%. If thrombin-pretreated platelets are allowed to equilibrate with a high concentration of serotonin (123 mu M), they release a smaller percentage of their total serotonin upon further thrombin treatment, in comparison with the percentage of serotonin released from control platelets equilibrated with the same concentration of serotonin calculations indicate that in thrombin-treated platelets reequilibrated with serotonin, two-thirds of the serotonin is in the granule compartment and one-third is in the extragranular compartment, presumably the cytoplasm. Analysis of the exchange of serotonin between the suspending fluid and the platelets showed that thrombin treatment does not alter the transport rate of serotonin across the platelet membrane and does not cause increased diffusion of serotonin from the platelets into the suspending fluid. The primary reason for the reduced serotonin accumulation by the thrombin-treated platelets appears to be loss of amine storage granules or of the storage capacity within the granules.

Sign in / Sign up

Export Citation Format

Share Document