E. Bienenstock, L. Cooper, and P. Munro. Theory for the development of neuron selectivity: orientation specicity and binocular interaction in visual cortex. J Neurosci, 2:3248, 1982.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....second plot it is clear that for high post synaptic frequencies, although LTP clearly dominates, the transition probabilities become smaller and smaller. Analogously, for low Ypost, the LTD probabilities become negligible. Note that, up to 60Hz, the behavior resembles the one of the BCM rule [26]. neurons are negligible) 4) a regulatory mechanism that blocks LTP when Ypost is too high, and stops LTD when Ypost is too low. This is a possible implementation of the mechanism required to store correlated patterns [11, 12, 31] It also resembles the one in the BCM rule [26] and it is ....

....of the BCM rule [26] neurons are negligible) 4) a regulatory mechanism that blocks LTP when Ypost is too high, and stops LTD when Ypost is too low. This is a possible implementation of the mechanism required to store correlated patterns [11, 12, 31] It also resembles the one in the BCM rule [26] and it is essentially based on the additional effect of the action potential that signals when the activity of the post synaptic neuron is too elevated or too low. To illustrate all these properties we plot the LTP and LTD transition probabilities in three different ways (see Figs. 3,4,5) Figure ....

Bienenstock, E. L., Cooper, L. N., and Munro, P. W. (1982) Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2, 32

....transitions in case of spontaneous activity of both pre and post synaptic neurons are negligible) 4) a regulatory mechanism that blocks LTP when pot is too high, and stops LTD when pot is too low. This is the mechanism required to store correlated patterns [7, 8] It also resembles the BCM rule [10] and it is essentially based on the extra effect of the action potential that signals when the activity of the post synaptic neuron is too elevated. ....

E.L. Bienenstock, L.N. Cooper, and P.W. Munro (1982) Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2, 32

....function is maximized. Usually a very simple mechanism is assumed to modulate the connection strength, which is based on information locally available at the synapse. As a consequence in connectionist systems synaptic plasticity usually depends only on the average preand postsynaptic activity [1, 3 5]. Many applications, e.g. those deriving from principles like maximization of coherent information, are not feasible without a signal transmitting information about a global property of the network. Synaptic plasticity rests on information that is not locally available at a synapse. Without a ....

....larger. The pattern of weights then approaches that stimulus. Figure 1D shows the receptive fields after convergence. Horizontal bars are learned by the network (corresponding to correlated activity) In contrast to this, learning with a BCM learning rule (standing for Bienenstock Cooper Munro [5]) where a threshold separating regions of LTP and long term depression is regulated by the average postsynaptic activity, results in the network learning vertical bars, the most common stimuli. Thus this example shows that exploiting the properties of the additional learning signal it is possible ....

Bienenstock, E.L., Cooper, L.N., and Munro, P.W. (1982). Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. J. Neurosci. 2, 32-48.

....which ones are preserved. 2 Synaptic plasticity and receptive field development Over the years many researchers have modeled receptive field development using rate based models. I have contributed to this e#ort over the years. We have shown how di#erent models of synaptic plasticity such as BCM [Bienenstock et al. 1982] and other higher order models[Blais et al. 1998] can develop oriented receptive fields with di#ering occular dominance [Shouval et al. 1996] given a natural image environment. Recently attempters have been made to use the STDP curve directly to model synaptic plasticity [Song et al. 2000] ....

Bienenstock, E. L., Cooper, L. N., and Munro, P. W. (1982). Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2:32--48.

....towards the soma (double arrow pointing downward) where it influences synaptic plasticity of all active synapses. See text for details. GABA mediated inhibition leads to higher firing rates and to smaller synaptic currents [48, 49] Alternatively a BCM like learning rule to normalize activity [7] could be combined with the calcium spike effect. Analysing such effects could help understand how calcium spikes and normal firing could interact leading to efficient learning. The cells habituate with respect to calcium spikes, the third term, and change all their weights by an amount ....

Bienenstock E., Cooper L.N., Munro P.W., Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex, J. Neurosci. 2 (1982) 32 -- 48.

....[Atick, 1992; Atick and Redlich, 1992 ] provided an explanation of the center surround structure of retinal ganglion receptive fields in terms of whitening or decorrelation of outputs in response to natural images. Several Hebbian learning algorithms for decorrelation have also been proposed [Bienenstock et al. 1982; Williams, 1985; Barrow, 1987; Linsker, 1988; Oja, 1989; Sanger, 1989; Foldiak, 1990; Atick and Redlich, 1993 ] many of which perform Principal Component Analysis (PCA) Although the PCA of natural images produces lower order components that resemble oriented filters [Baddeley and Hancock, ....

E. L. Bienenstock, L. N. Cooper, and P. W. Munro. Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2:32--48, 1982.

....axis. To provide some intuition regarding the behavior of ECA we have run ECA MAX, ECA MIN, Oja s rule, and two related procedures, BCM and BINGO, on the same density. BCM is a learning rule that was originally proposed to explain development of receptive fields patterns in visual cortex (Bienenstock, Cooper and Munro, 1982). More recently it has been argued that the rule finds projections that are far from Gaussian (Intrator and Cooper, 1992) Under a limited set of conditions this is equivalent to finding the minimum entropy projection. BINGO was proposed to find axes along which there is a bimodal distribution ....

Bienenstock, E., Cooper, L., and Munro, P. (1982). Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2.

....infinitely large values. Analogously, once depressed, synapses decrease to zero. These problems are usually fixed by constraining the learning rules, for instance by keeping the sum of weights constant. Alternatively, the postsynaptic activity can be used to adjust the threshold for potentiation (Bienenstock et al. 1982; K irkwood et al. 1996) to regulate neuronal excitability (Desai et al. 1999) or to scale all of a neuron s synaptic weights (Turrigiano et al. 1998) All these mechanisms stabilize postsynaptic activity and introduce competition, but the choice of constraint influences strongly the behavior ....

Bienenstock EL, Cooper LN, Mundo PW (1982) Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. J Neurosci 2:32-- 48.

....representations. Within the independent units framework, the main approach has been to introduce a sparseness constraint that does not actually involve direct activation based competition. This usually involves adding an extra factor to the learning rule that favors sparse representations (e.g. [42, 14, 43, 44]) or adding a sparseness bias into the activation function itself (e.g. 37] Thus, units are only competing over the long time course of learning (or against their own negative bias) and not directly with one another to represent the current input pattern (i.e. selection) Furthermore, the ....

E. L. Bienenstock, L. N. Cooper, and P. W. Munro. Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2(2):32--48, 1982.

....= V (x i V w i ) Here, NR is explicitly manifested by the second term, which provides a multiplicative correction that is independent of the speci c input x i but is determined by the neuronal output V . Interestingly, this neuronal regulation term guarantees P i w 2 i = 1. The BCM (q.v. [1] model is another example of complex interplay of neuronal regulation and synaptic competition. In the BCM approach, both Hebbian potentiation and depression are used in de ning the synaptic learning rule: Synapses are potentiated when the presynaptic neuron res frequently, while depressed ....

E. L. Bienenstock, L. N. Cooper and P. W. Munro. Theory for the development of neuron selectivity: Orientation specicity and binocular interaction in visual cortex. Journal of Neuroscience 2, 32-48, 1982. Chechik, Horn and Ruppin. Neuronal Regulation and Hebbian Learning 10

....vector of synaptic weights. d d m m 1 1 2 2 m m 1 2 2 dd dd Projections through m 1 1 Projected distribution Figure 1: The stable solutions for a two dimensional two input problem are m 1 and m 2 (left) and similarly with a two cluster data (right) The BCM feature extraction [8, 15] seeks multi modality in the projected distribution of these high dimensional vectors. A simple example is illustrated in Figure 1. For a two input problem in two dimensions, the stable solutions (projection directions) are m 1 and m 2 , each has the property of being orthogonal to one of the ....

E. L. Bienenstock, L. N Cooper, and P. W. Munro. Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. Journal Neuroscience, 2:32--48, 1982.

....of the noise Root Mean Square (RMS) and the dimension of the input space. We prove the existence of critical values for the noise RMS which define the selectivity properties of the BCM neuron. Introduction The BCM theory of cortical plasticity has been introduced by Bienenstock, Cooper and Munro [Bienenstock et al. 1982] to account for the changes observed in cell response of visual cortex due to changes in visual environment. This learning model allows modeling and theoretical analysis of various visual deprivation experiments such as monocular deprivation (MD) binocular deprivation (BD) and reversed suture ....

Bienenstock, E. L., Cooper, L. N., and Munro, P. W. (1982). Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2:32--48.

....locally available information for such a process, making the presynaptic action potential and the postsynaptic membrane potential a first choice. As a consequence, most learning rules investigated are modifications of the classic rule proposed by D.O. Hebb [1949; Stent, 1973; Sejnowoski, 1977; Bienenstock et al. 1982; Artola et al. 1990] He proposed, that synaptic efficacy is increased upon correlated pre and postsynaptic activity. On the other, the performance of the system is measured not on the level of individual synapses, but on the network level [Barlow, 1989; Barlow et al. 1989; Atick and Redlich, ....

....its original formulation a Hebb type learning rule leads to a positive feedback with a run away increase of synaptic efficacy. To solve this problem several mechanisms have been suggested: The average activity of the postsynaptic neuron defines a sliding threshold separating regions of LTP and LTD [Bienenstock et al. 1982]. This leads to a normalization of each neuron s average firing rate. In recent years experimental support for this hypothesis has become available [Kirkwood et al. 1996; Bear, 1997] An alternative is to normalize the total afferent synaptic weight of each neuron [Horn et al. 1998] a mechanism ....

Bienenstock, E., Cooper, L. N., & Munro, P. W. (1982). Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex.

....The response characteristics of the layer II III cells in barrel columns surrounding the D1 barrel column were measured after 7, 20 and 60 days of whisker trimming (see Fig. 5) and significant changes in the sensory evoked activity were found. Our present model is 128 based on the BCM theory [6,7], and it is the same model as the one used recently for explanation of the course of experience evoked plasticity in different layers of the rat barrel cortex [8] In this model we have used a biologically realistic inhibitory circuits (see Fig. 2) in which feedback inhibition, as described below, ....

....to the magnitude of their long latency activity (Fig. 4) As shown below, this functional circuit may help explain some crucial experimental observations. 2 BCM Model of the Barrel Column and Simulation Results Synaptic plasticity is modeled according to the BCM synaptic modification rule [6,7]. If we consider the case of a linear cell, the modification of the i th synapse with the weight m i at time t is proportional to the product of input activity at the i th synapse, d i (t) and a function f, in such a way that [ dm t dt c t t d t i M i ( h f ....

Bienenstock E. L., Cooper L. N & Munro P. W. (1982) Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2, 32-48.

....zero and there would be no modification of the connections impinging on this neuron. The necessity of having some kind of memory effect was also noticed by several authors. Clothiaux et al. 8] use a memory effect in the context of synaptic changes of feedforward connections based on the BCM model [6] and it is also a relevant ingredient in some models for invariant object recognition [13, 41] We can expect that the expansion will increase as T increases, although a quantitative analysis requires numerical simulations. On the other hand a high value of T will have the effect of making 12 ....

Bienenstock E. L., Cooper L. N. and Munro P. W. 1982, Theory for development of neuron selectivity: orientation specificity and binocular interaction in visual cortex J. Neurosci. 2, 32-48

....with the 2D alphanumeric pattern classification in order to investigate the computational properties of the BCM neural network with the so called feedforward inhibition. 1 Introduction The BCM theory was introduced in order to explain selforganization in the developing visual cortex [1]. Later it was used for explanation of the experience dependent plasticity in the mature somatosensory cortex [2] It was shown theoretically [3] and in simulations [4] that the BCM neural network with the so called feedforward inhibition can perform projection pursuit, i.e, it can find ....

....The parameter determines the length of the recent past over which the squared neuron s response is averaged. We call it a neuron s memory of its past activity. The smaller (larger) the shorter (longer) memory. The whole notion of the dynamic modification threshold M was biologically inspired [1]. From the above relations it follows that when 0 c k M all active synaptic weights weaken. When c k M , all active synaptic weights potentiate. 3 Results The network input x was the binary f1; 0g bitmap with 10 Theta 13 pixels. During training and testing, noisy representatives ....

Bienenstock E.L., Cooper L.N, Munro P.W. "Theory for the Development of Neuron Selectivity: Orientation Specificity and Binocular Interaction in Visual Cortex," J. Neuroscience, Vol. 2, pp. 32--48, 1982.

No context found.

E.L. Bienenstock, L.N Cooper, and P.W. Munro. Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2:32--48, 1982.

No context found.

E.L. Bienenstock, L.N Cooper,and P.W. Munro. Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. J. Neurosci. 2:32-8, 1982.

....of AMPAR by calcium dependent phosphorylation and dephosphorylation, LTP LTD induction) with plasticity of NMDAR subunit composition. Our model produces a frequency dependent LTP LTD curve with a sliding synaptic modification threshold similar to what has been proposed on theoretical grounds [2, 26] and observed experimentally [3, 4] 2 Control of synaptic strength by activity dependent regulation of AMPAR phosphorylation The bidirectional regulation of phosphorylation of two sites on the GluR1 subunit of the AMPAR, based on the work of Lee, Huganir and Bear, is schematized in figure 1 ....

.... depends on postsynaptic voltage and presynaptic frequency (figure 3a) or to a model that depends only on presynaptic frequency (figure 3b) We have also demonstrated that a phenomenological model for the modification of NMDAR subunit composition can account for the sliding modification threshold [2, 3, 4]. This mathematical model provides a link between synaptic plasticity observed in situ, and biochemical electrophysiological observations of glutamate receptor composition and function. Further, it combines plasticity of AMPA and NMDA receptors. A consequence of the proposed mechanism of the ....

[Article contains additional citation context not shown here]

E. L. Bienenstock, L. N Cooper, and P. W. Munro. Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2:32--48, 1982.

....many instances of learning in the adult. A covariance rule of Hebbian plasticity roughly states that the change in the efficacy of a given synapse varies in proportion to the covariance between the presynaptic and postsynaptic activities. As noted by many authors (e.g. Sejnowski 1977a, 1977b; Bienenstock et al. 1982; Linsker 1986; Sejnowski et al. 1988) a covariance type rule is preferable to a rule that uses the mere product of pre and post synaptic activities because the covariance rule predicts not only weight increases but also activity related weight decreases, and as a consequence allows convergence ....

Bienenstock, E., Cooper, L.N., and Munro, P. 1982. Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. J. Neurosci., 2, 32--48.

....field. In the other environment, natural image movies are used to provide motion over the receptive field. Using these more realistic environments, we reproduce the observation of Feidler et al. that simple Hebb rules are incapable of producing direction selective cells, and that the BCM rule(Bienenstock et al. 1982) is capable of producing direction selective cells. We also consider several other statistically motivated rules(Blais et al. 1998) which have similar form to the BCM rule, and explore how they develop direction selectivity. In addition we provide an insight into the reason direction selectivity ....

....We denote the input vector by d and the weight vector by m. Neural activity is given by the rectified product of the inputs and the weights, c = #(d m) The derivative of the sigmoidal is given simply by # # . We consider several synaptic modification rules such as the quadratic form of BCM(Bienenstock et al. 1982; Intrator and Cooper, 1992) as well as other statistically motivated rules that share the basic properties proposed by BCM(Blais et al. 1998) We include, for comparison, a stabilized Hebb rule used for extracting the principle component of the input (PCA) Oja, 1982) These rules have the ....

Bienenstock, E. L., Cooper, L. N., and Munro, P. W. (1982). Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience, 2:32--48.

No context found.

E. Bienenstock, L. Cooper, and P. Munro. Theory for the development of neuron selectivity: orientation specicity and binocular interaction in visual cortex. J Neurosci, 2:3248, 1982.

No context found.

E. L. Bienenstock, L. N. Cooper, and P. W. Munro, "Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex," J. Neurosci., vol. 2, no. 1, pp. 32--48, 1982.

No context found.

Bienenstock, EL, Cooper, LN & Munro, PW (1982) Theory for the development of neuron selectivity: Orientation specificity and binocular interaction in visual cortex. Journal of Neuroscience 2:32-48.

No context found.

Bienenstock, EL, Cooper, LN & Munro, PW (1982) Theory for the development of neuron selectivity: Orientation specicity and binocular interaction in visual cortex. Journal of Neuroscience 2:32-48.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC