Scene segmentation by spike synchronization in reciprocally connected visual areas. I. Local effects of cortical feedback

Abstract

To investigate scene segmentation in the visual system we present a model of two reciprocally connected visual areas using spiking neurons. Area P corresponds to the orientation-selective subsystem of the primary visual cortex, while the central visual area C is modeled as associative memory representing stimulus objects according to Hebbian learning. Without feedback from area C, a single stimulus results in relatively slow and irregular activity, synchronized only for neighboring patches (slow state), while in the complete model activity is faster with an enlarged synchronization range (fast state). When presenting a superposition of several stimulus objects, scene segmentation happens on a time scale of hundreds of milliseconds by alternating epochs of the slow and fast states, where neurons representing the same object are simultaneously in the fast state. Correlation analysis reveals synchronization on different time scales as found in experiments (designated as tower, castle, and hill peaks). On the fast time scale (tower peaks, gamma frequency range), recordings from two sites coding either different or the same object lead to correlograms that are either flat or exhibit oscillatory modulations with a central peak. This is in agreement with experimental findings, whereas standard phase-coding models would predict shifted peaks in the case of different objects.

What is signal and what is noise in the brain?

“…In section 3 we conclude that the reason for the high CV values in our models can be regarded as a slowly changing temporal covariation of the afferents to the observed neuron. Thus we have found a simple explanation for the data of Softky and Koch (1993) which is consistent with our general ideas presented in the introduction and modelled and discussed in more detail in our previous paper (Knoblauch and Palm, 2002).…”

“…Recently we have implemented a large model of two interacting visual cortical areas in order to investigate scene segmentation and "binding" by spike synchronization (Knoblauch and Palm, 2002). Each neuron in this model receives inputs from more than 1000 other model neurons.…”

“…This view has been challanged by the idea that the apparent noise inherent in brain activity that is not strictly related or temporally coupled to the experiment could be functionally significant. In this work we examine the ISI statistics and discuss these views in a recently published model of interacting cortical areas (Knoblauch and Palm, 2002). From the results of further single neuron simulations we can isolate temporally modulated synaptic input as a main contributor for high ISI variability in our model and possibly in real neurons.…”

“…They perhaps do so with respect to the ongoing experiment, but not with respect to what may be behaviorally relevant for the animal. This means, the observed variability could be caused by internal states of the brain possibly related to attention, expectation, motivation, "binding" to other percepts, or generally to the activation of global "cell assemblies" (Hebb, 1949;Palm, 1982Palm, , 1990Knoblauch and Palm, 2002). These internal states could change on a relatively fine time scale (tens to hundreds of milliseconds) and could play an important role in cortical function (e.g.…”

“…In section 2.1 we reconsider results from a recently published large neural network model of two interacting cortical areas (Knoblauch and Palm, 2002). The aim of our model was to understand the population activity that leads to local field potentials (LFP) and LFP-correlograms that have been ovserved in visual cortical areas (Engel et al, 2001;Eckhorn, 1999).…”