Scene segmentation by spike synchronization in reciprocally connected visual areas. II. Global assemblies and synchronization on larger space and time scales

Abstract

We present further simulation results of the model of two reciprocally connected visual areas proposed in the first paper [Knoblauch and Palm (2002) Biol Cybern 87:151-167]. One area corresponds to the orientation-selective subsystem of the primary visual cortex, the other is modeled as an associative memory representing stimulus objects according to Hebbian learning. We examine the scene-segmentation capability of our model on larger time and space scales, and relate it to experimental findings. Scene segmentation is achieved by attention switching on a time-scale longer than the gamma range. We find that the time-scale can vary depending on habituation parameters in the range of tens to hundreds of milliseconds. The switching process can be related to findings concerning attention and biased competition, and we reproduce experimental poststimulus time histograms (PSTHs) of single neurons under different stimulus and attentional conditions. In a larger variant the model exhibits traveling waves of activity on both slow and fast time-scales, with properties similar to those found in experiments. An apparent weakness of our standard model is the tendency to produce anti-phase correlations for fast activity from the two areas. Increasing the inter-areal delays in our model produces alternations of in-phase and anti-phase oscillations. The experimentally observed in-phase correlations can most naturally be obtained by the involvement of both fast and slow inter-areal connections; e.g., by two axon populations corresponding to fast-conducting myelinated and slow-conducting unmyelinated axons.

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

“…The more stimuli that are presented, the less time an assembly representing a single stimulus can be in the fast state because of the interplay of competition and habituation. These findings may be related to experimental results investigating attention and shrinking receptive fields (Moran and Desimone 1985;Reynolds and Desimone 1999; see also Knoblauch and Palm 2002). c Summed spike records of ten different activated patches representing the triangle (rows 6-9), ellipse (rows 3-5), and rectangle (rows 0-2) stimuli in population P .…”

“…Since axonal delays are quite small (see Appendix B.3) and the associative memory is fully connected (i.e., synapses between any pairs of neurons are possible), we rather interpret area C as one or several cortical columns of a higher visual area (Braitenberg and Schu¨z 1991), e.g., V4 or IT. In the second paper (Knoblauch and Palm 2002) we simulate a larger variant of this model to investigate more-global effects, where also area C is organized topographically.…”

“…But that model simulated two areas at the same hierarchical stage (i.e., left and right hemispheres), while our model simulates two areas at different stages where only the lower area receives input from the retina. Therefore even with zero conduction delays for the connection between areas P and C, the neurons in the higher area C should become activated some time after the activation of the lower area P. In the second paper (Knoblauch and Palm 2002) we show results from simulation experiments concerning inter-areal delays and zero phase lag, and discuss possible solutions to this problem.…”

Spike-timing-dependent synaptic plasticity can form “zero lag links” for cortical oscillations.

“…However, since the transmission delays between hemispheres can reach many tens of milliseconds ( [12], Fig. 1B), nonlocal synchronization e ects in cortex have been discussed controversially: modeling studies have indicated that in-phase synchronization requires synaptic delays smaller than 1 3 or 1 4 of the oscillation period [6,9,10]. For 50 Hz oscillations the zero-phase condition (delay ¡ 20 3 ms) is obviously not met.…”

Temporal Correlation Theory and Modeling the Segmentation of the Visual Information in the Brain (a review)

“…Это противоречит как экспериментальным данным, так и теоретическим представлениями ТВК, согласно которым активность нейронов, представляющих один объект, должна быть синфазной. Авторами предложено несколько вариантов модификации модели для преодоления этой трудности [37]. Один из вариантов был исследован более подробно методами имитационного моделирования.…”