Order-Based Representation in Random Networks of Cortical Neurons

Abstract

The wide range of time scales involved in neural excitability and synaptic transmission might lead to ongoing change in the temporal structure of responses to recurring stimulus presentations on a trial-to-trial basis. This is probably the most severe biophysical constraint on putative time-based primitives of stimulus representation in neuronal networks. Here we show that in spontaneously developing large-scale random networks of cortical neurons in vitro the order in which neurons are recruited following each stimulus is a naturally emerging representation primitive that is invariant to significant temporal changes in spike times. With a relatively small number of randomly sampled neurons, the information about stimulus position is fully retrievable from the recruitment order. The effective connectivity that makes order-based representation invariant to time warping is characterized by the existence of stations through which activity is required to pass in order to propagate further into the network. This study uncovers a simple invariant in a noisy biological network in vitro; its applicability under in vivo constraints remains to be seen.

Tradeoffs and Constraints on Neural Representation in Networks of Cortical Neurons

“…Of the two time-based schemes, rank-order (which is a further reduction of the time-to-first-spike scheme) captures practically all the information carried by the time-to-first-spike scheme. This is consistent with recently published results (Shahaf et al, 2008). A table of all the results used for construction of Figure 3a is presented in supplemental Item 2 (available at www.…”

“…In this study, using generic networks of cortical neurons as a model system, we follow the path of a stimulus-reconstruction approach to compare the representational efficacy of four types of popular schemes, two rate-base and two time-based: population-counthistogram (Schwartz, 1993;Hupé et al, 2001;Fiorillo et al, 2003), spike-count (Arabzadeh et al, 2006;Foffani et al, 2009;Jacobs et al, 2009), time-to-first-spike (Petersen et al, 2001;Foffani et al, 2004;Johasson and Birznieks, 2004;Gollisch and Meister, 2008;Gollisch and Meister, 2008), and rank-order Van Rullen and Thorpe, 2001;VanRullen et al, 2005;Shahaf et al, 2008). Notwithstanding limitations associated with the stimulus-reconstruction approach in relation to brain function, it served us well in the present context as a mean for estimating the total information content, embedded in a given response feature, about an input.…”

“…The immediate response may last for up to 25 ms following a stimulus, however more then 95% of its spikes are confined to the first 10 ms (Bakkum et al, 2008;Shahaf et al, 2008; supplemental Item 1, available at www.jneurosci.org as supplemental material); therefore, the first 10 ms following each stimulus were excluded from the data. While data beyond this limit might contain, in a very low probability, a small fraction of spikes that were directly activated by the stimulus, we chose not to exclude spikes beyond 10 ms since the data there is made of mostly synaptically mediated response.…”

Repeated stimulation of cultured networks of rat cortical neurons induces parallel memory traces

“…The finding that stimulus responses stabilized, probably means that subsequent presentation of the cue reactivated the trace and did not interfere with the established activity-connectivity equilibrium. Marom and coworkers showed that repeated stimulation of cultured cortical networks induced site-specific responses that enabled deduction of the stimulation site during up to 24 h (Shahaf et al 2008;Kermany et al 2010). This further supports the view that presentation of the original cue reactivates the trace.…”

Functional Connectivity of Neural Network in Dissociated Hippocampal Culture Grown on Microelectrode Array

“…Low-frequency stimulation of neuronal cultures is known to evoke network bursts whose patterns depend on the location of the stimulating electrode [5,6]. The induced reverberating bursts play a key role in changing functional connectivity of neural networks [7][8][9][10][11].…”