Order-Based Representation in Random Networks of Cortical Neurons

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.Author Summary: The idea that sensory objects are represented by the order in which neurons are recruited in response to stimulus presentation was put forward over a decade ago, largely based on computational biology considerations. While intensively analyzed in simulation studies, the general biological applicability of this highly compacted and efficient representation scheme, as an ensemble neural code, was never established. In recent years, algorithmic and experimental technologies advanced to a stage that allows for facing the challenge; here we took advantage of this progress. We let a large-scale random network of cortical neurons develop on top of a microfabricated, multielectrode array that enables electronic interrogation of the network, stimulating through various points in space, and simultaneously recorded the resulting activities from a large number of neurons. We applied state-of-the-art classification algorithms and asked how well the rank order representation scheme handles categorization tasks. We show that recruitment order is generally applicable as an ensemble code; it emerges spontaneously in a large “structureless” network of neurons as a functional code that is invariant to significant temporal variance in spike times and spike rates and flawlessly classifies inputs on a trial-to-trial basis.