Artificial neural network performance based on correlation analysis qualitatively comparable with human performance in behavioral signal detection experiments.

Standard Gaussian signal detection theory (SDT) is a widely used approach to assess the detection performance of living organisms or technical systems without looking at the inner workings of these systems like neural or electronic mechanisms. Nevertheless, a consideration of the inner mechanisms of a system and how they produce observed behaviors should help to better understand the functioning. It might even offer the possibility to demonstrate isolated pattern separation processes directly in the model. To do so, modeling the interaction between the entorhinal cortex (EC) and the hippocampal subnetwork dentate gyrus (DG) via the perforant path reveals the decorrelation network's mode of operation. We show that the ability to do pattern separation is crucial for high-performance pattern recognition, but also for lure discrimination, and depends on the proportionality between input and output network. NEW & NOTEWORTHY We elucidate the interplay of the entorhinal cortex and the hippocampal dentate gyrus during pattern separation tasks by providing a new simulation model. Functional memory formation and processing of similar memory content is illuminated from within the system. For the first time orthogonalized spiking patterns are evaluated with signal detection theory methods, and the results are applied to clinically established and novel tests.

A memristive plasticity model of voltage-based STDP suitable for recurrent bidirectional neural networks in the hippocampus.

Memristive systems have gained considerable attention in the field of neuromorphic engineering, because they allow the emulation of synaptic functionality in solid state nano-physical systems. In this study, we show that memristive behavior provides a broad working framework for the phenomenological modelling of cellular synaptic mechanisms. In particular, we seek to understand how close a memristive system can account for the biological realism. The basic characteristics of memristive systems, i.e. voltage and memory behavior, are used to derive a voltage-based plasticity rule. We show that this model is suitable to account for a variety of electrophysiology plasticity data. Furthermore, we incorporate the plasticity model into an all-to-all connecting network scheme. Motivated by the auto-associative CA3 network of the hippocampus, we show that the implemented network allows the discrimination and processing of mnemonic pattern information, i.e. the formation of functional bidirectional connections resulting in the formation of local receptive fields. Since the presented plasticity model can be applied to real memristive devices as well, the presented theoretical framework can support both, the design of appropriate memristive devices for neuromorphic computing and the development of complex neuromorphic networks, which account for the specific advantage of memristive devices.