Material Memristive Device Circuits with Synaptic Plasticity: Learning and Memory Victor Erokhin & Tatiana Berzina & Paolo Camorani & Anteo Smerieri & Dimitris Vavoulis & Jianfeng Feng & Marco P. Fontana # Springer Science+Business Media, LLC 2011 Abstract An important endeavor in modern materials science is the synthesis of adaptive assemblies with information processing capabilities similar to those of biological neural systems. Recent developments concern materials functionally similar to the memristor, a notional electrical circuit whose conductivity is dependent on past activity. This feature is analogous to synaptic plasticity: the ability of neurons to modify their synaptic connections as a result of accumulated experience—the basis of learning and the formation of memory. In this paper, we present the first evidence that memristive device-based organic materials show adaptive behavior similar to biological cognitive systems, using learning in the feeding neural network of the pond snail, Lymnaea stagnalis, as a specific biological reference. The synthetic reproduction of synaptic plasticity reported here can create new paradigms for novel comput- ing systems and give impetus to the search for bio-inspired nanoscale molecular architectures capable of learning and decision making. Keywords Organic memristive system . Synapse analog . Conducting polymer . Solid electrolyte . Heterojunction . Learning and memory 1 Introduction In recent years, the use of biological systems as paradigms for the fabrication of advanced functional materials has become a major research field, spanning many disciplines. However, little progress has been made, mainly because of the great distance separating the complexity of the human brain and what the available technology could offer. Recent developments in the practical realization of the so-called memristor (resistor with memory) [1–3], have opened new vistas for the fabrication of unconventional computing systems, and also of complex materials which can process information similarly to the brain: i.e., bio-inspired infor- mation processors. The memristor, which is at the basis of the recent flurry of activity in the field, was predicted theoretically on the basis of symmetry considerations many years ago [4]. Its conductivity is a function of the charge that has flown through it, thus providing the device with both nonlinearity and past-activity-dependent properties, i.e., memory; on this basis, the memristor is predicted to have properties similar to those of synapses in neural systems [5]. Recently, also silicon realization of the meristive devices has been reported [6, 7]. We have reported an organic memristive system [2] based on an electrochemically controlled polymeric junc- tion [8], in which conductivity varies as a function of its past activity, i.e., of the ionic charge that has passed through it, similarly to the original memristor proposal [4]. The main reason to call it a “memristive device” is the fact V. Erokhin : T. Berzina CNR-IPCF, Rome 43100, Italy V. Erokhin (*) : T. Berzina : P. Camorani : A. Smerieri : M. P. Fontana Department of Physics, University of Parma, Viale Usberti 7A, Parma 43100, Italy e-mail: victor.erokhin@fis.unipr.it D. Vavoulis : J. Feng Department of Computer Science, University of Warwick, Coventry CV4 7AL, UK J. Feng Centre for Computational Systems Biology, Fudan University, Shanghai, People’ s Republic of China BioNanoSci. DOI 10.1007/s12668-011-0004-7