Int. Journ. of Unconventional Computing, Vol. 4, pp. 23–32 Reprints available directly from the publisher Photocopying permitted by license only  c 2007 Old City Publishing, Inc. Published by license under the OCP Science imprint, a member of the Old City Publishing Group Biological Computing Using Perfusion Anodophile Biofilm Electrodes (PABE) J OHN GREENMAN 1,2 I OANNIS I EROPOULOS 1,2 , AND CHRIS MELHUISH 2 1 Faculty of Applied Sciences, University of the West of England, Bristol, UK 2 Bristol Robotics Laboratory, Universities of Bristol and of the West of England, Bristol, UK E-mail: John.Greenman@uwe.ac.uk Received: December 1, 2006. Accepted: December 19, 2006. This paper presents a theoretical approach to biological computing, using biofilm electrodes by illustrating a simplified Pavlovian learning model. The theory behind this approach was based on empirical data produced from a prototype version of these units, which illustrated high stability. The implementation of this system into the Pavlovian learning model, is one example and possibly a first step in illustrating, and at the same time discovering its potential as a computing processor. Keywords: artificial intelligence, biological computing, neurone-like, transistor-like, unit and connected assemblies, Pavlovian association learning. 1 INTRODUCTION We have recently described what we now term as perfusion anodophile biofilm electrodes (PABES) and have demonstrated their potential for basic binary type computing [1]. In particular we have shown that the biofilm system can maintain a dynamic steady state under one (of many possible) particular set of physicochemical conditions and then switch to a new steady state in response to changes in one of the parameters of the physicochemical environment (set by the operator) leading to a new condition. We also proposed how interconnecting units might be configured into logic gates (AND, OR, XOR) in order to perform basic binary logic operations [1]. Connections could be made via electrical or fluidic links. A schematic representation of the PABE unit and the type of physicochemical changes it will respond to is shown in Greenman et al. [1], and the mathematical modeling for these units is described below. 23 brought to you by CORE View metadata, citation and similar papers at core.ac.uk provided by UWE Bristol Research Repository