In this paper we examine the application of artificial neural networks to low level processing of tactile sensory data. In analogy to the term early msion, we call the first level of processing required in tactile sensing early taction. Associated with almost all existing realizations of tactile sensors, are fundamental inverse problems which must be solved. Solutions to these inverse problems are computationally demanding. Among such inverse problems, is the problem of 'deblurring' or deconvolution of data provided by any array of tactile sensors which is also assumed to be corrupted by noise. We note that this inverse problem is ill-posed and that the technique of regularization may be used to obtain solutions. The theory of nonlinear electrical networks is utilized to describe ene~y functions for a ~lass of nonlinear networks and to show that the equilibrium states of the proposed network correspond to ~r~d solutions of the delurring problem. An entropy regularizer is incorporated into the energy function of the network for the recovery of normal stress distributions. It is demonstrated by means of both computer simulations and hardware prototypes that neural networks provide an elegant solution to the need for fast, local computation in tactile sensing. An integrated circuit prototype of the proposed network which has been designed and fabricated is discussed as well.