Adam B. Csapo

A Conceptual Framework for the Design of Audio Based Cognitive Infocommunication Channels

Cognitive infocommunication channels are abstract channels which use a combination of sensory substitution and sensorimotor extension to convey structured information via any number of sensory modalities. Our goal is to develop engineering systems which are capable of using cognitive infocommunication channels in order to convey feedback information in novel and meaningful ways. Such applications could help reduce the cognitive load experienced by the user on the one hand, and help alleviate the undesirable effects of hidden parameters on the other. We describe the main challenge behind the development of cognitive infocommunication channels as a two-part problem which consists of the design of a synthesis algorithm and the design of a parameter-generating function for the synthesis algorithm. We use formal concept algebra to describe the kinds of synthesis algorithms which are capable of reflecting realistic forms of interaction between the user and the information which is to be communicated. Using this model, we describe an application example in which auditory signals are used to convey information on tactile percepts. Through an experimental evaluation of the application, we demonstrate that our approach can be used successfully for the design of cognitive infocommunication channels.

Sound of Vision - Spatial Audio Output and Sonification Approaches

The paper summarizes a number of audio-related studies conducted by the Sound of Vision consortium, which focuses on the construction of a new prototype electronic travel aid for the blind. Different solutions for spatial audio were compared by testing sound localization accuracy in a number of setups, comparing plain stereo panning with generic and individual HRTFs, as well as testing different types of stereo headphones vs custom designed quadrophonic proximaural headphones. A number of proposed sonification approaches were tested by sighted and blind volunteers for accuracy and efficiency in representing simple virtual environments.

The Spiral Discovery Network as an Automated General-Purpose Optimization Tool

The Spiral Discovery Method (SDM) was originally proposed as a cognitive artifact for dealing with black-box models that are dependent on multiple inputs with nonlinear and/or multiplicative interaction effects. Besides directly helping to identify functional patterns in such systems, SDM also simplifies their control through its characteristic spiral structure. In this paper, a neural network-based formulation of SDM is proposed together with a set of automatic update rules that makes it suitable for both semiautomated and automated forms of optimization. The behavior of the generalized SDM model, referred to as the Spiral Discovery Network (SDN), and its applicability to nondifferentiable nonconvex optimization problems are elucidated through simulation. Based on the simulation, the case is made that its applicability would be worth investigating in all areas where the default approach of gradient-based backpropagation is used today.