Abstract
This chapter is about the effects on perception of joint sensory stimulation. It shows that by combining various simultaneous stimuli, it is possible to elicit a psychophysiological effect that is different from the sum of the responses of each stimulus alone; in particular, a new cortical response is elicited besides the ones pertaining to each stimulation modality. This is believed to be particularly useful when designing mobile interfaces because of their needs to be maximally informative while minimally intrusive. Moreover, no technologic additional requirements are necessarily needed, besides proper synchronization protocols, with respect to standard technology, once more showing that often improving is a question of properly combining existing knowledge.
TopBackground
Our interaction with the world is mediated through the sensorial systems, allowing us to acquire information from the surroundings. Human perception is based on the psychophysiological properties of such interaction, even to make us interact with possibly mobile information devices in an increasingly easy way.
It is well known that electrical potentials (Reagan, 1972) as well as their magnetic correlate (Liberati, Narici, Santoni & Cerutti, 1992a) are measurable on the skull as the evoked effect of sensorial stimulation. Their topographic (Liberati, DiCorrado & Mandelli, 1992b) relevance (Liberati, 1992c) with respect to the background electrical activity of the unstimulated brain needs to be captured via quite sophisticated algorithms discriminating signal from noise, such as stochastic parametric identification (Cerutti, Baselli, Liberati & Pavesi, 1987; Liberati, Cerutti, DiPonzio, Ventimiglia & Zaninelli, 1989) and Kalman filtering (Liberati, Bertolini & Colombo, 1991b). Such tools do allow monitoring the psychophysiological effect (Chiarenza, Cerutti, Liberati & Mascellani, 1987) of even multimedia stimulation (Liberati et al., 1991a), also implying a coordination of brain activity in space and time (Liberati, Cursi, Locatelli, Comi & Cerutti, 1997).
Key Terms in this Chapter
Piecewise Affine Models: The evolution in time of the corresponding hybrid systems composed of both smooth dynamics and sudden jumps.
Latency: The amount of time between the stimulus administration and the peak of the evoked potential, measuring the neural average conduction time.
Neural Networks: A black-box nonlinear model whose main characteristics are to be composed by many nonlinear elements of the same simple kind composed in a regular structure whose parameters are identified from examples of data.
Logical Networks: A fast binary rule generator and variable selector able to build understandable logical expressions by analyzing the Hamming distance between samples.
Stochastic Parametric Identification: Computation of the parameters best suited to mathematically describe the process underlying the data within a general mathematical model.
Adaptive Bayesian Networks: Tree, automatically built from data, illustrating the causal relationship among the main variables and the class of outcomes.
Minimum Description Length Principle: Based on information theory, do state that the best model is the one minimizing both the variables and the bits to describe data in terms of them, thus minimizing its overall communication cost.
Evoked Potentials: Brain electrical activity detectable immediately following the administration of a stimulus.