AN INVESTIGATION OF PERIPHERAL FACTORS AFFECTING THE SPATIAL PERCEPTION OF TACTILE POINT STIMULI

Abstract

The earliest measures of tactile spatial acuity reflect the ability of human observers to localize and discriminate the simplest of stimuli: single or double point (punctate) indentations. Because punctate stimuli cover an extremely small area, they typically only activate a few peripheral afferents at a time. Therefore, many researchers have used single-point localization and two-point discrimination thresholds to probe the density of innervation at different body sites. This thesis explores the relationship between peripheral properties and the spatial perception of tactile point stimuli. In chapter 2, we simulate the neural responses of primary afferents to single and double points, capturing many realistic properties of the periphery: innervation density, the shape and size of receptive fields, and interactions between two-point stimuli. Furthermore, we model optimal performance in localization and discrimination tasks given these afferent responses, and compare it to human performance. We find that human performance is well below optimal, suggesting that humans do not make use of all the information present at the level of the primary afferents. Nevertheless, many human performance trends, resulting from peripheral properties, are predicted by our computational analysis. Using empirical methods, in Chapter 3, we further investigate one of these trends: surround suppression (the suppression of two-point responses relative to that of a single point) is thought to provide a magnitude cue during two-point discrimination (2PD), resulting in elevated performance even at zero separation between two points. We demonstrate that human observers do indeed show elevated 2PD performance at zero separation on a variety of tested body-sites; an alternative task involving orientation discrimination, however, does not show this same trend and is therefore unlikely to be contaminated by the same magnitude cue. In Chapter 4 we review and test a Bayesian model of two-point trajectory estimation that replicates a famous perceptual length contraction illusion. We provide evidence in support of the model: stimuli that give rise to poor spatial acuity also give rise to a stronger length contraction illusion. Overall, the three studies covered in this thesis elucidate many of the peripheral and stimulus properties that shape our perception of tactile point stimuli.