Search

At present, autonomous, robotic dexterous manipulation in unknown environments still eludes us. Children only a few years old lift and manipulate unfamiliar objects more dexterously than today’s robots. Thus, robotics researchers increasingly agree that ideas from biology can strongly benefit the design of autonomous robots.

We investigated the tactile perception of a liquid surface that can be clearly felt as a thin line by a hand moving in the liquid. Although this phenomenon was first reported by Meissner in 1859 and is quite well known, the underlying mechanism is poorly understood. This study aimed to clarify how we perceive the boundary between the atmosphere and water as a cutaneous sensation. We found that skin hair plays a major role in the perception on hairy skin, while surface tension does not significantly contribute to perception of a liquid surface. Furthermore, we found that glabrous skin has a smaller role than hairy skin in this sensation.

In this study, tactile afferents in monkey fingertips were mechanically stimulated, using a flat disc shaped probe, with several magnitudes of torque, in clockwise and anticlockwise directions. In order to prevent slip during the stimulation event, a sufficient normal force was also applied, with three different magnitudes tested. Recordings were made from afferents innervating the glabrous skin covering the entire distal segment of the finger. A Parzen window classifier was used to assess the capacity of tactile afferents to discriminate, concurrently and in real-time, the three stimulus parameters; namely, background normal force, torque magnitude and direction. Despite the potentially confounding interactions between stimulus parameters, classification accuracy was very high and was improved even further by selecting subsets of best performing afferents.

A series of experiments was performed in which psychophysical measurements were made at threshold and suprathreshold levels of stimulation under carefully controlled laboratory conditions. A wide range of stimulus parameters was explored including sinusoidal frequency, intensity, size of contactor, surround condition, skin temperature, and body site. These results are compared to experiments in which responses to sinusoidal displacements were measured from receptor nerve fibers in this and other laboratories. Our work has culminated in a model of cutaneous mechanoreception that proposes four discrete information channels that combine at threshold and suprathreshold levels to signal tactile perception.The psychophysical channels and their physiological counterparts are: 1) the Pacinian channel, mediated by Pacinian-corpuscle afferents; 2) the Non-Pacinian I channel, mediated by rapidly-adapting afferents (Meissner corpuscles); 3) the Non-Pacinian II channel, mediated by slowly-adapting Type II afferents (Ruffini endings); and 4) the Non-Pacinian III channel, mediated by slowly-adapting Type I afferents (Merkel cell-neurite complex). These channels operate over specific bands of vibratory frequencies and the channels partially overlap in their absolute sensitivities.

Until 1910, histological investigation of the skin had revealed a variety of structurally developed end organs in the skin, all of them connected to myelinated afferent nerve fibres. The function of these receptor structures had to remain a matter of speculation until it became possible to record electrophysiologically from identified single nerve fibres. Unitary recording techniques were pioneered by Adrian and Zottermann (1926). Adrian and Umrath (1929) were among the first to combine structure and function in their work on Pacinian corpuscles. Frankenhaeuser’s (1949) Lindblom’s (1965) and Tapper’s (1965) work perfected techniques linking structure and function.

We have selected material for this presentation by structuring the discussion around a series of critical questions that might be asked with a view toward modelling a biological tactile sensing system.

Mammals are covered by the epidermis, consisting of a soft inner portion, called the Malpighian layer, and a hardened outer cornified portion. The inner part maintains the tissue by proliferation and differentiation of its cellular components, the outer part protects the organism from the environment. On the surface of the epidermis inscriptions occur forming characteristic ridges and sulci (Fig. 1). The inner surface may be smooth or rough, forming branching ridges and mounds enclosing valleys and craters (Montagna and Parakkal 1974). Thus the architecture of the epidermis varies from site to site in relation to its surface and undersurface structures. In general the epidermis is thicker over glabrous skin than over hairy skin, it is thickest on the palm, sole, paw, muzzle and tail. In man its thickness varies from 0.8 to 1.4 mm. The epidermis merges with mucous epithelia of the eye, nose, ear, mouth, rectum and genitalia and its continuity is interrupted by appendages of the skin.

Unlike previous papers, this study presents the ultrastructural findings from investigations carried out exclusively on human biopsied material. However, with a view to representing the experimental results on the denervated and reinnervated receptors, as well as the results of cytochemistry and radioautography, the ultrastructure of some avian cutaneous receptors is also included.