VirTex

Psychophysics and coding of vibrotactile signals in the tactile system of rats and humans.

01.07.2016 bis 30.06.2019

Mammalian texture perception is based on mechanical interactions of the sensor (hair and skin) with a surface. It is long known by engineers that elastic object properties greatly transform the relative movements of two objects rubbing against each other. They have termed this effect 'stick-and-slip' movements, describing the charging of elastic energy (stick) and its sudden discharge of it (slip).
Skin and hair are highly elastic materials and correspondingly show prominent stick-and-slip behavior. For texture perception this may be a boon as - at least in the whisker system of rats - it has been clearly demonstrated that stick-and-slip movements occur in texture dependent ways and therefore carry texture information which can be exploited by the neuronal tactile system. In the present proposal we aim at demonstrating that slip movements or 'kinematic events' as we have called them in the rodent whisker system, are exploited as well in human fingertip-based perception.
Further, active tactile systems, like the rodent whisker and the primate finger, sensor movements profoundly alter the way how a texture is represented in the sensory system. The second conjecture to be studied in the present project is that active touch is employed in a goal directed way to manipulate slips and thus optimize perception.
If fast kinematic events play a role for perception, the vibrotactile signal must be registered at highest temporal resolution. This, however, is incompatible with memory storage, which, by nature, is space-limited. The third conjecture to be investigated is how the tactile system deals with the problem to compare incoming sensory information with compressed data in memory.
Toward these ends, we will employ a systematic comparative approach between human and rat tactile perceptional systems. We will present 'virtual tactile textures' - pulsatile skin indentations (or whisker deflections) controlled by real time feedback of finger/whisker position. The virtual environment is artificial, but it is critical for the questions posed here, because it will allow precise and simultaneous experimental control of all perceptional relevant variables.
The two species clearly show differences, but at the heart of both resides the same choice of physical parameters: "Integration of the signal versus extracting kinematic events", and identical computational problems: "How to factor in sensor movements into neuronal and perceptional processes". The comparative approach is highly promising because the rodent system offers invasive monitoring and interrogation of the underlying neuronal circuits and the human system is highly approachable via psychophysics. Last but not least we think it worthwhile to begin to systematically exploit the great knowledge base sampled by a large community of researchers on the whisker system for the purposes of understanding human perception.