Can a Computer Tell Differences between Vibrations?: Physiology-Based Computational Model for Perceptual Dissimilarity Prediction

What is it about?

Human tactile perception is the perceptual responses to a given vibration set, so investigating human perception for tactile signals is time-consuming and expensive. We proposed a computational model that can predict the perceptual dissimilarities of users for vibrations. Our model was inspired by physiological knowledge of vibrotactile stimuli transmission from the sub-neural network to the somatosensory cortex.

Featured Image

Photo by Killian Cartignies on Unsplash

Photo by Killian Cartignies on Unsplash

Why is it important?

We set parameters that simulate human somatosensory system to train our model. Interestingly, the fitted parameter values were close to the human perception system. Moreover, our model showed robust predictions for perceptual dissimilarities of users with more similar perceptual spaces than existing metrics across vibrotactile parameters or input stimulation types. We believe that our model has a potential to serve as a pilot tool for pattern design due to its flexibility for vibrotactile parameter tests, and it will be able to decrease the time and cost of the vibration pattern design process and accelerate it.