A Direction-Dependent Mechanotransduction Model to Convert Fingertip Forces Into Neural Spike Trains for Tactile...

Area of Science:

• Computational Neuroscience
• Biophysics
• Robotics

Background:

• Human mechanoreceptors exhibit directional sensitivity to shear forces, crucial for fine tactile perception and object manipulation.
• Existing computational models often overlook this directional tuning, limiting their accuracy and applicability in areas like prosthetic feedback.

Purpose of the Study:

• To develop a Direction-Dependent Mechanotransduction Model (DDMM) that accurately replicates the direction-specific encoding of human tactile afferents.
• To enhance the biological realism and effectiveness of computational models for tactile feedback systems.

Main Methods:

• Integrated multidirectional pressure and shear forces to modulate neural spiking based on shear vector alignment and neuron-specific attenuation.
• Transformed force inputs into afferent-specific currents (SAI, RAI, RAII) and converted them into spike trains using an Izhikevich neuron model.
• Validated the model using simulated fingertip interactions and experimental triaxial force measurements from human fingertip actions.

Main Results:

• Simulated interactions yielded directionally selective spiking frequencies (0-47.5 Hz), consistent with biological ranges.
• The model's directional tuning, quantified by PRSI (0.31-0.45), aligns with experimentally measured DSIs (0.23 ± 0.18).
• Experimental validation showed a ~350% increase in mean spiking frequency when neural attenuation profiles aligned with shear force direction.

Conclusions:

• The DDMM provides a biologically inspired and computationally efficient framework for encoding tactile force direction.
• This model has significant potential applications in neuroprosthetics, robotic manipulation, and advanced somatosensory modeling.