A Radial Modulus-Gradient Fiber for Chronic Recording and Decoding in Deep Brain

Fiber electronics provide the most promising platform for the detection, modulation, and reconstruction of biosignals in the brain. However, preserving stable communication between fiber electronics and cellular-scale targets in the deep brain is critical but challenging because of their mechanical mismatch. Here, our study fills this gap by developing a radial modulus-gradient fiber (RMGF), which can bridge high-modulus conductive components (MPa) and low-modulus brain tissue (kPa) to well eliminate the mechanical mismatch at the entire neural‒device interface. The RMGF exhibits strain-insensitive electrical properties (<0.2% resistance fluctuation over 700,000 stretching‒release cycles). As an example, the RMGF enables unprecedented five-month continuous tracking of single neurons in the dorsal lateral geniculate nucleus of freely moving cats, and allows reconstruction of visual stimuli with the use of only three neurons, with a high correlation coefficient of 0.95, approaching the theoretical limit of the unscented Kalman filter (0.97). The results indicate that dorsal lateral geniculate nucleus neurons maintain stable tuning properties (spatial frequency sensitivity, ON/OFF characteristics, and X-cell classification) and reveal a minimal effective ensemble for efficient encoding of information within deep thalamic circuits. This RMGF represents a platform for chronic recording at the single-cell level and investigating fundamental mechanisms in the deep tissues.