Jiqing Dai, Song Wang, Jianrui Zhang, Wenzheng Zhou, Wenchao Liu, Li Wang, Zhehao Ren, Yuanxia Chen, Haitao Zhang, Xinyi Zhu, Tingrui Pan, Kai Xiao
Biological perception represents a sophisticated process that seamlessly integrates adaptive sensory processing with neural computation. The core mechanism involves transforming external stimuli into potential signals with receptor-synapse synergy enabled by precise ion transport. Inspired by nature's paradigm, we present an adaptive neuromorphic tactile perception system that couples piezoionic sensing with synaptic computing functions via nanochannel-mediated ion transport. The biomimetic tactile device is a composite of two ionic hydrogel films separated by a PET nanochannel membrane. Similar to mechanosensitive ion channels, applied pressure drives ions to selectively traverse the nanochannels, producing fast-adaptive transmembrane potential in single units and slow-adaptive potential maintenance via force-driven coupling of piezoionic signaling with electrical relaxation across interconnected units. Leveraging adaptive sensory signaling within interconnected units, the neuromorphic system reduces the recording consumption of pressure information by 75% and achieves 92.3% texture recognition accuracy. This capability offers a promising pathway toward overcoming integration and computation bottlenecks in wearable intelligent sensing.