Biological intelligence operates through chemo-ionic signal processing, where neurotransmitters encode information as spatiotemporal chemical gradients that regulate ionic dynamics across neural synapses. Given the diversity of chemical neurotransmitters and ion species, developing an artificial chemo-ionic cascade synapse that can translate biochemical signals into tunable synaptic weights will be of great significance for brain-inspired computing and brain-computer interfaces. Here, we present an artificial dopamine (DA)-ionic cascade synapse by integrating a sensitive DA sensor with an ionic elastomer-based neuromorphic device. The oxidation of DA generates localized electric fields that electrostatically modulate ion migration within the ionic elastomer device, enabling chemical-to-ionic signal transduction and dynamic plasticity control. Consequently, biochemical cues like DA concentration can be directly reflected in tunable ionic synaptic weights, which can then be used to control a robotic platform for recognition tasks. This artificial synapse exhibits biochemical signal-driven behavioral selectivity in an object-grasping task, completing a perception-decision-execution loop. This work establishes a framework for processing biochemical information via native ionic dynamics, paving the way for chemically neuromorphic systems and embodied human-machine interaction.
Zhang et al. (Wed,) studied this question.