Feng Ding, Di Xue, Lifeng Chi, Lizhen Huang
The von Neumann architecture faces significant limitations, including low transmission efficiency and high energy consumption, when handling large-scale data and unstructured problems. Benefiting from the inherent merits of optical signals including high bandwidth, near-zero Joule heating, fast transmission speed, and immunity to electromagnetic interference, photonics provides a powerful pathway for high-speed neuromorphic computing. Together with the mechanical flexibility and largearea manufacturability of organic semiconductors, organic phototransistor (OPT)-based photonic synapses have therefore attracted extensive attention in recent years. This review provides a comprehensive overview of recent advances in OPT-based photonic synapses, covering operational principles, active materials, advances in bidirectional photoresponse process, as well as cutting-edge applications. Finally, the current challenges and opportunities in this field are highlighted. Distinct from previous reviews, this review emphasizes an in-depth exploration of bidirectional photoresponse mechanisms, a systematic dissection of material-structure-function correlations enabling integrated sensing-memory technology, and emerging.