<p>With the emergence of artificial intelligence and big data, computer hardware and systems have encountered significant pressure from rapidly growing data. A promising approach to resolve this embarrassing situation is the use of a neuromorphic device inspired by biological nervous systems that can overcome the von Neumann bottleneck. Piezotronic neuromorphic devices can spatiotemporally modulate electrical transport properties by piezopotential and associate external mechanical motion with electrical output signals directly in an active manner, showing great capability to sense, store, and process information of external stimuli. In this article, we discuss multifunctional applications of neuromorphic piezotronic devices, including bionic sensing, storage, logic computing, and electrical/optical synapses. In the context of future challenges and perspectives, we discuss ways to modulate novel neuromorphic devices with piezoelectric effects more effectively. It is believed that neuromorphic piezotronic devices are promising solutions for the next generation of interactive sensory/memory/computing, which facilitates the development of the Internet of Things, artificial intelligence, biomedical engineering, etc.</p> Graphical Abstract <p></p>