Researchers Unveil Groundbreaking Artificial Tongue That Learns Flavors

A team of researchers from the National Center for Nanoscience and Technology in Beijing has developed a pioneering artificial tongue capable of learning and distinguishing various flavors. This innovation represents a significant leap in taste technology, showcasing the potential to digitize the sense of taste with remarkable accuracy.

The new device, constructed from graphene oxide, not only identifies flavors but also learns from them, achieving nearly 99% accuracy in detecting sour, salty, bitter, and sweet tastes during laboratory tests. This artificial gustatory system mimics biological taste perception by utilizing layered graphene oxide membranes, which sense chemicals in solution and process these signals similarly to how human taste buds and neurons interact.

To address the unique challenges of taste, which must operate in a liquid environment, the team created a graphene oxide ionic sensory memristive device (GO-ISMD). This device enables ions to move within nanoconfined channels, allowing for interfacial adsorption and desorption. This process generates a memory-like electrical response, allowing the device to perform in-sensor computations.

When subjected to voltage pulses, the GO-ISMD replicates synaptic behavior, showing the ability to strengthen or weaken responses and retain memory effects. The thickness of the membrane directly affects the duration of this memory, with some signals remaining for up to 140 seconds, far exceeding predictions based on simple ion movement.

Advancements in Artificial Taste Technology

The researchers employed a technique known as reservoir computing to convert chemical signals into digital patterns. According to Yon Yang, a key researcher involved in the project, “Inspired by the biological taste system, we developed a smart system using our devices to ‘recognize’ chemicals based on their flavors.” This system comprises three main components: a sensing input, a reservoir layer, and a neural network.

The sensing module captures flavors and translates them into electrical signals, which are then processed by the reservoir layer to produce unique digital patterns. The neural network, trained on a computer, helps the system remember these flavors, enabling it to recall them in future interactions.

In their proof-of-concept study, the researchers tested four representative tastants: sour (acetic acid), salty (sodium chloride), bitter (magnesium sulfate), and sweet (lead acetate). The device achieved approximately 98.5% accuracy in distinguishing these flavors, with binary test accuracies ranging from 75% to 90% depending on the sample. It even classified complex mixtures, such as coffee and cola, with impressive results.

Despite these accomplishments, the authors caution that the current setup is still in the experimental stage. The device is described as bulky and energy-intensive, necessitating further miniaturization and circuit integration before it can be deployed in practical applications beyond the laboratory.

Future Implications for Technology

The potential applications of this technology are vast. Yan, another researcher on the team, highlighted its ability to bridge fields such as brain-inspired computing, chemical detection, and biomimetic systems. With advancements in production scaling, power efficiency, and the integration of multi-sensor arrays, this artificial tongue could lead to transformative innovations in healthcare technology, robotics, and environmental monitoring within the next decade.

By combining sensing and computing capabilities in one device, this graphene oxide system marks a notable advancement in the field of biomimetic gustation and neuromorphic engineering. The findings were published in the journal Proceedings of the National Academy of Sciences, signaling a promising future for tools that may enhance or reconstruct the human sense of taste.