Penn State Researchers Develop Novel Electronic Tongue that Advances AI
Artificial intelligence (AI) has made impressive progress in recent years, but it has yet to incorporate the psychological side of human intelligence. To address this gap, a team of Penn State researchers, led by Saptarshi Das, is developing an electronic tongue that mimics how taste influences our food choices based on both physiological needs and psychological desires. This groundbreaking technology could pave the way for more human-like AI systems.
The Importance of Emotion in Artificial Intelligence
While AI has excelled in mathematical models and data sets, it has largely ignored emotional intelligence. Das believes that adding the emotional aspect to AI is crucial for creating more realistic and effective AI systems. However, replicating human behavior and emotional intelligence in robots is challenging due to the difficulties of measuring and observing human behavior accurately.
Exploring the Relationship Between Taste and Eating Habits
Eating habits are a prime example of the interaction between physiological and psychological factors. Our sense of taste, or gustation, heavily influences our food choices based on flavor preferences, which differs from eating solely for nourishment when hungry. Das explains that individuals tend to choose foods they enjoy the most when given the option, even when they are not hungry.
The Development of an Electronic Tongue
The Penn State researchers have created an electronic tongue that replicates the taste perception process. The electronic tongue consists of tiny, graphene-based sensors called chemitransistors, which can detect different chemical molecules. These sensors convert chemical data into electrical impulses, mimicking the function of taste receptors on the human tongue. The research team also developed an electronic “gustatory cortex” using a transistor that remembers past signals. This cortex connects a “hunger neuron,” an “appetite neuron,” and a “feeding circuit.”
By utilizing materials such as graphene for chemical sensing and molybdenum disulfide for circuitry and logic, the researchers have created a versatile artificial gustatory system capable of detecting the five primary taste profiles: sweet, salty, sour, bitter, and umami. This technology holds significant potential, ranging from AI-driven diets based on emotional intelligence for weight loss to personalized meal offerings in restaurants. The research team’s future plans involve expanding the taste range of the electronic tongue and integrating both the tongue and the gustatory circuit into one chip.
The researchers envision this concept of gustatory emotional intelligence in AI translating to other senses, such as vision, hearing, touch, and smell, to further advance AI technology. As our understanding of the human brain improves, these technologies can become even more refined and closely replicate human behavior.
With the support of the United States Army Research Office and the National Science Foundation’s Early CAREER Award, the Penn State research team, consisting of Das, Pannone, Ghosh, Sen, Wali, and Ravichandran, aims to revolutionize the field of AI by introducing emotional intelligence and replicating human behavior more accurately.