An innovative “electronic tongue” that can potentially replicate human preferences for certain foods is in development. Researchers at Penn State University say this breakthrough comes as researchers aim to integrate emotional intelligence, which includes likes and dislikes influenced by taste, into artificial intelligence (AI).
AI has advanced significantly over the years. However, its current models largely disregard human psychology, including the nuances of emotional intelligence.
“The main focus of our work was how could we bring the emotional part of intelligence to AI,” says corresponding author Saptarshi Das, associate professor of engineering science and mechanics at Penn State, in a university release. “Emotion is a broad field and many researchers study psychology; however, for computer engineers, mathematical models and diverse data sets are essential for design purposes. Human behavior is easy to observe but difficult to measure and that makes it difficult to replicate in a robot and make it emotionally intelligent. There is no real way right now to do that.”
Eating habits serve as a prime example of this emotional intelligence. While hunger is the physiological drive to eat, our choices of what to eat are influenced by our sense of taste, a process known as gustation. As Das explains, even when one is not hungry, the psychological desire can drive them to consume a sweet treat over, say, a piece of meat.
“If you are someone fortunate to have all possible food choices, you will choose the foods you like most,” explains Das. “You are not going to choose something that is very bitter, but likely try for something sweeter, correct?”
To grasp the intricacies of taste, Penn State researchers looked to the human tongue, which translates chemical data into electrical signals sent to the brain’s gustatory cortex. In this brain region, complex neuronal networks shape our taste perception. To mimic this, the team crafted an electronic “tongue” using chemitransistors made of graphene to detect chemical molecules, combined with memtransistors constructed from molybdenum disulfide. The resultant “electronic gustatory cortex” connects a “hunger neuron,” “appetite neuron,” and a “feeding circuit.”
“This means the device can ‘taste’ salt,” says study co-author Subir Ghosh, a doctoral student in engineering science and mechanics.