Stanford researchers use mechanical testing and machine learning to mimic meat textures, advancing sustainable food options

Stanford engineers have developed a novel approach to evaluating food textures, potentially accelerating the creation of plant-based meats that closely mimic their animal-derived counterparts. Their research, published in Science of Food, combines mechanical testing with machine learning to assess food texture in a manner comparable to human sensory evaluations.

The study, led by Professor of Mechanical Engineering Ellen Kuhl and PhD student Skyler St Pierre, involved subjecting various food samples—including animal and plant-based hot dogs, sausages, turkey, and tofu—to mechanical tests that simulate chewing actions. These tests measured the foods' responses to pulling, pushing, and shearing forces. Subsequent machine learning analysis processed the mechanical data to identify patterns correlating with human perceptions of texture.

The findings revealed that certain plant-based products already replicate the texture spectrum of animal meats. "We were surprised to find that today’s plant-based products can reproduce the whole texture spectrum of animal meats," Kuhl noted. This suggests that plant-based alternatives have advanced significantly in their ability to mimic traditional meat textures.

The motivation behind this research stems from the environmental impact of industrial animal agriculture, which contributes to climate change, pollution, habitat loss, and antibiotic resistance. Transitioning to plant-based proteins could mitigate these issues. However, consumer acceptance remains a challenge, as many meat enthusiasts are hesitant to adopt alternatives. "People love meat," St Pierre acknowledged. "If we want to convince the hardcore meat eaters that alternatives are worth trying, the closer we can mimic animal meat with plant-based products, the more likely people might be open to trying something new."

Traditional methods of assessing food texture often lack standardization and transparency, hindering collaboration among scientists and the development of new plant-based recipes. The Stanford team's methodology offers a standardized, objective, and reproducible means of evaluating food textures, which could streamline the development process for plant-based meats.

This research aligns with broader efforts to reduce the environmental footprint of food production. A study cited by the researchers estimated that plant-based meats, on average, have half the environmental impact of animal meat. Despite this, consumer adoption has been slow; only about a third of Americans in one survey indicated they were 'very likely' or 'extremely likely' to buy plant-based alternatives.

By providing a more efficient and accurate method for assessing food textures, the Stanford team's work could play a crucial role in developing plant-based meats that appeal to a broader audience, thereby supporting environmental sustainability and public health goals.