If the near-starving grasshopper from the childhood fable, the Ant and the Grasshopper, had been given a piece of corn by one of the well-prepared ants, the grasshopper probably would have developed a preference for corn that would have persisted even when he was well-fed.
Based on a joint study between Dr. Spencer Behmer, a Texas A&M University assistant professor of entomology, and researchers at the University of Oxford, the United Kingdom, the grasshopper would likely have developed this preference based on its physical state at the time – its reserves were low, and it was hungry.
“When you’re deprived, things taste better, and their perceived value may be exaggerated,” said Behmer, who has a joint appointment with the Texas Agricultural Experiment Station.
Behmer studied the phenomena of ‘state-dependent learned valuation’ in a grasshopper, the African desert locust, Schistocerca gregaria, at the University of Oxford before moving to Texas A&M in August. He collaborated with Dr. Lorena Pompilio, who was then a graduate student, and Dr. Alex Kacelnik, her major adviser. Their study was published in Science on March 17.
“This work suggests that researchers may need to pay more attention to the state of the subject – whether vertebrate or invertebrate – at the time of learning,” Behmer said.
“African desert locusts were used because they are known to be good learners, which may be important as they are extreme generalists when it comes to their diet – they eat just about any kind of plant,” Behmer said. “Learning allows them to make quick decisions about whether they should eat a particular plant.”
During the study, locusts were trained under one of two conditions – hungry or well-fed – for separate parts of the day (morning or afternoon). In each state, they were repeatedly given a small piece of wheat, and at the same time, exposed to an odor, either peppermint or lemon grass.
Training lasted for three days, and the size and quality of the wheat presented was identical in both conditions. On the fourth day, half of the locusts were starved and half were well fed, and then each was exposed to the odors once again in a ‘Y maze.’
Results showed that the locusts nearly always chose the arm of the Y-maze containing the odor that they experienced while in the deprived state. Interestingly, the well-fed locusts also preferred the odor previously associated with the deprived state.
“The absolute size of the rewards were identical in both states,” Behmer said. “There should have been no preference.”
“The sensitivity of the taste receptors on the locusts’ palps (finger-like projections near the mouth) change according to the internal state of the animal,” he said. “The hairs on the palps, which are similar to a human tongue, enable the locust to taste its food, and if the locust is deprived of a nutrient, these hairs tend to be more easily stimulated when the missing nutrient is eventually encountered. The strong signal from these hairs pairs with a signal from the smell.”
“The locust forms an association between the two and a signal is then sent to the brain, where it gets processed,” he said. “It seems locusts assign greater value to the wheat experienced during the deprived state, and this is why they prefer the odor associated with the deprived state even though no preference should exist.”
A link between preferences and physical state had been made in previous studies of vertebrates – humans and birds. But this link had never been studied in invertebrates.
Behmer said an intrinsic cognitive process doesn’t seem to be involved in decision-making in locusts, whose brains are tiny compared to humans. They contain an estimated 360,000 neurons. The human brain contains 100 billion neurons.
He plans to continue studying learning mechanisms at Texas A&M, using Schistocerca americana, a grasshopper native to Texas.
From Texas A & M