Pigeons, Like Humans, Can Behave Irrationally
Gambling may seem like a uniquely human activity. Twinkling slot machines and croupiers in starched white shirts may be about as far from the natural world as we can get. Yet one team of researchers, led by psychologist Thomas Zentall at the University of Kentucky, has taken a particular interest in how animals gamble. The group reasons that if we can identify irrational behaviors in animals, such as gambling, we might discover common brain mechanisms related to such seemingly complex behaviors.
According to behavioral ecologists, Zentall recounts, animals should never gamble because evolution has honed them over many thousands of years into optimal foragers. That is, animals should expend the least amount of energy and time to consume the greatest number of calories. Yet this is not always the case.
In a recent series of experiments Zentall and his colleagues have found that pigeons make some of the same common reasoning mistakes as humans do. For example, they exhibit a strong tendency to select a riskier option over a smaller, safer reward. In one avian version of a casino, pigeons had to choose between a low-probability payoff of 10 food pellets (versus zero) and a high-probability payoff of three pellets. (The expected value is two pieces of kibble in the first case and three in the second.) Although at first the birds chose the more profitable three-pellet option, over time they switched strategies and went for the suboptimal 10-pellet gamble again and again. Research on human gamblers reveals a complementary trend. Compulsive gamblers pay little attention to their losses, tending to remember when they won but not the frequency of winning.
Other studies have shown that pigeons fall prey to the sunk cost fallacy, just as humans do. We might sit through a disappointing movie on the off chance it improves and thus redeems our ticket purchase, or we might stick with a failing business because we hope that our fortunes will change. Similarly, pigeons will continue working on a challenging task to earn a snack rather than switching mid-task to a much easier activity with the same reward. "There's something fundamental about this tendency," Zentall says. "It's not just something cultural for us, such as a belief that we should finish what we've started."
At the annual meeting of the American Psychological Assocation this past weekend, Zentall presented new research on the pigeon version of yet another cognitive bias, the "less is more" heuristic. When making rapid judgments between two things we tend to give greater weight to the average quality of our options rather than the overall quantity. For example, in one famous experiment done by behavioral scientist Christopher Hsee, participants were asked to rate two collections of dinnerware. One set consisted of 24 pristine plates. The other set contained 31 perfect pieces plus nine broken ones. The participants tended to place a higher value on the smaller set--even though the second option contained more flawless dishes.
Rhesus macaques display similar behavior. They like but do not love to eat a slice of cucumber as a snack. Yet if you let a monkey choose between a grape plus a cucumber or just a grape, the monkey will choose just the grape. Like humans, these monkeys appear to judge their choices by the average quality of the offer, rather than the quantity, suggesting that this cognitive shortcut has deep evolutionary roots.
Now for the pigeons. Instead of grapes or plates, the pigeons were presented with peas, which they find delicious. They consider milo seeds, also known as sorghum, less appealing but still palatable. When given the option of either a sole pea or a pea and a milo seed, however, the birds chose the pea and the milo seed. They appeared to behave more rationally than either humans or monkeys.
To look more closely at this surprising behavior, the team divided the pigeons into two groups to see if the birds' level of hunger might play a role. When the pigeons were hungrier,* they made the optimal choice, going for the pea and the milo seed. When the pigeons were only somewhat hungry, they suddenly behaved like humans and chose just the pea. "If it's really important to them they go for quantity," Zentall explains. "If they're not so hungry they go for quality."
Zentall suggests that across species, quality may be easier--that is, faster--to judge than quantity. In the wild pigeons typically face competition from their fellow birds, so the bird that reacts the fastest to the sight of food is most likely to snag the morsel. Our ancestors likely faced similar pressures.
As for why pigeons seem to outperform us some of the time? Zentall suggests that motivation may be the answer. Our biases are not inviolable rules of behavior--they are tendencies we reveal when making quick decisions. When humans are tested in the lab, the stakes are typically very low. Given sufficient motivation, we, too, become more likely to think through a scenario and make the better choice.
*Correction: An earlier version of this blog stated that well-fed pigeons made the optimal choice of the pea and milo seed. It should have stated that hungry pigeons made the optimal choice. Somewhat hungry pigeons were more likely to choose only the pea.