We’ve all been in situations that seem too good to be true: something unexpectedly good happens, and instead of just enjoying it, we pause, looking for the trick or catch.
It turns out monkeys may do this as well. Knight et al. (2013) taught macaque monkeys to accept (or reject) offered treats by pushing (or not pushing) a button. First, a researcher would show the monkey the treat to be offered (a pellet or mini marshmallow), then take it away. The button would then light up, indicating to the monkey that the official offer was forthcoming. Finally, the monkey was presented with the treat again. If he pushed the button, “accepting” the offer, then he would receive the treat. (Unsurprisingly, the monkeys never “rejected” any of the offered treats. But the researchers wanted to make sure the option was available.)
After the monkeys learned the task, the researchers changed it up a bit to see how they would react. Instead of offering the same treat that was first presented, the researchers offered a different treat. So on some trials, the monkeys were offered an unexpectedly worse treat (presented with a mini marshmallow but offered a pellet), while on others, they were offered an unexpectedly better treat (presented with a pellet but offered a mini marshmallow).
On the worse-than-expected trials, the monkeys still accepted all of the offers, but their latencies to accept (i.e. push the button) were significantly longer than when they were offered the same treat that was presented to them. This result agrees with a previous study where macaques exhibited confusion and negative reactions when they found an unexpected less-desired food instead of the expected more-desired food hidden under a cup. And it makes evolutionary sense for animals to desire the outcome (in this case, food) that has a greater value to them.
Based on this logic, we might expect the monkeys to accept the better-than-expected offers faster than the worse-than-expected offers (and possibly even the expected offers). However, the latencies to accept the better-than-expected offers were actually more than twice as long as the worse-than-expected offers. Moreover, the monkeys exhibited significantly more negative responses, frequently avoiding looking at the mini marshmallows by averting their eyes or head (a type of behavior often seen in fear tests).
We know from previous research that animals behave in a way to maximize the value of outcomes (for example, getting the best possible food), so why would the macaques respond negatively to receiving a better-than-expected treat?
Because of another important principle that guides our actions: consistency. Animals, including humans, love routine and consistency. And with good evolutionary reason, as the unexpected can often be dangerous. We therefore like to be able to make accurate predictions about our world, and generally aren’t happy when our expectations are wrong.
So in this case, the simple inconsistency between expectation (presented treat) and outcome (offered treat) could have caused the longer acceptance latencies, regardless of the direction of the outcome (better or worse).
But why were the macaques’ reactions to better-than-expected offers even more negative than those to worse-than-expected offers? The researchers suggest that the macaques may also have a sort of “too good to be true” mentality. After all, unexpected outcomes are much more likely to be negative than to be positive. (If you go to the same banana tree every day, you’re more likely to find fewer bananas and more predators than you are to find an unexpected overabundance of bananas and zero predators.) The monkeys who survive to pass on their genes are the ones who assume all surprising outcomes are bad, and respond cautiously to seemingly good surprising outcomes, just in case.
It seems that, at least in this study, the principle of consistency takes precedence over the principle of maximizing outcome value. This result can give us insight into the kinds of considerations primates take into account when making decisions. Based on the results of this study, if a monkey had to choose between an action with a less predictable but potentially greater outcome value, or an action with a more predictable but smaller outcome value, which action might it choose?
Knight, Emily J., Kristen M. Klepac, and Jerald D. Kralik. “Too Good to Be True: Rhesus Monkeys React Negatively to Better-than-Expected Offers.” PloS one 8.10 (2013): e75768.
Tinklepaugh, Otto Leif. “An experimental study of representative factors in monkeys.” Journal of Comparative Psychology 8.3 (1928): 197.