Look Who’s Talking

Although they don’t have language the way humans do, animals can clearly communicate with each other (otherwise, why would they make sounds at all?). We usually view animals’ vocalizations as reflexive (a yelp in response to pain) or very general (a threatening growl), but research in the past few decades has shown that animal vocalizations can be remarkably specific.

320px-Präriehund_P1010308Some of the most specific communication has been found in the alarm calls of prairie dogs. Prairie dogs live in burrows in the ground, making them vulnerable to predators, so it’s very important for them to be able to communicate the presence of a predator to the rest of the colony. Moreover, because different predators have different attack strategies (a hawk would attack from the air, whereas a dog would attack from land), it’s also important for prairie dogs to communicate what kind of predator is nearby so the most appropriate escape response can be made.

Researchers have found that prairie dogs do indeed make different alarm calls depending on the species of predator spotted. However, just because a prairie dog makes a predator-specific alarm call doesn’t mean that other prairie dogs interpret it as such; it’s possible that prairie dogs can make distinctions between predators when giving alarm calls yet not respond to these distinctions when hearing other prairie dogs’ alarm calls. They could rely on other cues, such as the behavior of the prairie dog that made the alarm call or the predator itself, to make the appropriate escape response.

In order to test whether prairie dogs can distinguish between different types of predator calls, researchers ran a playback experiment. First researchers recorded alarm calls made by prairie dogs in response to seeing different types of predators. They later played back these calls in the absence of a predator and looked at the prairie dogs’ responses.

4838957259_c0a458c8ba_mIf the prairie dogs truly understand that different types of calls are associated with different types of predators, then they will respond appropriately based on the predator. The researchers found that the prairie dogs did just this. When they heard an alarm call to a hawk, they immediately ran into their burrows, the typical response to a hawk predator. When they heard an alarm call to a dog, they became alert but didn’t immediately enter their burrows, just as they would respond to the presence of a dog.

These results demonstrate something very important about prairie dog alarm calls: these calls are referential, meaning that they refer to an object (or in this case, animal), rather than being reflexive. This is significant because, until the last few decades or so, it was thought that the referential nature of human language was one of the characteristics that set it apart from other animal communication.

240px-Kissing_Prairie_dog_edit_3Further research on prairie dog alarm calls has shown just how specific they can be. Multiple studies have shown that prairie dogs have different alarm calls for different humans, and for the same human wearing different colored shirts, suggesting that prairie dogs can encode color and shape information into their calls. Researchers have also found regional differences between calls, where the acoustic differences between calls for the same predator are greater between colonies that are geographically farther away.

And prairie dogs aren’t the only species with specific vocalizations: vervet monkeys, Diana monkeys, Campbell’s monkeys, ground squirrels, and chickens (to name just a few) have all demonstrated different alarm calls for different types of predators.

So if these vocalizations are so specific and referential, do they count as language? It’s definitely a contested issue (as you would expect). Few scientists would argue that these vocalizations are the same type of language as that of humans. However, these vocalizations suggest that the cognitive structures involved in human language may have evolved earlier than we previously thought. And they’ve certainly made scientists rethink what exactly a “language” is!


Here’s an interesting video about prairie dog alarm calls by Con Slobodchikoff, who has done nearly all of the research out there on prairie dog vocalizations. And here’s a video with Robert Seyfarth (who, along with Dorothy Cheney, has done an incredible amount of research on vervet vocalizations) about whether such specific vocalizations count as language. Finally, here’s a video of vervet alarm calls in action!


Sources Cited:

Ackers, Steven H., and C. N. Slobodchikoff. “Communication of stimulus size and shape in alarm calls of Gunnison’s prairie dogs, Cynomys gunnisoni.” Ethology 105.2 (1999): 149-162.

Frederiksen, J. K., and C. N. Slobodchikoff. “Referential specificity in the alarm calls of the black-tailed prairie dog.” Ethology Ecology & Evolution 19.2 (2007): 87-99.

Kiriazis, Judith, and C. N. Slobodchikoff. “Perceptual specificity in the alarm calls of Gunnison’s prairie dogs.” Behavioural processes 73.1 (2006): 29-35.

Slobodchikoff, C. N., S. H. Ackers, and M. Van Ert. “Geographic variation in alarm calls of Gunnison’s prairie dogs.” Journal of Mammalogy (1998): 1265-1272.

Slobodchikoff, C. N., and R. Coast. “Dialects in the alarm calls of prairie dogs.” Behavioral Ecology and Sociobiology 7.1 (1980): 49-53.

Slobodchikoff, C. N., C. Fischer, and J. Shapiro. “Predator-specific alarm calls of prairie dogs.” American Zoologist 26 (1986): 557.

Slobodchikoff, C. N., et al. “Semantic information distinguishing individual predators in the alarm calls of Gunnison’s prairie dogs.” Animal Behaviour 42.5 (1991): 713-719.

Slobodchikoff, C. N., Andrea Paseka, and Jennifer L. Verdolin. “Prairie dog alarm calls encode labels about predator colors.” Animal cognition 12.3 (2009): 435-439.


When Less is Worth More

Often when problem solving, we don’t have the time or cognitive capacity to painstakingly evaluate each solution. Instead, we make an educated guess or rely on our common sense or a rule of thumb. These shortcuts (called “heuristics”) may save us time and cognitive processing, but they don’t always lead to the optimal solution.

Take, for example, the “less is better” effect: valuing a single, high quality object more than that same high quality object plus an object of lower quality. The addition of the lower quality object somehow decreases the subjective value of the higher quality object, even though the quantity of the two objects is greater than the high quality object alone.

286px-HFG_Hans_(nick)_Roericht-_TC100-_dish_set-dhubThis may seem ridiculous, but humans have clearly demonstrated this effect. In one experiment, Hsee (1998) asked participants to indicate how much they would be willing to pay for a set of dishes. In one condition, the set contained 24 dishes. In another condition, the set contained 40 dishes, but 9 of them were broken. Even though the second condition contained more unbroken dishes than the first, the amount that participants were willing to pay for the first set was greater than for the second set. The inclusion of some broken dishes decreased the value of the second set, even though it actually contained more unbroken dishes than the first set.

It’s important to note that participants weren’t comparing the two sets of dishes directly; each participant was only asked to name a price for one set of dishes. Their responses, then, were based on their valuation of the particular set of dishes without additional information or comparison.

When Hsee asked participants to name prices for both sets of dishes, the “less is better” effect disappeared. Instead, participants said that they would pay more for the larger set of dishes than the smaller one. This indicates that humans use comparison information (when it’s available) to make valuation judgments, and we can use this information to overrule heuristics and arrive at the optimal choice. But when we don’t have that comparison information, our heuristics can sometimes steer us wrong.

What about nonhuman animals? Do they rely on similar heuristics when problem solving, and thus make similar suboptimal choices?

Kralik et al. (2012) investigated whether rhesus macaques would also demonstrate the “less is better” effect. In both the lab and a more naturalistic setting, they gave the macaques two options: a more-preferred treat (like a grape), or a more-preferred treat AND a less-preferred treat (like cucumber).

8774415224_ac7e9deb0e_mThey found that the macaques chose the grape alone significantly more than the grape + cucumber. At first glance, this doesn’t make sense; when given the choice between more food and less food, it’s more evolutionarily advantageous to choose more food. However, it seems that the macaques were more concerned with the quality of the choice than the quantity, and somehow the presence of the less-preferred cucumber lowered the quality of the more-preferred grape. So the decision was instead between lower-quality food (grape + cucumber) and higher-quality food (grape).

It turns out that focusing on quality more than quantity might actually make evolutionary sense: research on the foraging behavior of animals suggests that, when food is scarce, focusing on the quality of food in a particular location (rather than quantity) may lead to more optimal foraging decisions (Do I keep eating at this tree or move on to another one?). Additionally, when food is abundant, averaging the quality of all the food in a particular location will maximize outcome.

So it seems that when solving this problem, macaques rely on a heuristic that may maximize outcome in the wild, even though it doesn’t maximize outcome in this particular task. And, unlike humans, macaques aren’t able to use comparison information to override the heuristic and make the optimal choice. Perhaps factoring in comparison information requires higher cognitive processes that macaques don’t possess, or perhaps macaques just need more experience with the task in order to factor in comparison information.

2802429078_ec607610f9_mEither way, the results of these experiments suggest that some heuristics may have an evolutionary origin that predates the split between humans and other primates. And they may have developed even earlier than that: Pattison & Zentall (2014) recently found that dogs also demonstrate the “less is better” effect!

If you’ve enjoyed reading about animal cognition research on this blog, check out Inside Animal Minds, a short NOVA series on animal cognition. The first episode airs tonight at 9 pm on PBS!


Sources Cited:

Hsee, Christopher K. “Less is better: When low-value options are valued more highly than high-value options.” Journal of Behavioral Decision Making 11.2 (1998): 107-121.

Kralik, Jerald D., et al. “When less is more: Evolutionary origins of the affect heuristic.” PloS one 7.10 (2012): e46240.

Pattison, Kristina F., and Thomas R. Zentall. “Suboptimal choice by dogs: when less is better than more.” Animal cognition (2014): 1-4.