Introduction

Imagine you are swimming in the sea on a warm sunny day, immersed in your thoughts, when suddenly a group of people starts screaming and swimming hurriedly toward the shore. What do you do? Quick! This requires a fast decision!

(A) Continue swimming in a relaxed way.

(B) Stay still, in the same position.

(C) Swim as fast as you can to the shore.

(D) Other—what?

What did you decide to do? And why? Let us take a guess. We predict that you interpreted the screaming and swimming behavior of the group as a sign of danger. And this triggered your own defense response which, most likely, was to start swimming as fast as you could to the shore. Correct?

When you first think about it, this could seem a hasty reaction, since you had no direct information that there actually was a threat in the water. So, why did you decide to swim quickly to shore? Imagine you had decided to continue swimming in the sea, when suddenly you noticed a prominent triangle sticking out of the water, coming straight toward you! Ah, a shark! If the screaming and swimming behavior of the group had not triggered a previous defense response, you would only have detected the shark once it was very close, putting your life in danger. Oh no!

This example shows that, in many situations, waiting until you actually experience the danger first hand can be risky. So, we can see that using information transmitted by others can be a clever strategy to avoid encounters with threats, like predators. Many animal species, from invertebrates (animals without backbones, like flies or crabs) to fish, birds, and mammals use a wide range of social cues to communicate the presence of dangers. These cues can include chemicals or sounds [1–3]. For instance, when fish are exposed to chemical alarm cues (like small pieces of skin) released from other fish that are injured, they display defense responses, such as increased use of a shelter or freezing in place [4]. When pigeons sense a threat and take off in flight, this elicits escape responses in other pigeons [2]. Similarly, the fear response of monkeys that are scared by the presence of a snake (like shaking the cage or screaming) can trigger fear in other monkeys as well [5].

Can you see how powerful this cycle of “social transmission” of fear can be? Let us go back to the first example, when the group of people detected the shark in the water and started screaming and swimming to escape the shark. You used these defense responses as a threat warning signal, triggering your own escape. And, consequently, your own response of swimming quickly to shore could have triggered a reaction in someone else nearby. Like the domino effect, the message about an imminent danger can be broadcast across the environment.

Now, the million-dollar question is: how do the defense behaviors displayed by someone that detects a threat influence the defense behaviors of an observer that has not detected the threat directly? What exactly is the information from the threat-detecting person that the observer is using? Could this information be visual or auditory? Or olfactory? Or tactile?

From the Wild to the LAB

To answer these questions, we first needed to recreate this situation in the lab. We needed to expose an individual to an unpleasant event that triggered a defense response that could be detected by others (the observers). Then we could see if the detection of that defense response led to a defense response in the observer, just like the example with the shark.

To recreate this situation in the lab, we used a manipulation called fear conditioning, in which animals learn to predict unpleasant events. What does this mean? Do you know the story of Pavlov’s dog, and how it learned to associate a sound or light with getting food? Pavlov was a Russian psychologist who was studying drooling in dogs. He developed a setup in which he would play a sound (or flash a light) for dogs before feeding them, and then he looked at how much the dogs drooled. Now, imagine that, after doing this for a while, Pavlov just played the sound without presenting the food—how do you think the dogs reacted? When Pavlov did this experiment, he discovered that dogs began to drool just by listening to the sound. Why? Because they learned that the sound always came before the food.

Fear conditioning uses the same type of manipulation, except the animals learn to predict unpleasant, instead of rewarding, events. The protocol consists of playing a sound, which is then immediately followed by an unpleasant stimulus—a mild electrical shock (Figure 1A, Step 1). After a set of sound-shock pairs, the sound itself elicits fear responses, because the animal becomes able to predict the shock (Figure 1A, Step 2). In our experiments we used rats, which are social animals that live in groups. This is important, because we wanted to study social transmission of fear between individuals in a group.

So, this was our first step: to condition rats to fear a tone (represented by *bip) after it had been paired with an unpleasant shock (represented by *bzzzt) to the foot (Figure 1A).

Now, which type of fear responses do rats display in this situation? When there is a threat, animals can display different types of defense behaviors such as flight (running away), fight, or freeze. However, when the animals are in a confined space with no way escape, most vertebrates (animals with backbones, like our rats) tend to stop moving. This is a defense response that might allow them to avoid being detected. If the animal does not move, it would not call attention to itself. Since, in our experiment, rats cannot escape from the box, that is exactly the defense behavior that we observed when the tone was played—rats became motionless (represented by *freeze…; Figure 1A, Step 2).

Looks like we are ready to tackle our question: how does the freezing behavior displayed by rats that have learned to fear the tone, called the demonstrators, influence the behaviors of other rats for whom this sound does not mean anything, the observers?

I Freeze. Do You Freeze?

To study this question, we put two rats together in a box, called the social box—one demonstrator and one observer. This box had a partition in the middle, separating the rats, but still allowing them to see, hear, smell, and touch each other (Figure 1B, Day 4 social interaction).

Ready to perform the experiment?

Let us just briefly recap: we wanted to look at the defense response of the observer rat while witnessing the defense behavior of the demonstrator, which has previously learned to fear the tone. For that, we placed the rats in the social box, and after a few minutes we played the tone. Next, we looked at the behavior of both the demonstrator and the observer rats (Figure 1B, Day 4 social interaction).

So? What happened? As expected, when the demonstrator rat was presented with the tone (represented by *bip), it froze (Figure 1C). What about the observer rat? How did it react? Once the demonstrator started freezing, the observer rat also displayed a defense response of its own, which was ... any guess? Yes! It also froze (Figure 1C).

Does this remind you of anything? These observations are similar to the example with the shark, where the group of people were the demonstrators, you were the observer, and the defense responses were screaming and swimming hurriedly. But wait! Here is an important catch: not all the observer rats froze—only the ones that previously experienced the foot shock themselves did. What does this mean? Imagine that you did not know that there are dangers in the ocean. Would you have reacted the same way to the crowd? This gives us important information: to be able to respond to the fear response of the demonstrator, observer rats need to have previously experienced the stimulus that caused the fear—the foot shock, in this case.

To summarize what we found, observer rats responded to the defense response of the demonstrator with a defense behavior of their own (freezing). But this only happens if the observer rats had previously experienced the fear-inducing stimulus.

So the next question we asked was, which sensory cues of the demonstrators could be triggering freezing in the observer rats? In other words, how is fear being transmitted from the demonstrator to the observer?

I Freeze, You Freeze. But How Do We Communicate?

Could it be that the demonstrator is warning the observer through some kind of “verbal communication”—making some kind of sound that warns the observers? Previous work from other scientists showed that rats emit alarm calls in threatening situations, such as when they are in the presence of predators. These calls could potentially warn other rats of approaching threats; however, it is still not clear how important these vocalizations are in fear transmission. So, we decided to investigate this!

The alarm calls of rats cannot be heard by humans, since they are emitted in a frequency above the general range of the human ear. So, to “listen” to what rats were “telling” each other, we used two special microphones, one placed over each side of the social box. When we performed the experiment, we found that only one pair of rats, from the eight pairs tested, emitted distress calls. This shows that, in these conditions, the demonstrator rats do not need alarm calls to communicate fear to the observer rats.

So ... what else could be sending the message to the observer rats? Could freezing by itself be the cue?

Imagine you are in a room with someone else and suddenly that person stops moving. What type of information allows you to detect the transition from movement to immobility? Well, since you can see it, it is the visual information that allows you to detect that the other person stopped moving. So, if that is the case, what would happen if the lights were off?

When we performed the same experiment, but in the dark, we found that both demonstrators and observers still froze when the tone was played. In other words, visual cues were not necessary for the observer rats to respond to the fear of the demonstrator. Surprised?

Time for another quick summary: rats were not “talking” to each other, and visual cues were not crucial for the social transmission of fear. What else might be happening to transmit the fear signal, then?

Let us give you a hint. When someone stops moving, there is a transition between sound, caused by movement, and silence, due to freezing. When rats are in the social box, they normally move around, producing rustling sounds. But when the demonstrator rat freezes, there is silence, since the sound disappears when there is no movement.

Could it be that the observer rats detected the demonstrator rats freezing because of the transition from the sound of movement to silence?

To test this hypothesis, we first recorded the sound of a rat exploring the box for a few minutes (Figure 2A, Step 1, represented by *swish). Then, we placed the demonstrator and observer rats in the social box and played the tone (Figure 2A, Step 2, represented by *bip). After the tone was played, when both the demonstrator and the observer were already freezing, we played the sound of the rat moving around (Figure 2A, Step 3, represented by *swish). This experiment was done in the dark, so that observer rats could not see what the demonstrator was doing.

What happened? What would you expect? If silence was the cue that caused freezing in observers, then you would expect that playing the sound of the rat moving should prevent their freezing. And … that is exactly what happened! When we played the sound of a rat exploring the box (represented by *swish), the observer stopped freezing (Figure 2B, red bar). And importantly, freezing restarted immediately when the sound stopped playing—during silence (Figure 2B, after the red bar).

Aha! Interesting, right? This tells us that the observer rats seemed to be using the “sound” of silence to detect freezing by the demonstrator rat.

Finally, if this is indeed the case, we should be able to cause freezing in observer rats that are alone in the social box, just by simulating the changes in sound produced by the movement of another rat. So, we tried this experiment. First, we placed a rat alone in the social box and played the sound of another rat moving around (Figure 3A, represented by *swish). Then, we played the “sound” of silence, to mimic a rat freezing (Figure 3A, represented by *silence). And, “Eureka”! When we performed this experiment, we observed that rats froze during the moments of silence (Figure 3B, white bars). Cool, is not it? The onset of silence is enough to cause freezing in rats. Let us put the pieces together. Observer rats freeze in response to the display of freezing by demonstrators. And this social transmission of fear does not rely on alarm calls or on visual cues, but instead, on the transition from sound to silence, due to the lack of movement of the demonstrator rats.

Now, Back to the Wild: I Freeze, You Freeze, He Freezes

Why is this discovery important, especially if you are an animal in the wild?

Animals use auditory cues to detect the presence of a predator through the behavior of other individuals, e.g., alarm calls [3]. However, a drawback of this kind of auditory signal is that, when an alarm call is emitted, the animal sounding the alarm is protecting others but also calling attention to itself. Freezing, on the other hand, is a behavior that animals already use to defend themselves, to avoid being detected, but—as we show here—freezing can also serve as a cue to warn other animals of an approaching threat. Since most vertebrates freeze when threatened, this behavior is a cue that can spread very quickly in an environment. Also, this is not just true for rats—the other animal could be a mouse or a dog. Silence could be used to rapidly spread the news of danger between different kinds of animals.

So, the take-home message of our study is this: be cautious when there is a sudden silence. Shhhh…

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.