Fighting Fish Offer Clues in How the Sexes See the World

Male and female animals often behave quite differently, but Columbia University neurobiologist Amy Norovich wants to know if they literally see the world differently, too. “We know that males and females of a species can see the same thing and arrive at a different behavioral outcome, but we have no idea how visual information is processed in a sex-specific manner,” she says.

To find out, Norovich studies a unique model organism: Betta splendens, commonly known as the Siamese fighting fish. These gorgeous, long-finned fish — often kept as pets — are infamous for their sexually divergent aggression. Female bettas can occupy the same tank and be completely happy; females and males can coexist in relative peace, too. But two males? “They will fight until one of them dies,” Norovich says. “Sometimes, both will die. It’s a really extreme behavior.”

Norovich is working to characterize the neural circuits that connect the betta’s visual system with the muscles that control its gill covers, which aggressively flare when males see each other. Identifying such circuits could someday explain “how visual information influences social behavior, which is extremely relevant for understanding human behavior,” she says. An edited version of our interview follows.

‌What exactly are Siamese fighting fish, and why are they so aggressive?

Siamese fighting fish are native to Southeast Asian countries, like Thailand, Cambodia and Laos, where they’re found in shallow bodies of water, like rice paddies. Several hundred years ago, people started domesticating them: They would catch them and fight the males against one another, then take the victor and breed it. So hundreds of years of selective pressure has been applied to these fish to increase their natural aggressive tendencies.

How is the fighting fish’s aggressive behavior useful, scientifically?

Aggression in these fish is an innate behavior — something that they engage in without any prior training or experience. As soon as a male fish sees another male fish, it initiates an aggressive response: It’s like flipping a switch. And it’s an incredibly robust response in a way that a lot of other animal behaviors are not. We let the fish see each other for 30 minutes, and the behavior persists for the full period. So, by behavior standards, it’s fairly easy to elicit and quantify.

Another advantage is that aggression in bettas is sexually dimorphic — meaning that it is exhibited by males but usually not females, despite the fact that they share the same genome. When a female sees a male, her response can be somewhat defensive but is ultimately one of interest — she visually tracks the male but will rarely flare her gill covers, for example. So we’re interested in using this species to understand the neural basis of how males and females react differently to the same visual stimulus: in this case, a male fish.

Why not just use traditional laboratory animals, like mice?

Mice are not an ideal model for studying visually guided behavior because they don’t rely a lot on vision for social cues. The most important sensory cues for them are chemosensory: odors and pheromones. Understanding how a set visual stimulus drives behavior — and how this behavioral response differs between males and females of a species — requires a different model organism, one that relies heavily on visual cues and responds to them in a stereotyped manner.

What are you trying to learn from the fish?

At the moment, we’re studying the male fish’s aggressive display behavior. If you keep male fish in separate tanks and allow them only visual contact, they’ll engage in a display that mimics certain aspects of an actual fight; in particular, they extend their fins and gill covers to increase their apparent body size. Studying the display allows us to isolate how visual information drives aggressive behavior without worrying that what we’re seeing is a reflection of other senses engaged during a real fight, like pain or touch.

Right now, we’re identifying regions of the male fish’s brain where neurons are active during aggressive display. It’s a first pass at identifying brain regions that are involved in a male’s response to another male. We’ll then compare these results with the activity in female brains during interaction with a male to identify potential sex differences.

What have you discovered so far?

We’ve borrowed a tool used in mice to probe the brains of these fish for a molecular signature of neural activity. We’ve been successful so far in identifying a few brain regions that show differential activity between fish that have experienced an aggressive social encounter and fish that have not.

What’s the next step to show a connection between vision and this sex-specific behavior?

It’s one thing to identify neurons that are active during behavior and another entirely to place them in the context of a neural circuit. We’re using viruses to map connections between the visual system and downstream circuits that control behavior. We use a virus that has a natural ability to infect the nervous system and travel from neuron to neuron across a synapse that has also been modified to express a fluorescent protein to enable us to visualize the cells. We’re injecting a version of this virus into the fish’s retina that will hop forward from the eye to neurons downstream of visual circuits. We have another version of the virus that travels in the opposite direction, which we will inject into the muscle that elevates the gill cover. So we have two external points of access that will allow us to begin mapping connections between populations of nerve cells.

The idea is to combine our viral tracing data with our map of brain cells that are active during the behavior. We can then start to construct a representation of how visual information ultimately leads to this downstream, sex-specific behavior of gill-cover flaring.

‌Is there an open question that you hope will be answered over the course of your career?

Surprisingly, we don’t understand in great detail how hormones influence brain function. Betta fish provide a unique opportunity for studying this: There’s this quirk of their biology where if you treat a female fish with testosterone, you can turn her into a male — complete with a male behavioral repertoire.

We don’t know at this point if the circuitry behind male aggression is unique to males or if shared circuitry is regulated by other sex-specific means. But something has to switch in response to this hormone to change the behavioral state of the animal. Betta gives us a cool opportunity to examine this in an adult animal in transition between states. That’s something that I think is really promising that no one else is really looking at right now.

Interviews with other society fellows are here.