Australasian Science: Australia's authority on science since 1938

Does a Fly Know If It’s in Control?

Credit: Nick Valmas (QBI)

A tethered fly walks on a trackball controlling an object on a digital display, allowing its brain activity to be recorded at the same time. The fly moves the object to the front when it’s paying attention to it. Credit: Nick Valmas (QBI)

By Leonie Kirszenblat & Bruno van Swinderen

What do the brain waves of a fly placed in a virtual reality arena tell us about self-awareness in animals?

When you step on your car’s accelerator, you know that it will go faster. We all know that our actions have consequences, but are animals also self-aware of their actions?

You may find this surprising, but even the tiny fruit fly that hovers around your fruit bowl is calculating her every move. Although her brain is infinitely smaller than a human brain, it is capable of many of the same operations, and may offer some clues to how our own minds work.

Being self-aware and in control of your actions changes how you see and interact with the world. Imagine you are driving to the airport. When you have driven this route several times, you learn how to get there. But if you’re always a passenger you may have no idea how to get there if you are suddenly confronted with the task of getting there on your own. This is one example of how, when you are in control, you pay more attention and learn better than when you are a passive observer.

We know that even insects can learn more efficiently when they learn through their own actions. Yet we still know very little about how our brains might be operating differently when we are in control compared with when we are just observing. This has important implications for learning in the classroom and beyond.

Why would we want to study this in the brain of a fruit fly? Fruit flies have been used in scientific research for more than 100 years, predominantly for genetic studies, and in the past few decades have contributed greatly to the progress of neuroscience. They have much smaller brains than us, containing only 100,000 neurons compared with more than 100 billion in a human brain.

But even though they are relatively simple creatures, they are capable of complex behaviours such as learning, memory and selective attention. To understand these behaviours we can perturb the fly brain in a number of ways that would not be possible in humans.

Since flies cannot tell us what they are paying attention to, we must treat them like human infants, interpreting their behaviours to figure out what they are thinking. This can be done by placing a fly in a virtual reality arena where they can control objects on a screen using their body movements. In the arena, flies can move objects by walking on a ball, similar to how we use a joystick to play a computer game or a steering wheel to drive a car. Flies quickly “understand” that their movements are linked to the virtual objects they see around them, because they adapt their walking behaviour to place objects of interest directly in front of them.

In our study, we let each fly control an object and then we replayed the exact same visual sequence that they were not able to control, thus alternating between control and replay. While the fly was in the virtual reality arena, its brain activity could be recorded by sticking a tiny silicon probe right through the fly’s head to record the electrical signals coming from multiple sites in the brain. The novelty of this type of experiment, aside from the virtual reality environment, was that we could record brain signals across the entire brain, allowing us to examine how the different brain regions interact while the fly was behaving. We hypothesised that there would be differences in brain activity when the fly was in control.

First, we needed to try to understand what the brain of a fly is “seeing” when the fly is paying attention to an object. One way to see how a brain is responding to an object in the environment is to make the virtual object flicker on and off. When an object flickers at a particular frequency, this can be detected in the brain as electrical waves of the same frequency.

In humans, this “frequency tag” in the brain can be used to infer what a person is paying attention to. When you focus your attention on a particular object, the corresponding brain signal gets stronger. This principle can also be used to study what a fly perceives at any given time.

We hypothesised that when the fly was in control of the object, the brain response to the flickering object would be stronger since the fly would be paying more attention to it. Surprisingly, we found that the strength of the fly’s brain response to the object was identical whether it was in control or not. At first we wondered whether this meant that control makes absolutely no difference for a fly; that they are just responding reflexively to changes around them – whether these were caused by them or not.

But when we delved a little deeper by looking at how the different brain regions interact with one another, we found that when a fly was in control over the object there was increased synchrony between brain regions compared with when it saw the same visual sequence but was not in control. This suggests that the fly may perceive the object differently when it’s in control compared with when it’s experiencing the replayed scenario.

The key difference between these two brain states was reflected in how the brain regions interact. Crucially, we would never have been able to discover this if we had just recorded from one brain region at a time.

The biggest effects we saw were in the centre of the fly’s brain, which is where sensory stimulation from the outside world – such as touch, vision and olfaction – converges, and where flies must make sense of all this information to select the best course of action. It has been likened to the basal ganglia in humans, whose disruption has been associated with loss of movement control in disorders such as Parkinson’s disease, as well as neuropsychiatric disorders that affect attention and goal-directed behaviour.

An important implication of our study is that to understand how the brain produces behaviour in the fly, we need to look at how brain regions communicate, particularly those regions that must coordinate a lot of information. In the real world, when animals adapt their behaviour to achieve a goal they must be able to match their visual perception to their own action. When this cannot be done (as in our replayed scenario), this must be recognised by the animal or else its behaviour may become maladaptive.

In other words, animals need to know when they are not in control, such as a fly being buffeted by wind, so that they modify their behaviour to get back in control. In the fruit fly, co­ordination in the central brain is likely to be crucial for matching sensory feedback to actions, which only occurs when the fly is in control.

In the future, we would like to know more about how the brain coordinates information to produce behaviour. The brain uses a variety of neurotransmitters to transmit signals from one neuron to another. We still don’t know whether particular neurotransmitters could give rise to the sort of brain activity we see when the fly is in control.

We suspect that one neurotransmitter that might be involved is dopamine, which is important for goal-directed behaviour and attention in both humans and flies. The advantage of the fruit fly, Drosophila, is that we can ask this question by looking at mutants: what happens to the behaviour and corresponding brain activity in flies that lack certain neurotransmitters, like dopamine, in a virtual reality experiment?

Although the anatomy of the fly brain is very different to ours, understanding how the fly brain operates may help to uncover common principles of attention, self-awareness and learning in humans and other animals.

Leonie Kirszenblat is a PhD student supervised by Bruno van Swinderen at the Queensland Brain Institute, The University of Queensland.