How Humans and Other Animals Navigate Constantly Moving Environments?
Scientists may sometimes come across as slightly unhinged, yes? Reading about their latest research using Virtual Reality (VR) to fool fruit flies, one would be forgiven for thinking so. However, the tiny VR flight simulator used in this specific research has a deeper purpose; probing the inner workings of the insect brain, which can reveal deep insights into human neuroscience.
In a new study published in Current Biology, a long-standing question is reopened about how humans and other animals navigate constantly moving environments
To function in a constantly moving environment, humans and fruit flies (Drosophila melanogaster) both rely on innate behaviour, optomotor response, allowing humans to stabilise their gaze in response to motion.
Example: When you turn the handlebars of a bicycle, the front wheel twists, causing the visual world to sweep past the eyes in the opposite direction. Yet you are able to easily keep your gaze on the road ahead of you.
If the bike was modified so that turning the handlebars right caused the wheel to turn left and vice versa—as in the famous “backward bike” experiment—you would initially struggle to steer it. But, says Simon Sponberg, a biomechanics researcher at the Georgia Institute of Technology, because the human optomotor reflex is flexible, it would eventually adjust.
Do Insects, or in This Case Fruit Flies, React Similarly?
To find out, Benjamin Cellini and Jean-Michel Mongeau, two biomechanics researchers at Pennsylvania State University, stuck a fruit fly in a tiny container and bathed it in cold air until it fell asleep. Then, they carefully glued a metal pin to the animal’s thorax and suspended it in a magnetic field, creating an invisible “tether” that held the fly in place while allowing its body to rotate.
Once securely fastened, the team placed the fly in a small dome lined with LED bulbs, a contraption describes as a sort of planetarium for insects. By generating different visual displays, the researchers could create a “virtual reality flight simulator” that tricked the insect into thinking it was flying.
To test the flexibility of the optomotor response in flies, Cellini and Mongeau used high-speed video to track the insects’ head and wing movements. Feeding the information into the visual display of the VR system allowed scientists to manipulate the fly’s perception of reality and introduce errors between what the insect expected to see and what it actually saw.
For example, when the fly rotated its body, the researchers made the world around it spin faster or slower than normal. Alternatively, if the fly turned to the left, the visual field might also move left instead of swooshing in the opposite direction.
If humans were to encounter this type of “augmented reality” by wearing “upside-down” goggles, our optomotor response would eventually adjust. At first, the conflict between the feedback from our visual system and other sensory input would make us confused, even nauseous. But over time, we would learn how to move around in this altered world. If our visual field spun unusually slowly we would compensate with faster, more exaggerated movements.
But despite spending extended periods of time in the VR domes, the flies never managed to adapt. Instead of altering the speed or direction of their movement to cope with changes in their visual environment, their behaviour remained fixed, even when the world around them began to move in unusual ways.
In some trials, when there was a particularly large discrepancy between what the fly expected and what it saw, the insect would become disoriented and begin to rotate in an erratic, unstable fashion, providing the most convincing evidence that the optomotor reflex in flies is not flexible.
New Data Challenges Earlier Findings
Those results challenge the findings of some earlier papers. The new data suggest the neural circuits that underlie the optomotor response in fruit flies behave differently from those in humans and other vertebrates.
Another important aspect of the research is the authors’ discovery that the flies’ optomotor response behaves very much like a linear system. It means that by using relatively simple mathematical models, the team was able to accurately predict how the insects would behave in different augmented reality scenarios.
Benefits of the Research
This ability to anticipate fly behaviour, along with evidence that the optomotor response is fixed instead of flexible, will provide fundamental new insights into the brain architecture of a genetic model system.
It could also benefit scientists working in the fields of neuroscience, biomechanics and evolutionary biology.
The research might also be of interest to engineers working on animal-inspired robots, potentially paving the way for even “crazier” experiments, creating cyborg insects perhaps?
Current Biology publishes original, peer-reviewed research.
