Editor’s note: This is Part II. You can read Part I here.
Hello, I am Danae Madariaga, a senior at Alberto Blest Gana high school. I have participated in a data collection project with Etienne, Tim, and Derek for three months. Throughout this time, I have learned many things such as the use of Google Colab to analyze my data that I uploaded to the cloud. This makes it easier for scientists from around the world to analyze my data as well.
Really though, the most important thing that I have learned is that being a scientist is not easy! It is a very hard job that requires perseverance and patience. I have also learned how to optimize my time to perform my experiments in a consistent manner. Working with Backyard Brains was my very first job and a very pleasant experience, especially with all the new tools and plants I have now!
I aim to study the neuroscience behind this question by replicating past studies on grasshopper vision. Grasshoppers can sense an approaching object and quickly hop away to avoid collision with the object because their visual system includes a type of neurons (descending contralateral movement detector, DCMD) that underlies the animal’s visual and motor sensitivity to approaching objects, such as predators. I use Backyard Brains’ open source SpikerBox and the SpikeRecorder iPad app to record and visualize the activity of the DCMDs in the form of electrochemical action potentials, or spikes.
Throughout this project, I aim to bring neuroscience research out of the far, far away university labs and design and perform a low-cost and reproducible project using open source and DIY tools, to explore and learn from the neural basis of the grasshopper’s escape mechanism.
Somewhere in Portland, there’s a restaurant that serves grasshopper sushi rolls. Is it safe? Is it good? I don’t know!
Because I am a grasshopper researcher this summer in Ann Arbor, Michigan, I have other questions in mind: How do people catch these bugs? If you’ve ever tried to catch one, you know that it is nearly uncatchable when its skeletal muscles get to work as you approach with silent steps, trying to capture it for an afternoon snack.
Catching grasshoppers in Ann Arbor is my exciting challenge this summer. Finding out why they are hard to catch is my neuroscience project.
Why are they hard to catch? Because they can quickly jump away when a person or another insect or object approaches it. How are they able to quickly hop away to escape a potential predator or avoid collision with an object? To address this specific question, I will look into the movement detector neurons in the grasshopper’s brain—the organ that fascinates me.
Just as I can see it with my eyes, the grasshopper can see me if I come to it. Or if I show it scenes from Star Wars when spaceships are flying toward the viewer, the grasshopper can see them too and would hop, hop away. That is what researchers Rind and Simmons found in 1992 in their research on the vision of the locust, or a kind of grasshoppers that form swarms. The grasshopper’s nervous system includes a type of visual neurons, called descending contralateral movement detector (DCMD), that receives visual info from the eyes and sends that info to the legs, and underlies the grasshoppers’ ability to visually detect and react to an approaching object, be it a spider looking for a crunch or an astronomically speedy spaceship.
In human language, the brains of these bugs are capable of serious mathematics. In a paper published in 1995, researcher Hatsopoulos and colleagues came up with an equation that describe how the DCMD neurons sense and respond to approaching and receding objects: velocity, or speed, of the approaching image:
multiplied by an exponential function of size of object’s image on the retina: On a high-level consideration of the computational way the brain of the grasshopper functions, the activity of the DCMD neuron is related to how fast the image is coming toward the eye of the grasshopper and the image size on the eye that changes with decreasing distance between object and the eye. The peak in firing is reached before the collision of the object and the grasshopper, and the bug can leap away using their legs to avoid being hit or eaten.
In the world of scientific research, disagreements founded upon experimental evidence and thoughtful arguments give rise to scientific progress. In the two above-cited papers, I see several discrepancies between the two groups of researchers. While Rind and Simmons concluded that there was good correlation between the neuron’s activity and the object’s acceleration during the exposure of the grasshopper to approaching objects, Hatsopoulos and colleagues used both their computation and experiment to conclude that the correlation was poor. The two papers generally agree that the DCMD neuron’s responses depend on the size and speed of the object. Keeping these ideas in mind, I will see what results my project will yield and I look forward to contributing to the discussion.
Art by Tanner @ All Hands Active, Ann Arbor, MI
I hope to demonstrate that a fun and educational neuroscience project can be done outside of the far far away university labs! I use Backyard Brains’ Neuron SpikerBox that amplifies and visualizes the activity of the DCMDs in the form of electrochemical action potentials, or spikes. I also have an iPad with the SpikeRecorder app, which provides visual stimuli (growing or receding black dots on a white background) in front of the grasshopper’s eye as well as records the DCMD activity.
Hop, hop away. This is how the grasshopper stays alive. This is how it continues to exist and eats plants and destroys our crops. But it is also our food and art and stories. And it is my friend (euphemism for “study organism”) this summer. Please check out how I chase, catch, maybe eat, and perform electrophysiology on the grasshoppers to record the activity of the DCMD neurons in an open-source and DIY style.
iPad: ~$400 or if already have – As screen for visual stimuli and recording and visualizing spikes, in a cumulative SpikeRecorder app
Microscope: $200 or ask around or use a phone app! – 20X magnification is sufficient; For surgery on grasshopper and electrode placement
Magnetic stirrer: $100 or be creative! -To heat, melt and mix the wax and rosin mixture
RadioShack mini speaker/amplifier: $15 – To hear the spikes when the grasshopper sees the balls being thrown at it
Backyard Brains Neuron SpikerBox: $100 or DIY – A bioamplifier that allows us to hear and see spikes in living neurons; has cork board piece on top for mounting animal
Backyard Brains Micromanipulator: $100 or DIY – For precise placement of electrodes
1-Channel Electrodes: Reference and Recordingincluded with manipulator or DIY – Hook electrode around neck connective of grasshopper and reference electrode in abdomen, both connected to SpikerBox
Painter’s Tape$2 – To restrain grasshopper and tape it down to corkboard piece on the SpikerBox
Thread: $2 – To pull grasshopper’s neck up to expose neck where recording electrode is placed
Wax and violin rosin mixture: $6 – Heated to be mixed together; for placing on thread around grasshopper’s neck to keep it in place
Build the SpikerBox, micromanipulator, and electrodes. Build instructions for these items are in the files section! Gather the materials:
Anesthetize the grasshopper by placing it in a plastic container (also its home in the lab) and keeping in the fridge (not freezer) for 15-20 minutes or until it is inactive. This keeps the animal still and painless (if insects indeed feel pain) during the upcoming surgery. (Dragonflies are also our faves.)
After anesthesia, tape the grasshopper belly up on the cork board piece on the BYB SpikerBox. Tape all the legs and the abdomen. Leave the head and a little of the thorax exposed—these areas are where electrodes will be placed. (I find masking/painter’s tape to be the easiest to work with.)
Then, use a piece of black thread to pull the head back until the neck connectives (two white strips under the neck skin) are visible. This step is tricky, so be patient. Also, the thread must not block the eye opposite of the side of the neck that the recording electrode will be placed. In the picture above, I plan to put the recording hook electrode (where needle is pointing in photo) around the left neck connective of the grasshopper. So the right side of the grasshopper (our left) must not be blocked by anything.
Create a 50/50 mixture of rosin and bee wax in a glass petri dish. Put the dish on magnetic stirrer, on very low heat (I’ve broken several dishes due to ignorance of kitchen basics). After a few minutes, the solid mixture is melted into a liquid.
Use a needle to pick up the warm rosin-wax liquid and place it around the thread holding back the head of the grasshopper. When the liquid cools and molds into a wax texture, the thread is secured to the neck of the grasshopper.
Now, another step as tricky as pulling back the grasshopper’s head: Placing the hook silver wire electrode around the neck connective to pick up the activity of the DCMD neurons. The BYB manipulator electrode comes as a simple straight wire. For this experiment, I modify it by using tweezers to bend the tip into a small hook, to place around the connective.
Then, under the microscope, use a tiny needle to pierce a hole on the left side of the grasshopper’s neck, next to the connective. Then, carefully place the hook electrode into the hole and use the BYB Micromanipulator to adjust so the electrode is deep enough to hook around the connective.
Depending on the grasshopper and the spot in the neck where the hole is made, there might be green blood oozing out from the hole (top photo). If there is no blood coming out, I immediately put a drop of Vaseline on the recording electrode and nerve cords to isolate them from the rest of the body and keep the pierced hole from drying out (which would otherwise happen within minutes).
Connect the BYB manipulator electrodes to the SpikerBox. Then, place the reference electrode (needle) either on the mid-thorax or abdomen. I find that the reference electrode grounded in the thorax yields better signals.
The prep is now finished! It looks something like this:
Almost done! On the iPad SpikeRecorder app, set the parameters for a new experiment. Here, I am testing the response of the DCMD neurons to an approaching black ball on a white screen. The balls of different sizes will approach the grasshopper’s eye at different speeds. Between each trial (each trial is a pair of an approach velocity and an object size) is an intertrial interval of 45 seconds for the neuron to fire again, based on literature and my own experiments.
The black balls originate from the center of the iPad screen and fill up the screen to simulate approach and collision with the grasshopper’s eye.
After the iPad is ready, tape it upright to the wall so the angle between the center of the screen/ball and the center of the grasshopper’s eye is minimal. Use a level and ruler to measure. Then, connect the iPad to the SpikerBox (green cable) and also connect a RadioShack mini speaker to the SpikerBox (blue cable) to hear the spikes. Then, turn off the lights and close the doors to the experimental room for good contrast of the black ball on the white screen.
Otherwise, make a portable experimental room! I use a card board box from the recycle bin of a restaurant. I turn on the iPad and close the flaps, and the ball begins to come at the grasshopper’s eye. And the grasshopper’s DCMD neuron will activate. And I will get experimental data!
Note: You might know that electrophysiology recording is troubled by lots of noise,including body and other electronic sources. Choose a spot without many electronics being used, and minimize body movement. I stay at least 5 feet away and sit as still as possible while the experiment is going on.
Finally, when the experiment is over, thank the grasshopper. Unhook the electrodes, untape it from the cork board, and bring it back to the field where it can feel most at home.
With the ideal ITI determined, I can move on to the set of core experiments: testing to see how the DCMD neuron behaves when simulated black balls of different sizes and velocities approach the grasshopper’s exposed eye. So my little friends spend about 2 hours on top of the SpikerBox for these experiments.
I continue to process the data in MatLab for better visualization. Here are the results for balls approaching from a constant initial distant of 10cm, 6cm in size, and with various velocities (-2, -4, -6, -8m/s).
Perievent histogram: showing DCMD firing frequency 2s before and 2s after the simulated collision between the eye and the object:
Raster plot: showing DCMD spiking pattern across each pair of S and v over time. DCMD firing peaks around collision for objects approaching at -2m/s, and after collision for objects approaching faster: