Hello, my name is Olivier and I am a junior in neuroscience at Michigan State University, but for the rest of the summer Ann Arbor is my home as I work on a research project as a Backyard Brains intern! My goal is to improve upon the roboroach by adding further directional control. Left and right was a good start, but it’s only halfway there! Presently the roboroach consists of an electrode placed on the cockroach that feeds to a ground wire in the thorax and two stimulation wires, one in each antenna, which can be used to electrically stimulate the cockroach to change the direction the cockroach is facing. I am attempting to changes to the device and surgery to make the cockroach move more forwards and backwards using stubby antennae-like structures called cerci. Cerci are located at posterior end of the abdomen and serve as sensors to alert the cockroach to oncoming danger, usually a predator, and initiate an escape reflex that quickly moves the cockroach out of harm’s way. Cerci do this by responding to the changes in air current.
The goal is to commandeer both cerci in order to steer the insect directly forwards or backwards (more…)
We have been slowly rolling out our RoboRoach kits (we started shipping our beta units in late March 2011, and we have shipped 14 so far), and today we received some reports from the field. Robert Uglesich, a Professor at Cooper Union in Manhattan, used our RoboRoach kits this summer to teach high school students about microstimulation, analog electronics, and the neural basis of behavior. He reported today:
“I wanted to write to let you know that the summer program ended today and the students had a blast working with the RoboRoach kits. These last few weeks they were so enthused that they were getting to campus and starting to work before I even came in.”
There was a brief write-up today in the New York Times, but, more interestingly, the students made their own website summarizing their results. Their chief findings were:
1) The cerci nerves adapt more slowly to the stimulation than the antenna nerves. Also, stimulating cerci enables forward motion (our current antennal prep only allows right and left).
2) Randomizing the stimulation to the antenna nerves by using music patterns (instead of the constant 55 Hz stimulation we have set on the circuit) causes habituation to occur more slowly as well.
Thus, a three channel stimulator, with more heterogeneous stimulation patterns, would enable a RoboRoach system with forward, left, and right turning control that adapts more slowly. Nice work students! Impressive! Can we join your lab?
Our current efforts on the RoboRoach are to make the circuit lighter and reduce assembly time. Our current production version (gen3) weighs 6.25 g, an improvement from our first (9 g) and second (8 g) prototypes. It still takes about 6 hours to build however, so we are researching circuit configurations and interconnects that take less time to assemble and test.
We’ve been busy in the trenches of our R&D department designing a working prototype of the “RoboRoach,” a remote control system to enable left and right turning in running cockroaches. This began as a University of Michigan Senior Engineering Project we sponsored in Winter 2010, and over the past year we’ve incrementally been working to improve it, especially after getting great feedback from our colleagues at the latest Society for Neuroscience conference (thanks Cindy). By taking the small, lightweight control circuitry out of the toy “HEXBug Inchworm” and modifying it with low-power 555 timer chips to create biphasic pulses, we can deliver 55 Hz stimulation to the antennae nerves of large discoid and fuscia cockroaches.
We gave a seminar recently on Feb. 25th to the undergraduate students at Grand Valley State University (thanks for the invite John and Merritt!), and we publicly unveiled a working prototype! Samara Woolfolk andStewart Berkos were the first eager students to try it out, and they agreed it worked. We believe in crowd-based peer review of real-time experiments. Being in pure speaking mode, we didn’t take any pictures or videos of the event, but here is Samara and Stewart posing after lunch…
And below is a movie of a basement experiment… Are you convinced?
We plan to offer the RoboRoach as a beta product soon. Stay tuned! In the meantime, we need to do a couple more experiments to ensure:
1) the wires stay in the antenna (principal failure point).
2) the preparation works reliably (at present we only get evoked movement in ~25% of our preps (n=30)…we are working hard to figure out the causes).
Let us know if you have any suggestions/advice! To the NeuroRevolution!