Mental health research tool: Robot aquarium fish of horror

The live predator in this test was later replaced by a robot
Credit: NYU Polytechnic

Researchers at NYU Polytechnic have achieved a major breakthrough in underwater robotics: They have proven that a robotic replica of something that will eat you can be as terrifying as the real thing.

Fortunately, for the streets of New York abused so heavily in Godzilla and all those Night at the Museum movies, the 'you' they were trying to frighten with a giant predator wasn't a human.

It was a zebrafish – one of dozens of species of fish raised in huge numbers for the domestic aquarium market that are, nevertheless, so similar to one another in behavior and appearance that it's not worth the time it would take for most people to identify one.

During the past decade, however, zebrafish have become important enough that fully qualified, properly funded, real-life scientists can get away with spending a ridiculous amount of time and effort trying to frighten them.

After the Human Genome Project decoded the text – but not the meani

Zebrafish in domestic setting ZFin

ng – of human DNA code a decade ago, someone discovered that 70 percent of the 20,000 genes in a human have a counterpart in zebrafish. Of the genes that cause diseases in humans, 84 percent also exist in some form in the zebrafish.

Zebrafish are also plentiful, cheap and grow up fast enough to keep scientists working rather than wondering how they ended up probing the mysteries of the universe by doing cruel things to zebrafish.

The fish have made some real contributions, too. On Tuesday, Scottish scientists studying multiple sclerosis announced that they had made an important connection showing higher levels of activity in the brain slowed the deterioration of myelin around nerve fibers called axons in people with MS.

They did the test by dripping one of two poisons into tanks of zebrafish and confirming later that the fish whose brain activity had dropped 40 percent following also made less myelin to coat their own nerve cells. That means it might someday be possible to find a way to keep victims of MS and other degenerative diseases from being poisoned by Scottish scientists while swimming.

Researchers at the Translational Genomics Research Institute in Phoenix, Ariz. announced plans to implant tumors into the Zebrafish so they can study the result of treatments for pancreatic cancer by watching development of the tumors through the translucent skin of the fish.

All the crew at NYU Polytechnic wanted to do was see how easy it was to scare a zebrafish, how it was possible to tell a zebrafish was scared, and get a professor of mechanical and biospace engineering to build a robotic replica of the predatory, Amazonian (originally) red tiger oscar to see if it was more likely to frighten the zebrafish than a real oscar of the same size. (Oscars aren't at the very top of a zebrafish's fright list, but they're among the most aggressive fish widely available as aquarium pets in the U.S. If you're a minnow, guppy, zebrafish, goldfish, more than one goldfish, goldfish hoping to be eaten by a piranha, rat, frog, mouse, catfish, thumb, crab, snake or anything else in the water with an oscar, you don't want to be in the water with an oscar.)

The point wouldn't normally be just to scare the fish. The research was designed to establish a baseline of behavior so researchers would know if a zebrafish was mildly stressed or completely freaking out when they're used to help study stress and mental illness in humans. In human trials, the oscar would be much larger and would be able to fire you.

Tiger oscar, waiting for someone to eat. CC

The robot oscar was designed as a CAD model in SolidWorks, then 3D-printed in ABS plastic and spray painted to look like the real thing. A servomotor mounted outside the tank and an Arduino UNO control board to run it made a couple of robot-oscar's fins wave and let it wiggle a bit. Even without being able to swim around, fake-oscar scared the zebrafish even more than the real thing. (Full paper is available here; it is scheduled for June issue of the journal Zebrafish.)

Being able to tell you've terrified a fish isn't enough to justify the experiment or purchase of the servomotor at a local hobby store, however. Extending avoidance-response testing to non-mammal species could verify neurological findings thought to be limited to more complex land animals, the report concluded. It could also demonstrate the validity of much less advanced species for similar experiments.

Finally, the ability to put a robot fish in the main tank would give researchers a way to put the fish and its predator together without the risk of one eating the other – an ethical consideration of the rights of the zebrafish that has left such experiments on the blacklist, until now.

To express your thoughts on Computerworld content, visit Computerworld's Facebook page, LinkedIn page and Twitter stream.
Windows 10 annoyances and solutions
Shop Tech Products at Amazon
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.