Gooooooooal! psych students teach rats to play soccer … well, sort of
Posted on December 07, 2007 by Suzanne Ramsey
No one expected them to bend it like Beckham. They were rats after all — not highly trained athletes — with itty bitty legs, beady red eyes and no detectable soccer skills. But psychology students set out two weeks ago to teach them how to play soccer, or at least how to push a sucrose-coated pingpong ball into a goal.
Soccer playing rats was the final project for students in Dan Gottlieb’s animal learning class. Although Gottlieb didn’t expect the experiment to result in anything “scientifically useful,” he hoped it would be a fun project and the students would discover something about how rats learn.
“They all have been training their rats to push a ball into one of two goals,” Gottlieb said prior to the end of the project. “Then we put all the rats together and see what happens. I’m skeptical that it will work, but who knows?”
The experiment culminated Dec. 5 with a two-on-two soccer match held in Guion’s psychology lab. On game day, the students and the four-legged footballers gathered in the lab around a tiny soccer stadium, complete with a midfield line and two goalie boxes painted on the chalkboard green “turf.”
The arena was constructed by Alison Carr ’08, as evidenced by a black fingernail she sported, thanks to a mishap with a power tool.
The eight rats were divided into two teams — the Green Team and the Red Team — and each wore a number painted on its back in its team’s color. Four would play at a time. “We decided if we put them all in at once it would be chaos,” Gottlieb said.
Most of the rodents were named for famous learning theorists — Robert Rescorla, Burrhus Skinner, Michael Domjan, Edward Thorndike, Ivan Pavlov, Randy Gallistel, John Gibbon and Hermann Ebbinghaus.
The exception to this was Butters, who was originally named for scientist Allan Wagner.
Butters would eventually become a prolific scorer with five goals to his credit during the game, although he didn’t always score for his own team. “No, Butters! No!” was shouted more than once as the rat rolled the pingpong ball into his own goal.
Initially, the class tried to “clicker train” the rodents, but “they were scared of the clicker,” Carr said. Instead, the students — Abby Adkins ’09, Kia Bryan ’09, Doe Buchli ’10, Alison Carr ’08, Alex DiFeliceantonio ’08, Anna Fure ’08, Molly McLemore ’10, Cheryl Seaver ’09 — used food and lots of repetition.
“During the first week, I wanted to get him accustomed to interacting with the ball, so each time he moved the ball with his nose or paw I rewarded him with food pellets,” Bryan said of her rat, Herman.
“The next phase was to train Herman to put the ball in the goal. At first, I trained Herman in a small training box that was placed in front of the goal. When he put the ball in the goal, I rewarded him with food pellets.
“Next, I removed the box and allowed him to roam around the field. Herman would get very distracted, so I would either put pellets under the ball to entice him, or simply tap on the ball to get his attention.”
In time, Bryan said, Herman became adept at scoring goals and even learned to distinguish between the two goals, one of which was painted black and the other, white.
Carr, who trained Robert, called the training technique “shaping.”
“We rewarded them for successive approximations of pushing the ball — first for being near it, then touching it, then pushing it, etcetera,” Carr said. “They had to learn on their own how to actually get it in the goal.
“Even if you couldn’t tell in the final game, they all had it pretty well down. Three out of four [on the Red Team] learned to move it by grabbing it with their paws and dragging it instead of pushing it with their noses.”
In the first half of the soccer match, however, rats on both teams scurried around the field and seemed more interested in each other than the ball. “C’mon … or not,” one student pleaded as her rat, Randy, showed and then lost interest in the ball.
Eventually, Green Team captain, the aforementioned Butters, scored a goal for the Red Team. Shortly thereafter, Butters scored again, this time for his own team. “Butters is a genius,” Gottlieb announced wryly.
At the end of the first half, the game was tied 1-1.
In the second half, Butters scored three more times, twice for the Green Team and once for the Red Team. Herman also scored three goals for the Green Team and the game ended with a 6-2 victory for Green.
“I learned it is a whole lot harder to train a rat to push a ball into a goal than I would have ever thought,” Carr said later. “Actually, the biggest surprise was how much their behavior changed every time a new rat was added to the situation.
“Our team scored tons of goals in a team training we had the other day, but the context seems to become novel again with the addition of the Green Team on game day … because they weren’t nearly as good as before.”
Overall, Gottlieb thought the experience was a positive one. “I thought the experiment went well in that the students took it seriously and put a lot of time into it,” he said. “The goal of the experiment was simply to have the students work on a project in which they would need to train their animals over the course of multiple days. Most labs for the course are one-day labs, which is difficult when we are interested in learning processes.
“By observing the behavior of their rats, day after day, I was hoping the students would learn something about the difficulties and constraints faced by researchers studying basic learning mechanisms. I think students are often surprised by how little we know, and my hope is that this project helped them to be more amazed that we know anything at all.”
As for a rematch, Gottlieb said it won’t be necessary. “There was no question as to which team had the best soccer players,” he said.
Soccer playing rats was the final project for students in Dan Gottlieb’s animal learning class. Although Gottlieb didn’t expect the experiment to result in anything “scientifically useful,” he hoped it would be a fun project and the students would discover something about how rats learn.
“They all have been training their rats to push a ball into one of two goals,” Gottlieb said prior to the end of the project. “Then we put all the rats together and see what happens. I’m skeptical that it will work, but who knows?”
The experiment culminated Dec. 5 with a two-on-two soccer match held in Guion’s psychology lab. On game day, the students and the four-legged footballers gathered in the lab around a tiny soccer stadium, complete with a midfield line and two goalie boxes painted on the chalkboard green “turf.”
The arena was constructed by Alison Carr ’08, as evidenced by a black fingernail she sported, thanks to a mishap with a power tool.
The eight rats were divided into two teams — the Green Team and the Red Team — and each wore a number painted on its back in its team’s color. Four would play at a time. “We decided if we put them all in at once it would be chaos,” Gottlieb said.
Most of the rodents were named for famous learning theorists — Robert Rescorla, Burrhus Skinner, Michael Domjan, Edward Thorndike, Ivan Pavlov, Randy Gallistel, John Gibbon and Hermann Ebbinghaus.
The exception to this was Butters, who was originally named for scientist Allan Wagner.
Butters would eventually become a prolific scorer with five goals to his credit during the game, although he didn’t always score for his own team. “No, Butters! No!” was shouted more than once as the rat rolled the pingpong ball into his own goal.
Initially, the class tried to “clicker train” the rodents, but “they were scared of the clicker,” Carr said. Instead, the students — Abby Adkins ’09, Kia Bryan ’09, Doe Buchli ’10, Alison Carr ’08, Alex DiFeliceantonio ’08, Anna Fure ’08, Molly McLemore ’10, Cheryl Seaver ’09 — used food and lots of repetition.
“During the first week, I wanted to get him accustomed to interacting with the ball, so each time he moved the ball with his nose or paw I rewarded him with food pellets,” Bryan said of her rat, Herman.
“The next phase was to train Herman to put the ball in the goal. At first, I trained Herman in a small training box that was placed in front of the goal. When he put the ball in the goal, I rewarded him with food pellets.
“Next, I removed the box and allowed him to roam around the field. Herman would get very distracted, so I would either put pellets under the ball to entice him, or simply tap on the ball to get his attention.”
In time, Bryan said, Herman became adept at scoring goals and even learned to distinguish between the two goals, one of which was painted black and the other, white.
Carr, who trained Robert, called the training technique “shaping.”
“We rewarded them for successive approximations of pushing the ball — first for being near it, then touching it, then pushing it, etcetera,” Carr said. “They had to learn on their own how to actually get it in the goal.
“Even if you couldn’t tell in the final game, they all had it pretty well down. Three out of four [on the Red Team] learned to move it by grabbing it with their paws and dragging it instead of pushing it with their noses.”
In the first half of the soccer match, however, rats on both teams scurried around the field and seemed more interested in each other than the ball. “C’mon … or not,” one student pleaded as her rat, Randy, showed and then lost interest in the ball.
Eventually, Green Team captain, the aforementioned Butters, scored a goal for the Red Team. Shortly thereafter, Butters scored again, this time for his own team. “Butters is a genius,” Gottlieb announced wryly.
At the end of the first half, the game was tied 1-1.
In the second half, Butters scored three more times, twice for the Green Team and once for the Red Team. Herman also scored three goals for the Green Team and the game ended with a 6-2 victory for Green.
“I learned it is a whole lot harder to train a rat to push a ball into a goal than I would have ever thought,” Carr said later. “Actually, the biggest surprise was how much their behavior changed every time a new rat was added to the situation.
“Our team scored tons of goals in a team training we had the other day, but the context seems to become novel again with the addition of the Green Team on game day … because they weren’t nearly as good as before.”
Overall, Gottlieb thought the experience was a positive one. “I thought the experiment went well in that the students took it seriously and put a lot of time into it,” he said. “The goal of the experiment was simply to have the students work on a project in which they would need to train their animals over the course of multiple days. Most labs for the course are one-day labs, which is difficult when we are interested in learning processes.
“By observing the behavior of their rats, day after day, I was hoping the students would learn something about the difficulties and constraints faced by researchers studying basic learning mechanisms. I think students are often surprised by how little we know, and my hope is that this project helped them to be more amazed that we know anything at all.”
As for a rematch, Gottlieb said it won’t be necessary. “There was no question as to which team had the best soccer players,” he said.
