Forget dogs: These rats could be the future of search and rescue - Science Club

your daily dose of science and nature

Friday, February 11, 2022

Forget dogs: These rats could be the future of search and rescue

A giant pouched rat pulls a small ball on its vest, alerting researchers that it has located a mock victim.APOPO

Think search and rescue animal, and you’re likely to picture a dog in an orange vest. But a Belgian nonprofit based in Tanzania wants you to imagine something else: the African giant pouched rat. Donna Kean and her colleagues at APOPO, a nonprofit that trains pouched rats to save lives, have spent the past 2 decades working with the curious animals (Cricetomys ansorgei) to sniff out tuberculosis and track down land mines. Now, they’re moving on to search and rescue.

Science caught up with Kean to chat about the new project, known as RescueRats. This interview has been edited for clarity and length.

Q: Why send rats into the rubble instead of dogs?

A: The rats have a comparable sense of smell, and they’re just as trainable as dogs. They also aren’t tied to just one trainer, which is what you find with dogs. And their size is useful because they will be able to penetrate into areas with dense rubble and debris that dogs just wouldn’t be able to. The life span of our rats in captivity is roughly 8 years but we have rats that have lived up to 10 to 11 years in retirement.

Donna Kean holding a giant pouched rat
Donna Kean has spent the past 5 months training rats for search and rescue at APOPO’s research site in Tanzania.APOPO

The giant pouched rats are endemic here in Tanzania, where we’ve had a land mine detection program with them for 20 years. We have all the training facilities in place, and we have amazing trainers who are local. APOPO’s mission is focused on humanitarian projects, so this is a great way to build local capacity.

Q: How do you train a rat?

A: The first thing we did was train them to return to their starting point. A trainer released the rat in an empty room and let it wander around. We trained the rat to come back to the starting point when we played a beep; they were fed a reward of powdered rat pellets mixed with avocado and banana through a syringe when they did so.

Then we trained the rats to pull a rubber ball on their backpack. The ball is attached to a microswitch, which emits a beep. In the real world, it would provide a signal to rescuers. When they were reliably doing that, we’d start introducing a target human. The goal is for the rat to go over to this mock victim, pull the ball, return to where they were released from when they hear the beep.

Q: What was your success rate?

A: Training started in August and is ongoing. On average, the rats took 14 sessions to reliably return to base point within 3 seconds of the beep in stage one, and 10 sessions to pull the ball for 3 seconds in stage two. Our most proficient rats were already reliably performing the basic behavioral sequence (go to victim, pull ball, and return to base point) within seven sessions in the third stage. So far, all nine rats have mastered stages one and two; six have reliably learned the entire sequence.

Q: You were training very young, newly weaned rats. How were they to work with?

A: There are a lot of individual differences in personality, aptitude, and ability. A couple of them don’t perform as well as the others. So for example, we’ll have some that’ll go straight to the target. They look like they’re doing the correct behavior, but they don’t pull the ball; they just come back to where they’re released from.

But you know, from all of my experience working with animals, they can always surprise you. So the ones that are not doing very well can all of a sudden be the ones that shape up and are top of the class.

Q: How far could a rat theoretically go from its handler?

A: For search and rescue, dogs don’t penetrate debris, they just sniff around the outside of it. So the rats will only be deployed after human and canine search teams have already worked the debris site. We plan to release rats into debris sites from multiple entry points. We think they could probably go on average between 10 and 20 meters, with a maximum of about 30 meters, so we can cover a large area.

Q: Might a rat running around a disaster site scare someone?

A: There may be issues if people have a fear of rats or animals. But wherever they’re introduced and used for this purpose, there can be marketing campaigns so people know that this can happen. They’ll also have a backpack on, and they’ll have a microphone, light, and camera. There can be audio coming from the backpack that potentially says, “I am a RescueRat, I’m here to help you,” that kind of thing.

A giant pouched rat pulls a small ball on its vest, alerting researchers that it has located a mock victim.APOPO

Think search and rescue animal, and you’re likely to picture a dog in an orange vest. But a Belgian nonprofit based in Tanzania wants you to imagine something else: the African giant pouched rat. Donna Kean and her colleagues at APOPO, a nonprofit that trains pouched rats to save lives, have spent the past 2 decades working with the curious animals (Cricetomys ansorgei) to sniff out tuberculosis and track down land mines. Now, they’re moving on to search and rescue.

Science caught up with Kean to chat about the new project, known as RescueRats. This interview has been edited for clarity and length.

Q: Why send rats into the rubble instead of dogs?

A: The rats have a comparable sense of smell, and they’re just as trainable as dogs. They also aren’t tied to just one trainer, which is what you find with dogs. And their size is useful because they will be able to penetrate into areas with dense rubble and debris that dogs just wouldn’t be able to. The life span of our rats in captivity is roughly 8 years but we have rats that have lived up to 10 to 11 years in retirement.

Donna Kean holding a giant pouched rat
Donna Kean has spent the past 5 months training rats for search and rescue at APOPO’s research site in Tanzania.APOPO

The giant pouched rats are endemic here in Tanzania, where we’ve had a land mine detection program with them for 20 years. We have all the training facilities in place, and we have amazing trainers who are local. APOPO’s mission is focused on humanitarian projects, so this is a great way to build local capacity.

Q: How do you train a rat?

A: The first thing we did was train them to return to their starting point. A trainer released the rat in an empty room and let it wander around. We trained the rat to come back to the starting point when we played a beep; they were fed a reward of powdered rat pellets mixed with avocado and banana through a syringe when they did so.

Then we trained the rats to pull a rubber ball on their backpack. The ball is attached to a microswitch, which emits a beep. In the real world, it would provide a signal to rescuers. When they were reliably doing that, we’d start introducing a target human. The goal is for the rat to go over to this mock victim, pull the ball, return to where they were released from when they hear the beep.

Q: What was your success rate?

A: Training started in August and is ongoing. On average, the rats took 14 sessions to reliably return to base point within 3 seconds of the beep in stage one, and 10 sessions to pull the ball for 3 seconds in stage two. Our most proficient rats were already reliably performing the basic behavioral sequence (go to victim, pull ball, and return to base point) within seven sessions in the third stage. So far, all nine rats have mastered stages one and two; six have reliably learned the entire sequence.

Q: You were training very young, newly weaned rats. How were they to work with?

A: There are a lot of individual differences in personality, aptitude, and ability. A couple of them don’t perform as well as the others. So for example, we’ll have some that’ll go straight to the target. They look like they’re doing the correct behavior, but they don’t pull the ball; they just come back to where they’re released from.

But you know, from all of my experience working with animals, they can always surprise you. So the ones that are not doing very well can all of a sudden be the ones that shape up and are top of the class.

Q: How far could a rat theoretically go from its handler?

A: For search and rescue, dogs don’t penetrate debris, they just sniff around the outside of it. So the rats will only be deployed after human and canine search teams have already worked the debris site. We plan to release rats into debris sites from multiple entry points. We think they could probably go on average between 10 and 20 meters, with a maximum of about 30 meters, so we can cover a large area.

Q: Might a rat running around a disaster site scare someone?

A: There may be issues if people have a fear of rats or animals. But wherever they’re introduced and used for this purpose, there can be marketing campaigns so people know that this can happen. They’ll also have a backpack on, and they’ll have a microphone, light, and camera. There can be audio coming from the backpack that potentially says, “I am a RescueRat, I’m here to help you,” that kind of thing.

No comments:

Post a Comment