“I griped about it at lunch one day to Bill Weist and Dr. Leslie Squier, our visiting psychologists from Reed College. I'd been trying to train one otter to stand on a box, I told them. No problem getting the behavior; as soon as I put the box in the enclosure, the otter rushed over and climbed on top of it. She quickly understood that getting on the box earned her a bite of fish, But. As soon as she got the picture, she began testing the parameters. 'Would you like me lying down on the box? What if I just put three feet on the box? Suppose I hang upside down from the edge of the box? Suppose I stand on it and look under it at the same time? How about if I put my front paws on it and bark?' For twenty minutes she offered me everything imaginable except just getting on the box and standing there. It was infuriating, and strangely exhausting. The otter would eat her fish and then run back to the box and present some new, fantastic variation and look at me expectantly (spitefully, even, I thought) while I struggled once more to decide if what she was doing fit my criteria or not.My psychologist friends flatly refused to believe me; no animal acts like that. If you reinforce a response, you strengthen the chance that the animal will repeat what it was doing when it was reinforced; you don't precipitate some kind of guessing game. So I showed them. We all went down to the otter tank, and I took the other otter and attempted to get it to swim through a small hoop. I put the hoop in the water. The otter swam through it, twice. I reinforced it. Fine. The psychologists nodded. Then the otter did the following, looking up for a reward each time: swam through the hoop and stopped, leaving its tail on the other side. Swam through and caught the hoop with a back foot in passing, and carried it away. Lay in the hoop. Bit the hoop Backed through the hoop. 'See?' I said. 'Otters are natural experimenters.”