Think like a machine!

We have a machine that can look at a picture and tell us what it is. Most of the time, it tells us the right answers. But sometimes it makes mistakes. We don't know why the machine makes the mistakes it makes; do you?

Your job in this experiment is to predict when our machine will make a mistake. On each trial, you'll see a few images that will look to you like a familiar object. But, for one of them, the machine gave a different answer than it was supposed to. For each set of images, we want you to predict which image the machine got wrong.

When you are ready to begin, click "Start Experiment". By completing this survey, you are consenting to be in this research study. Your participation is voluntary and you can stop at any time.

(NOTE: When the experiment is done, a submit button should appear. If that fails to occur, the study should "auto-submit" after one minute.)

Which one is the worst example?

We have some images of familiar objects. But some of these images are "bad examples" — they don't really look like what you'd expect when you think of that object. Can you tell us which ones are the worst examples?

On each trial, you'll see a few image examples of a familiar object. For each set of images, we want you to select the worst example of that object.

When you are ready to begin, click "Start Experiment". By completing this survey, you are consenting to be in this research study. Your participation is voluntary and you can stop at any time.

(NOTE: When the experiment is done, a submit button should appear. If that fails to occur, the study should "auto-submit" after one minute.)