Explain this reasoning [for the selection processor hypothesis in your own words. Can you think of other ways to test the hypothesis?

READINGS FROM THE BOOK:

Leslie's Answer: The Selection Processor Hypothesis According to Leslie and his collaborators, there is a long time lag between when the capacity for metarepresentation first emerges (during the second year) and when children generally pass the false belief test (toward the end of the fourth year), because there are actually two very different abilities here. The first is the ability to attribute true beliefs to someone else. The second is the ability to attribute false beliefs. These two abilities emerge at very different times in development. On Leslie's model, young children are able to attribute true beliefs from a relatively early age. But they are only able to attribute true beliefs. They can only succeed on the false belief task when they learn to "switch off," or inhibit, the default setting that other people's beliefs are true. Leslie presents this in terms of a mechanism that he calls the selection processor. The selection processor is set up to favor true beliefs. So, the selection processor's default setting favors the true belief candidate. In the false belief task, this is the belief that Sally believes that the marble is in the box. But in this case, there is evidence to the contrary. The child knows 358 Mindreading: Advanced Topics that Sally did not see the marble being moved from the basket to the box. But for this countervailing evidence to be effective, the selection processor's default setting needs to be overridden. This is what separates children who pass the false belief task from children who fail. The ones who pass are able to inhibit the bias in favor of the true belief candidate. So, for Leslie, the problem is not with TOMM itself. TOMM is in place from the pretend play stage. It is just that it initially only works to attribute true beliefs. Success on the false belief task comes only when the young child acquires a more general capacity for executive control. Is there any way of testing this general hypothesis? One way to test this hypothesis would be to alter the false belief task to make greater demands on mechanisms of inhibition and control. If the task makes greater demands on inhibitory control, and inhibitory control is the factor that explains success rather than failure on the false belief task, then one would expect that success rates on the altered task would be lower than on the original task. Exercise 14.1 Explain and assess this reasoning in your own words. Can you think of other ways to test the hypothesis? A study published by Leslie and Pamela Polizzi in 1998 reported a number of experiments adopting this general strategy. Here is a representative example. Children are presented with a scenario in which a girl (let's call her Sally, for continuity) is asked to place food in one of two boxes. The twist to the tale is that one of the boxes contains a sick kitten. Because eating the food might make the kitten worse, Sally wants to avoid putting the food into the box with the kitten in it. So, Sally has what Leslie, German, and Polizzi term an avoidance-desire. The point is that avoidance-desires are inhibitory. An avoidancedesire is a desire not to do something. There were two conditions: In the true belief condition, the kitten is moved from Box A to Box B in front of Sally. In the false belief condition, the kitten is moved without Sally seeing. Children in each condition are asked to predict which box Sally will put the food in. There is no question here about whether the children understand false belief. All the children were able to pass the standard false belief task and all of them answered correctly when they were asked where Sally thought the kitten was. In the true belief condition, the child knows that the kitten is in Box B (since she saw the kitten being moved there) and she knows that Sally wants to avoid putting the kitten and the food in the same box. So, she needs to predict that Sally will put the food in Box A. But for that she needs to be able to make sense of Sally's inhibition of what most people would normally want to do - which is to give food to a kitten. It turned out that a very high percentage (well over 90 percent) of the children in the experiment were able successfully to predict where Sally would put the food in the true belief condition. Now consider the false belief condition. The child still knows that the kitten is in Box B and she still knows that Sally wants to make sure that the kitten does not get the food. But now she also needs to take on board the fact that Sally did not see the kitten being Why Does It Take Children So Long to Learn to Understand False Belief? 359 moved from Box A to Box B. So, as on the standard false belief task, she needs to inhibit her own knowledge of where the kitten is. Now the children are being asked to do two things at once - to inhibit their own knowledge of where the kitten is, as well as to make sense of Sally's inhibition of the normal desire to give food to a kitten. There is a double inhibition required. According to Leslie, German, and Polizzi this is why the success rate in the false belief condition is so much lower than in the true belief condition. It turned out that only 14 percent of children in the study succeeded in the false belief condition (as opposed to 94 percent in the true belief condition). Their hypothesis is that the double inhibition places much higher demands on the selection processor than the ordinary false belief tasks.