Second, suppose that you are playing a visual problem-solving game, such as Rush Hour, or completing a jigsaw puzzle. (Rush Hour involves a square board loaded with cars and trucks in various positions," and the object is to move these in a sequence so as to allow a designated car to leave the board through the only exit.) Here the problem solving in part consists of internal mental operations (let us suppose), but also in part it consists of the active physical manipulation of pieces on a board or pieces within the puzzle. One might suppose that these manipulations-rotations of puzzle pieces, or trials of car move sequences-were simply presented to you in a computer simulation, or through an automation of trial moves from which you select the best outcome. I suspect that this would change the problem-solving task in a significant way; but note that even here problem solving requires active engagement with a part of the world beyond the head, namely, the various visual displays from which one must select. In both cases, the problems are solved by utilizing, exploiting, or manipulating a set of resources, some of which are outside of the head. These are not simply inputs to those that are inside the head, because the very process of problem solving involves them as much as it involves resources inside the head. There is nothing bounded by the skull that counts as solving these kind of problems. (Wilson, 2010, pp. 180-181).
This might be clearer by an analogy. Suppose you have to solve the problem of printing the numerals for the first one hundred prime numbers on a sheet of paper. Also suppose that you have a computer with a program that computes these numerals and sends them to the printer. All you have to do is hit the "return" key. Now, I'm ok with saying that the problem solving involves the stuff that the computer and printer do, but why go further with Wilson and say that the cognitive contribution also includes the stuff the computer and printer do? That is, why accept EC? What's wrong with the view that the cognitive processing dedicated to solving the problem ends with your hitting the return key? That is, what is wrong with the anti-EC interpretation?