accident, and so on (Of course, each student wanted his or her cat to have the
longest tail.) Before they began spinning, students predicted that if they built 30 cat
tails in this way, they would get about equal numbers of cats with short, medium,
and long tails. After several trials they noticed they were tending to get medium
tails, which they explained by pointing out that you would have to be “real lucky” to
get a big number every spin, or “real unlucky” to get a small number every spin. As
this was our last session with these students, we could not explore what they might
have generalized from this experience; but we believe that understanding why such
processes produce normal-shaped distributions is a critical part of coming to trust
how process signals rise up through the noise.