Table 2.

Results of simulation of the repeats and no-repeats sequences in Walthew and Gilchrist (2006). Note that for nearly one-quarter of the no-repeat condition, the next trial is actually more likely to be sparse.

Number of preceding three trials on sparse side	Proportion of trials		Probability that next trial will be on the sparse side
	no-repeat condition	repeat condition	no-repeat condition	repeat condition
0	23	30	0.55	0.33
1	55	44	0.31	0.33
2	21	22	0.17	0.33
3	1	4	0.07	0.33