A third covariance matrix is relevant for the AB interaction because this effect does not average over either A or B. Instead, the interaction assesses whether the B difference is the same at each level of A. Table 12.11 presents scores that address this question. For each subject, a given score represents the noise effect (i.e., reaction time when noise is present minus reaction time when noise is absent) at a particular level of the angle factor.
#' @section Variables:
angle0
reaction time when noise is present minus reaction time when noise is absent at angle 0
angle4
reaction time when noise is present minus reaction time when noise is absent at angle 4
angle8
reaction time when noise is present minus reaction time when noise is absent at angle 8
References
Maxwell, S. E., Delaney, H. D., \& Kelley, K. (2018). Designing experiments and analyzing data:
A model comparison perspective (3rd ed.). New York, NY: Routledge.