BEAST Software






 






The sequencing of behavior can have particular power to analyze changes in behavior. Three routines that include graphical functions are intended for analysis of sequence between two acts. Log survivorship analysis is similar to that for 1act sequences except that the assumptions will not allow standard statistical testing. Instead BEAST uses a randomization technique to establish confidence limits for the transition rates.
The log survivor curve (blue) tracks almost along the top line of the 95% confidence interval from the randomization test. Note that a randomization test may not have smooth lines as you would obtain from a parametric statistic.
The statistical test indicates that this was indeed true. The confidence limits for transition rate are 0.291  0.575 whereas the calculated rate in 0.267. In this case, the result is trivial since there were few intervals less than 4 sec since the players had a pause between each iteration of the RockPaperScissors game.
Two complementary procedures use very different methods to test similar hypotheses based on the frequency with which one act precedes and follows another and the latency between one response and another.
