In two preceding papers (Gaito, 1985a(Gaito, , 1985b, I presented statistical data for two important variables in the suppression of kindling behavior (duration of stimulation and intertrial intervals), In this article, the relative contribution of each of these (plus other reliable variance components) is contrasted with the associated error components . In both cases, reliable variance contributed a substantial proportion of the overall variance, whereas lesser amounts of variance were associated with error components when suppression was complete, The overall results suggest that the suppression process is a transient time-dependent one whose magnitude decreases with short durations of stimulation and with long intertrial intervals., ad aĨ n a preceding paper (Gaito, 1985a), the effect of various durations of stimulation with I-Hz or 3-Hz sine waves on 6O-Hz kindling behavior was indicated. A second paper provided the results of varying intervals between I-Hz and 60-Hz stimulation (Gaito, 1985b). In this paper, the magnitude of these important variables is evaluated via statistical analyses.
RESULTS
Duration VariableThe durations of stimulation used were 0, 5, 15, 30, 60, 120, 180,300, and 600 sec (Gaito, 1985a). The results for the I-Hz and 3-Hz stimulation effects were similar. Thus, statistical data are presented only for the I-Hz events.To determine the magnitude of the duration effect, I used a type of intraclass correlation (Haggard, 1958), the Gaito Utility Coefficient (Uc) (Gaito, 1958(Gaito, , 1973. The obtaining of these coefficients can be illustrated by using the first effective threshold intensity ETI l point only in a simple one-factor ANOVA design with the data as shown in Table I .The Uc values for separate ETI points are shown in Table 2 and represent proportion of variances. The F values are significant at ETI2 , ETI J , and ETI 4 • As the F values increase from 1.34 at ETI l to 25.61 at ETI 4 , the Uc values for the durations dimension increase from .046 to .779, and the error decreases from .954 to .221. From this table, it appears that the sharpest effect of the I-Hz stimulation duration variable is between ETI l and ETI 2 • The magnitude reaches a stable point at ETI J • This stability