102
 The tone by block interaction was also examined using

 paired -t-tests with a Bonferroni correction based on four

 comparisons, requiring a P < .0125 for significance. The

 SCR for the shock no-shock (mean=21.80, sd=19.89) was

 significantly higher than the low tone (mean=10.46,

 sd=12.24) for blocks 1 [T(1,46) = -4.655, P < .0001], and

 block 4 (shock condition: mean=12.27, sd=19.02; no-shock

 condition: mean=4.17, sd=8.48), [T(1,46) = -3.433, P <

 .01], but not block 2 (shock condition: mean=12.21,

 sd=16.89; no-shock condition: mean=7.46, sd=9.86), [T(1,46)

 = -2.264, P = .0284] and block 3 (shock condition:

 mean=9.63, sd=14.39; no-shock condition: mean=6.54,

 sd=11.59), [T(1,46) = -1.449, P = .1542]. The t-tests are

 presented in Table C-28 in Appendix C.

 The condition by group interaction was explored using

 t-tests with Bonferroni corrections separately for the shock

 and no-shock conditions. Since there were not significant

 differences between the LH NCS and the RH NCS during the no-

 no-shock condition [T(22) = -.257, P = .7995] or the shock

 condition [T(22) = -1.338, p = .1946], the two groups were

 combined into one control group. Using the bonferroni

 correction, the analyses had to reach a p-value of .05/3=

 .017 to be considered significant. T-test tables are

 provided in Appendix C, Tables C-29 and C-30. During the

 shock anticipation, the RHD patients (mean=5.15, sd=8.55)

had significantly smaller responses then the CONS