IOl
it on the same side as the sound as compared to the opposite side (4.8% vs. 2.8%; t19 =
2.447, p = .024), regardless of whether or not a somatosensory stimulus was delivered
to the opposite hand. However, there was only a marginally significant difference
between false reports of Somatosensation on the same side as the sound and the no
sound conditions (4.8% vs. 3.5%; t19 = 2.054, p = .054). The difference in false alarm
rates between the no sound condition and the false reports of Somatosensation on the
opposite side of the sound also was not significant (3.5% vs. 2.8%; t19 = 1.209, p = .241).
Although the design of this experiment was not perfectly suited to conduct signal
detection analyses, and not all response biases could be ruled out with this design, we
nonetheless conducted signal detection analyses to obtain an estimate of bias free
changes in sensitivity to somatosensory perception with sound. The d' values were
calculated from the hit (correct localization of the somatosensory stimulus) and false
alarm (see above and Table 1) rates for each subject and subjected to the same
statistical analyses as the percent correct data. Consistent with the analyses on the
discrimination and false alarm rates, an ANOVA revealed a significant difference in
sensitivity between the three auditory stimulus conditions (F1,19 = 7.05, p = .002). There
was significantly higher sensitivity for discriminating the side of the somatosensory
stimulus when the sound was on the same side (d' = 2.00) as compared to the opposite
side (d' = 1.30; t19 = 3.689, p = .002) and the no sound conditions (d' =1.14; t19 = 3.352, p
= .003). There was no decrease in sensitivity when the sound was delivered to the hand
opposite the somatosensory stimulus as compared to the no sound conditions (t19 =
.561, p = .581). These differences in d' values indicate that perceptual sensitivity,