2.9.2 Reaction Time and Task Performance
Task performance was quantified using Hit rate. ‘Hit’ trials were
defined as trials in which a participant responded to the cue between
150 and 1000 ms after cue onset (Table 2A, Figure 2F ). A
secondary measure of task performance known as ‘Percent Double Press’
was computed to characterize systematic performance atypicalities in ASD
by dividing the number of duplicate responses (two or more responses in
an epoch) by the total number of button presses (Table 2B,
Figure 2G ). For perfect task performance, each stimulus would elicit a
single button press. In this case, the percentage of double presses
would be 0. For a participant who systematically pressed the response
button two times for each stimulus, the percentage of double presses
would be 50% because half of all presses represent duplicate responses.
To examine differences in reaction time as a function of the preceding
sensory cue type (A, V, AV), participant age, and clinical diagnosis
(ASD, TD), a general linear mixed model (GLMM) was computed
(Table 2B ). Table 2 details all model factors. Main effect
terms were included for participant group and preceding sensory cue. Age
was included as a covariate. Because investigation of group effects was
a key focus of the current work, two-way interaction terms were included
for all combinations of factors involving ‘Group’ (Group*Age, Group*Cue,
Group*RT quartile). Note that ‘RT quartile’ (fastest 25% of trials,
middle 25-50% of trials, middle 50-75% of trials, slowest 25% of
trials) was not included as a main effect in this GLMM since these RT
quartiles were constructed on the basis of the outcome measure (RT).