obtain a measure of standardized reactivity for each subject, we calculated the standard deviation across all valences, within subject for each session, and divided each reactivity score within session by this standard deviation. This approach produces an effect size statistic, called a D score (Greenwald, Nosek, & Banaji, 2003), tailored to repeated measures with a scaling similar to a Cohen’s d (Cohen, 1988). The resulting standardized difference score takes into account individual variability in physiological response while scaling individual reactivity differences over the four slide valences at each session.
