If a study uses reaction time as a dependent variable, stable individual differences (such as variations in participants' nervous systems and muscles) will naturally cause some individuals to be faster or slower. In a between-subjects ANOVA, these differences add to the within-groups variance ($MS_W$), decreasing the $F$-ratio. In a repeated-measures ANOVA, these differences are measured and subtracted from $MS_W$, increasing the test's sensitivity.