Table 1.

Behavioral data.

	Frequency difference of stimuli (f2–f1) in Hz
	−4	−2	2	4	Difficulty effect	Sign effect
PCR (%)	75.9 ± 4.4	65.3 ± 3.5	70.5 ± 4.3	86.1 ± 3.7	n/a (p < 0.001)∗	n/a (p = 0.002)∗
RT correct (ms)	590.2 ± 44.8	608.0 ± 48.3	554.5 ± 47.1	541.6 ± 44.7	−15.4 ± 9.0 (p = 0.002)∗	−51.1 ± 28.6 (p = 0.001)∗
RT incorrect (ms)	615.9 ± 64.9	593.9 ± 60.7	651.5 ± 58.1	678.6 ± 68.2	24.5 ± 19.0 (p = 0.014)∗	60.2 ± 34.8 (p = 0.002)∗

Proportion of correct responses (PCRs) and response times (RTs) as a function of the physical frequency difference f2–f1. Mean values ± 95% confidence interval (CI) are shown. ‘Difficulty effect’ compares easy (±4 Hz) and difficult (±2 Hz) trials in a paired t-test. ‘Sign effect’ compares between trials with positive (2 and 4 Hz) and negative (−2 and −4 Hz) frequency differences in a paired t-test. PCRs and RTs showed significant effects of difficulty and sign. RTs showed both effects for correct and incorrect trials, however, in opposing directions (cf. interactions in ANOVA of RTs). PCRs were logit-transformed before testing, due to non-normally distributed residuals. We omitted average differences of logit-transformed PCRs for both effects to avoid confusion (indicated by n/a). Asterisks indicate statistically significant results.