Table 1. Subjective duration (i.e. difference from the 2 s standard stimulus) estimated by the Quest algorithm for each stimulation frequency (mean of all 30 participants).

frequency (Hz)	mean duration estimates (s)	adjusted p-value	Bayes factor
3.91	0.66	<0.001	0.002
7.71	0.59	<0.001	0.002
15.71	0.50	<0.001	0.01
25.00	0.51	<0.001	0.001
31.14	0.35	= 0.01	0.12
41.71	0.19	= 0.04	0.38
55.43	−0.02	= 0.07	0.74
62.29	−0.18	= 0.88	7.06
71.14	0.01	= 0.35	3.47
82.86	−0.17	= 0.88	6.99
165.7	−0.22	used as reference

One-tailed paired-samples t-tests (corrected for multiple comparisons with the Bonferroni-Holm method) indicate that for frequencies up to 41.7 Hz the estimated duration significantly differed from the duration estimated for the reference frequency; stimuli flickering at these frequencies were perceived as longer. The last column shows Bayes factors, corresponding to the ratio between the probabilities of the null and alternative hypotheses. Bayes factors <0.3 can be taken as evidence for an effect of flicker frequency on perceived duration, while Bayes factors ≥3 indicate that there is likely no such effect. Intermediate values allow no conclusive decision for either of the two hypotheses.