TABLE V.

Pairwise effects of high vs low variability for each source (between-talkers, within-talkers, or phonetic contrasts). For condition labels (A–H), refer to Table I and Fig. 1. *Significant after Holm–Bonferroni correction, α = 0.05.

Variability effects	Conditions	Δ RT (ms)	Interference (%)	β	s.e.	t	df	p
Between-talker variability (multiple vs single talker)
One exemplar and one contrast	E > A	48.3	6.21	0.043	0.012	3.816	24.9	< 0.001*
Multiple exemplars and one contrast	G > C	16.0	1.95	0.010	0.006	1.519	23.8	0.142
One exemplar and multiple contrasts	F > B	6.8	0.64	0.003	0.004	0.760	22.6	0.455
Multiple exemplars and multiple contrasts	H > D	15.1	1.38	0.008	0.006	1.399	23.1	0.175
Within-talker variability (multiple vs one exemplar)
Single talker and one contrast	C > A	39.7	5.10	0.027	0.009	3.128	24.4	< 0.005*
Multiple talkers and one contrast	G > E	7.3	0.89	−0.008	0.005	−1.464	28.7	0.154
Single talker and multiple contrasts	D > B	26.2s	2.46	0.014	0.005	2.985	28.7	< 0.006*
Multiple talkers and multiple contrasts	H > F	34.5	3.22	0.019	0.005	4.197	29.2	< 0.001*
Word-choice variability (multiple vs one phonemic contrast)
Single talker and one exemplar	B > A	287.7	36.98	0.140	0.011	12.31	23.0	≪ 0.001*
Multiple talkers and one exemplar	F > E	246.2	29.79	0.099	0.007	15.09	29.5	≪ 0.001*
Single talker and multiple exemplars	D > C	274.2	33.54	0.130	0.007	18.29	23.0	≪ 0.001*
Multiple talkers and multiple exemplars	H > G	273.3	32.79	0.130	0.006	21.57	24.1	≪ 0.001*