Table 2.

Estimates of model coefficients (log odds), standard errors (SE), lower (CI_lower) and upper (CI_upper) 0.95 confidence intervals, z statistics (z-value) and associated p-values of the model fitted for study 1.

Term	log odds	SE	CIlower	CIupper	z	p
Intercept	−2.678	0.275	−3.229	−2.166	−9.735
Stimulus state diff	2.860	0.254	2.391	3.378	11.281	0.000
Intermediate	0.197	0.209	−0.240	0.643	0.941	0.347
Song	0.264	0.210	−0.175	0.692	1.257	0.209
z.Musicality	0.295	0.247	−0.217	0.768	1.191	0.234
Stimulus state diff: Intermediate	0.593	0.239	0.127	1.079	2.477	0.013
Stimulus State diff: Song	0.397	0.236	−0.084	0.882	1.680	0.093
Stimulus state diff:z.Musicality	0.016	0.234	−0.442	0.488	0.070	0.945
Intermediate:z.Musicality	−0.309	0.173	−0.683	0.084	−1.787	0.074
Song:z.Musicality	−0.086	0.174	−0.481	0.324	−0.490	0.624
Stimulus state diff: Intermediate:z.Musicality	0.616	0.225	0.142	1.119	2.742	0.006
Stimulus state diff:Song:z.Musicality	−0.010	0.223	−0.468	0.476	−0.044	0.965

The model estimated detection of a difference between original sound and target sound (reference level: Stimulus State same) as a function of Stimulus Variant (Speech, Song, Intermediate. Reference level: Speech) and Musicality as assessed by the GoldMSI questionnaire and z-transformed (that way the estimates refer to an individual with average musicality).