Skip to main content
. Author manuscript; available in PMC: 2010 Feb 26.
Published in final edited form as: Atten Percept Psychophys. 2010 Feb;72(2):470–480. doi: 10.3758/APP.72.2.470

Figure 4.

Figure 4

Results from 18 listeners in Experiment 1. (A) Proportion “saxophone” responses as a function of target stimulus. Steps along the abscissa correspond to the series of target sounds, with 1 corresponding to the French horn endpoint and 6 to the tenor saxophone endpoint. Separate lines denote performance in the French horn − saxophone filtered (abbreviated “Horn − Sax”), unfiltered (“Control”), and saxophone − French horn filtered (“Sax − Horn”) context conditions. Standard error bars are depicted for each of the 18 trial types. (B) Mean 50% crossovers for each context condition derived from fitted logistic regression functions. The ordinate corresponds to the location in the series of six target stimuli (1 = French horn endpoint; 6 = tenor saxophone endpoint [neither shown]), where listeners are projected to respond “French horn” and “saxophone” with equal frequency. Error bars denote standard errors of the means. *p < .05.