Robert F. Port
Catherine L. Rogers
Dept. of Linguist., Indiana Univ., Bloomington, IN 47405
Charles S. Watson
Gary R. Kidd
Indiana Univ., Bloomington, IN 47405
It has been assumed that subjects trained to detect increments in the frequency of all components of complex tonal patterns (broad focus) would be less accurate in detecting changes in a single target tone than subjects who have been trained to detect changes in only that component [e.g., Watson et al., J. Acoust. Soc. Am. 60, 1176--1186 (1976)]. In several experiments, using a number of 750-ms ten-tone patterns, subjects were trained using one of three methods: in the first two, a S/2AFC procedure was used to train subjects to detect frequency increments in a specific target tone (group one) or to detect frequency increments that could occur in any of the ten components (group two), and in the third, subjects were trained only to identify the individual patterns. Subjects trained using these methods were tested on their ability to detect changes in various components of the patterns, including the target tone for the first group. In all of these experiments, only very slight differences in performance were found among the different groups. These results suggest that lengthy experience with a given pattern allows a listener to discriminate small differences in frequency in any of the individual components of that pattern, relatively independent of the nature of that experience.