Linguistic Security, Ideology, and Vowel Perception 255versus rural, but not age. However, Koops et al. (2006) suggest that Houstonians
are in fact implicitly aware that this feature is associated with age. They used
eye-tracking technology to determine to what extent subjects ¿ xate on potentially
homophonous competitors presented on a computer monitor (as shown in Fig-
ure 11.2); in the middle of the screen, is a photograph of one of three women: a
“young” woman, a “middle-aged” woman, and an “older” woman.
Figure 11.1 Accuracy of subjects’ judgments of shadowed lexical items, according
to number of repetitions of shadows, and frequency of lexical items.
(HF = “high frequency,” etc.; Goldinger 1998.)