Skip to main content
. Author manuscript; available in PMC: 2019 Jul 16.
Published in final edited form as: J Cogn Neurosci. 2017 Jan 27;29(5):919–936. doi: 10.1162/jocn_a_01096

Figure 1.

Figure 1

(A) Spectrogram showing original stimuli and changes made during audio-morphing for a word–word minimal pair “blade” (pink) and “glade” (blue). Spectrograms show original tokens and low-ambiguity (5%/95%) and high-ambiguity (35%/65%) stimuli created using STRAIGHT software and time-aligned spectral averaging (see Rogers & Davis, 2009, for details). (B) Table showing example stimulus pairs and numbers of items illustrating changes made at syllable onset and offset in items with different types of change across lexical conditions (point of change highlighted in bold). (C) Proportion of responses matching Item 1 averaged over the phonetic continua in each of the phonological environment conditions; inset bar graph shows the mean position of the category boundary in each condition.