Figure 3.

Average fundamental frequency (F0) values across blocks (5 trials per block) of unaltered (baseline) and frequency-altered feedback (FAF) trials when participants were instructed to ‘compensate’ or ‘ignore’ changes in auditory feedback. The median F0 value was calculated for the initial 50 ms of vocal productions F0, normalized to the target note from each block. The vertical dotted line indicates the commencement of FAF trials. (A) Nonsingers’ median 50 ms data across blocks of unaltered and FAF trials, when auditory feedback was gradually (-2 cent increments across trials to -100 cents) shifted in pitch. Black squares represent nonsingers’ F0 values when instructed to ‘compensate’ for FAF. Gray diamonds represent nonsingers’ F0 values when instructed to ‘ignore’ the FAF. Error bars depict standard error of the mean. (B) Singers’ median 50 ms data across blocks of unaltered and FAF trials, when auditory feedback was gradually (-2 cent increments across trials to -100 cents) shifted in pitch. Black squares represent singers’ F0 values when instructed to ‘compensate’ for FAF. Gray diamonds represent singers’ F0 values when instructed to ‘ignore’ the FAF. Error bars depict standard error of the mean. (C) Nonsingers’ median 50 ms data across blocks of unaltered and FAF trials, when auditory feedback was constantly shifted down by -100 cents. Black squares represent nonsingers’ F0 values when instructed to ‘compensate’ for FAF. Gray diamonds represent nonsingers’ F0 values when instructed to ‘ignore’ the FAF. Error bars depict standard error of the mean. (D) Singers’ median 50 ms data across blocks of unaltered and FAF trials, when auditory feedback was constantly shifted down by -100 cents. Black squares represent singers’ F0 values when instructed to ‘compensate’ for FAF. Gray diamonds represent singers’ F0 values when instructed to ‘ignore’ the FAF. Error bars depict standard error of the mean.

Keough et al. BMC Neuroscience 2013 14:25   doi:10.1186/1471-2202-14-25
Download authors' original image