Examing the evidence for a pitch centre in human auditory cortex: a multi method approach

This PhD used a combination of psychophysical, functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) methods to evaluate the evidence for a pitch centre within auditory cortex according to the four pitch criteria: 1) pitch selectivity, 2) pitch constancy, 3) covariation with pitch salience, and 4) accounting for confounding factors, that were described in an article by Hall and Plack (2009). An fMRI study re-examined pitch criteria 1, 3 and 4 using stimuli and a subtractive study design informed by Penagos et al. (2004), but extended this work by addressing some of their limitations. Results indicated that the representation of pitch compared to noise is widely distributed across auditory cortex, while the evidence for an effect of pitch salience was questionable given that the weak pitch salience condition was not significantly different from matched noise at a group level. These findings raise concerns regarding fMRI‟s sensitivity to pitch salience effects in the context of high individual variability. An ERP „adaptation‟ study evaluated pitch criteria 1, 2 and 4 using pitch and timbre stimulus parameters that had been previously matched for discriminability, and sequences either varied in pitch, timbre or both across listeners. Findings from both sensor and source-based analyses suggested that pitch responses may be influenced by timbre (i.e., non-invariant); although further research is required. Thus, evidence failed to support the notion of pitch constancy at the level of the auditory cortex. Further studies using psychophysical listening paradigms continued this work, and again seemed to confirm a lack of evidence for pitch constancy. Reaction times and accuracy data demonstrated that timbre changes (non-target stimulus) interfered with listener‟s ability to discriminate pitch (target stimulus), and vice versa. Overall, these convergent results suggest that there is no modular representation of pitch (pitch centre), but rather pitch processing sites are distributed throughout multiple areas of primary and non-primary auditory cortex and are seemingly non-invariant to other stimulus parameters related to its perception (e.g., timbre). Under this assumption, the spatio-temporal model of pitch perception may best describe the neural mechanism underpinning pitch perception. Several recommendations are made to address challenges to interpretation identified throughout this PhD, which are likely to guide further research in this area.