My Ph.D. work is focused on investigating certain aspects of pitch processing and absolute pitch in musicians. Below, you can find PDF versions of conference posters presented in the last few years.My Ph.D. work is focused on investigating certain aspects of pitch processing and absolute pitch in musicians. Below, you can find PDF versions of conference posters presented in the last few years.
Organization for Human Brain Mapping, San Franscisco, 2009Anatomical Markers of Musicianship and Absolute Pitch as Revealed by Deformation-Based Morphometry. Patrick Bermudez & Robert J. Zatorre Montreal Neurological Institute, McGill University  A number of studies have investigated aspects of gross cerebral morphology associated with musical training (e.g. Gaser and Schlaug, 2003; Keenan et al., 2001), mainly with the use of manual segmentation and voxel-based morphometry of magnetic resonance images (MRIs), but none have applied deformation-based morphometry to the description of anatomy in musicians either with or without absolute pitch (AP). This work reveals previously unreported anatomical differences between musicians and non-musicians as well as those correlated with absolute pitch proficiency by analyzing deformation data recovered during the nonlinear registration of anatomical images. Methods Deformation vectors recovered during nonlinear spatial normalization of anatomical images were used to analyze relative local expansion and contraction of tissue when fitting each subject to a standardized template by computing the Jacobian determinant for each vector in the deformation field. T1 MRIs were linearly registered to the ICBM152 nonlinear 6th generation template with a 12-parameter linear transformation (Collins et al., 1994; Grabner et al., 2006), RF inhomogeneity corrected (Sled et al., 1998) and tissue classified (Tohka et al., 2004). For the musician vs. non-musician contrast (71 and 64 subjects, respectively), each of the grey and white matter tissue classes was then averaged across subjects to create study-specific grey and white matter templates which served as targets for subsequent nonlinear registrations with a 4 mm node-spacing between vectors in the deformation grid. In a second analysis using a subset of 49 musicians who underwent computerized testing of absolute pitch proficiency, AP-test score was correlated with determinants computed from a deformation grid with 2 mm node-spacing in a region of interest encompassing the medial temporal regions, basal ganglia and midbrain. Displacement data were convolved with a 4 mm FWHM 3D Gaussian blurring kernel. Results Among the more prominent areas of difference between the musician and non-musician groups were greater expanses of tissue in the right anterior middle frontal gyrus and deep in the region of the right intra-parietal sulcus (IPS). Among musicians, an ROI analysis using a higher degree nonlinear normalization revealed a right lateralized correlation between AP test score and matter in the body of the hippocampus. Discussion High proficiency practice of instrumental music places extraordinary demands on several cognitive systems. The precise integration of auditory, sensory, motor and spatial information is crucial to the successful execution of a musical performance and the IPS is conjectured to have an important role in auditory and spatial transformations (Grefkes and Fink, 2005). For instance, a previous fMRI experiment (Foster and Zatorre, 2007) demonstrated a correlation between BOLD signal and performance on a melody transposition task in an IPS location similar to that which is reported in the present work. Our finding may therefore reflect the many years of intense musical practice incurred by musicians and the consequent demands placed on abilities subserved by these areas. Little is known concerning the neuroanatomical correlates of absolute pitch, though the role of hippocampal and parahippocampal areas in memory formation is well known. Our finding is consistent with the suggestion that possessors of absolute pitch may incur somewhat different training effects as compared to their non-possessing musician cohort (Bermudez et al., 2008). References & acknowledgements Bermudez, P., Lerch, J. P., Evans, A. C. & Zatorre, R. J. (2008) Neuroanatomical Correlates of Musicianship as Revealed by Cortical Thickness and Voxel-Based Morphometry. Cereb Cortex, doi: 10.1093/cercor/bhn196. Collins, D., Peters, T. & Evans, A. (1994) An automated 3D non-linear deformation procedure for determination of gross morphometric variability in the human brain. Visualization in Biomedical Computing: Proc SPIE, 2359, 180-194. Foster, N. & Zatorre, R. J. (2007) A role for the intra-parietal sulcus in performing musical trasnposition judjments. Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, NeuroImage. Gaser, C. & Schlaug, G. (2003) Brain structures differ between musicians and non-musicians. J Neurosci, 23, 9240-5. Grabner, G., Janke, A. L., Budge, M. M., Smith, D., Pruessner, J. & Collins, D. L. (2006) Symmetric atlasing and model based segmentation: an application to the hippocampus in older adults. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv, 9, 58-66. Grefkes, C. & Fink, G. R. (2005) The functional organization of the intraparietal sulcus in humans and monkeys. Journal of Anatomy, 207, 3-17. Keenan, J. P., Thangaraj, V., Halpern, A. R. & Schlaug, G. (2001) Absolute pitch and planum temporale. Neuroimage, 14, 1402-8. Sled, J. G., Zijdenbos, A. P. & Evans, A. C. (1998) A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging, 17, 87-97. Tohka, J., Zijdenbos, A. & Evans, A. (2004) Fast and robust parameter estimation for statistical partial volume models in brain MRI. Neuroimage, 23, 84-97. This work was funded by the International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR).
|
THE NEUROSCIENCES AND MUSIC – III, Montreal, 2008Electrophysiological indices of absolute pitch perception. Patrick Bermudez, Robert J. Zatorre, Andreas Wollbrink, Sibylle Herholz & Christo Pantev Institut für Biomagnetismus and Biosignalanalyse, Münster; Montreal Neurological Institute, McGill University This work was funded by the Deutsche Forschungsgemeinschaft, International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR).
|
Organization for Human Brain Mapping, Florence, 2006Differences in cortical thickness between musicians and non-musicians. Patrick Bermudez, Jason P. Lerch, Alan C. Evans, Robert J. Zatorre Montreal Neurological Institute, McGill University A number of studies have investigated aspects of gross cerebral morphology associated with musical training (e.g. Gaser & Schlaug, 2003; Schneider et al., 2002), mainly with the use of manual segmentation and voxel-based morphometry of magnetic resonance images (MRIs). Automated methods for the extraction of cortical thickness from MRIs have been used successfully in experimental and descriptive studies of cerebral anatomy in various populations (e.g. Lerch et al., 2004). In the work described here, we use such a method to compare the cortical thickness of musicians and non-musicians with predicted differences in auditory, motor and dorsolateral frontal cortices. Methods Subject groups were 53 non-musicians and 53 musicians (10 years or more of musical experience, 21 with absolute pitch). T1 MR images were RF inhomogeneity corrected, linearly registered to the MNI152 symmetric template with a 9-parameter transformation and tissue classified. Following these steps, deformable models were used to first fit the white matter surface and then expand outward to find the gray matter/CSF intersection. Cortical thickness is defined as the distance between the linked vertices of the white and gray surfaces. Musician and non-musician thicknesses were then contrasted and evaluated according to the general linear model and significance thresholds established according to the False Discovery Rate theory. Results & Discussion The musicians versus non-musician contrast of cortical thickness reveals significantly greater thickness for musicians in the superior temporal surfaces, the motor cortices, broad areas of the prefrontal lobes, the left lingual gyrus and the right parahippocampal gyrus. The auditory and motor cortices have previously been shown to be morphologically distinct in musicians with the use of other techniques. As a measure, cortical thickness is complimentary yet more specific and constrained than the typical VBM measure which usually conveys information about extent, shape and position concurrently. The mid-dorsolateral frontal areas are thought to be of particular importance in subserving working memory function, an aspect of cognition very heavily relied upon in music perception and production. The parahippocampal areas are suspected to accept multi-modal input and play important roles in various types of memory formation. Results for separate analyses of musicians with and without absolute pitch will also be presented.. References & acknowledgements Gaser, C. & Schlaug, G. (2003). J Neurosci 23:9240-9245. Lerch, J.P., Pruessner, J.C., Zijdenbos, A. et al. (2005). Cereb Cortex 15:995-1001. Schneider, P., Scherg, M., Dosch, H.G., Specht, H.J., Gutschalk, A. & Rupp, A. (2002). Nat Neurosci 5:688-694. This work was funded by the International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR).
|
Organization for Human Brain Mapping, Toronto, 2005Differences in gray matter between musicians and non-musicians. Patrick Bermudez & Robert J. Zatorre, There exists a controversial literature regarding gross morphological differences in the brain structure of musicians as compared to non-musicians. Several authors have shown changes in diverse cerebral regions associated with musical training (e.g. Gaser et al., 2003; Schneider et al., 2002), but these have not always been consistent across studies. Here we present new data bearing on such differences with the use of voxel-based morphometry (VBM) of magnetic resonance brain images (MRIs). Methods Subject groups were made up of both males and females, 51 non-musicians and 43 musicians (10 years or more of musical experience, 22 with absolute pitch). Images were linearly registered to the MNI symmetric 152 template with a 12-parameter cost-minimization fit (Collins et al., 1994) and then RF inhomogeneity corrected (Sled et al., 1998) and tissue classified (Zijdenbos et al., 1998). The gray matter class was extracted and blurred using an 8 mm Gaussian kernel (Figure 1). Musician and non-musician images were then contrasted and evaluated according to the general linear model and random field theory correction for multiple comparisons (Worsley et al., 1996). Results & Discussion The main result is a greater gray matter concentration in musicians as compared to non-musicians in the right lateral surface of superior temporal gyrus, posterior to Heschl’s gyrus (auditory belt cortex; Figure 2). VBM results derived from images that are linearly transformed to a stereotaxic space concurrently communicate information about size, spatial extent and morphology. We must therefore be cautious in our interpretation. These and adjacent areas of the superior temporal gyrus have been previously implicated in a number of functional imaging studies, as well as in older brain-lesion studies, as being important for processing of pitch. They have also been implicated as differential morphological markers in volumetric studies using manual segmentation (Zatorre et al., 2002). Conclusion VBM is a data driven technique that does not rely upon apriori definitions of anatomical circumscription and, in this way, is free from systematic errors that can arise from arbitrary definitions. Our results suggest an experience dependent difference between musicians and non-musicians in areas known to be important in pitch processing. References & acknowledgements Collins, D.L., Neelin, P., Peters, T.M., and Evans, A.C. (1994). J. Comput. Assist. Tomogr., 281, 567–585. Gaser, C. and Schlaug, G. (2003). The Journal of Neuroscience, 23, 9240–9245. Schneider, P., Scherg, M., Dosch, H.G., Specht, H.J., Gutschalk, A. and Rupp, A. (2002). Nature Neuroscience, 688-694. Sled, J.G., Zijdenbos, A.P., and Evans, A.C. (1998). IEEE Trans. Med. Imag., 17, 87–97. Worsley, K.J., Marrett, S., Neelin, P., Vandal, A.C., Friston, K.J., and Evans, A.C. (1996). Human Brain Mapping, 4, 58-73. Zatorre, R.J., Belin, P. and Penhune, V.B. (2002). Trends in Cognitive Sciences, 6, 37-46. Zijdenbos, A.P., Forghani, R. and Evans, A.C. (1998). Proc. Int. Conf. Med. Im. Comput. Assis. Interven., MICCAI, 439–448. This work was funded by the International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR).
|
Organization for Human Brain Mapping, New York City, 2003Tonal working memory in absolute pitch musicians. P. Bermudez & R.J. Zatorre, Montreal Neurological Institute, McGill University, Montreal, PQ, H3A 2B4 Introduction In music, working memory is of particular importance since sounds unfold over time and must be held for relationships between events to be discerned. This maintenance and manipulation of information in real time contributes to many complex cognitive processes and seems to be subserved by discrete areas of the frontal cortex. It has been suggested that Absolute Pitch (AP), the ability to identify the names of musical pitches without reference to a standard, may diminish typical working memory requirements in the processing of tonal information amongst its possessors. Based on event-related potential (ERP) experiments1, it has been hypothesized that AP subjects do not update their working memory representation upon hearing new tones because the tones correspond to a fixed standard and, therefore, do not require maintenance in working memory. Evidence for this assertion comes from the absence or reduction of an ERP component (the P300, thought to be an index of working memory updating2) among AP subjects3,4. We hypothesized that regions involved in auditory working memory would therefore be more active in non-AP than AP musicians when performing an oddball-type task as part of an event-related fMRI paradigm. Methods Subjects Nine highly trained musicians were screened and categorized into one of two groups: AP and non-AP possessors. This was done by means of a computer-administered AP test which presented second-long synthetic tones ranging from C3 to B5 and prompted the subject for chroma and octave judgments for each note (see Figure 1 for distributions of average performance). Scanning Scanning was performed on a 1.5T magnetic resonance imaging scanner using temporally sparse acquisitions in an event-related design (Figure 2). The functional volume consisted of 20 contiguous 5 mm-thick axial T2* gradient echo EPIs aligned in plane with the Sylvian fissure. A T1-weighted volume was acquired for anatomical localization in each individual. Stimuli consisted of a frequent tone (C4, 261.62Hz) and an oddball tone (G4, 391.99Hz) of 1 s duration. There were two runs of 120 volume acquisitions at 10 second intervals with 5 stimuli per interval. Stimulus presentation was grouped into three trial types: 1) oddball, where 1 of the 5 stimuli presented (20%) was an infrequent tone, 2) frequent, where all stimuli were frequent tones and, 3) silence. Subjects were instructed to respond with a left button mouse click to frequent stimuli and a right button click to oddball stimuli. Results Both the AP and non-AP groups showed activity in primary and secondary auditory cortices related to the perception of auditory stimuli, as well as left posterior parietal activity. The principal contrast of interest, where activity related to the perception of the oddball stimulus is compared between groups, revealed the hypothesized difference in the frontal cortex thought to reflect differential use of working memory (Figure 3). References: 1) Klein et al. (1984) Science, 223, 1306-1309. 2) Donchin & Coles (1988) Behav. Brain Sci., 11, 357-374. 3) Hantz et al. (1992) Music Percept., 10, 25-42. 4) Wayman et al. (1992) J. Acoust. Soc. Am., 91, 3527-3531. This work was funded by the International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR). |
Society for Neuroscience, Orlando, 2002CONDITIONAL ASSOCIATIVE MEMORY FOR MUSICAL STIMULI IN NON-MUSICIANS: RELATIONSHIP TO ABSOLUTE PITCH. P. Bermudez & R.J. Zatorre*, Montreal Neurological Institute, McGill University, Montreal, PQ, H3A 2B4 An fMRI study was conducted to test the hypothesis that Absolute Pitch (AP), the ability to identify the names of musical pitches without reference to a standard, involves, in part, the ability to store and retrieve an arbitrary association between a stimulus attribute (the pitch of a sound) and a verbal label (note name). The posterior dorsolateral frontal cortex (pDFC) has been implicated in conditional associative learning with many types of stimuli, and a prior PET study showed pDFC activity in AP musicians, but not controls, while listening to single tones. It was therefore predicted that non-musicians would also show activation in the pDFC while performing a task designed to have limited analogy to AP. 8 normal volunteers with no formal musical knowledge underwent 2 fMRI scanning sessions: one pre-training baseline session and one post-training session. During offline training, subjects learned to associate each of 4 chordal stimuli with one of 4 numbers. fMRI was performed using a clustered acquisition in a block design. During the active condition, subjects heard one chord every 10s and identified it with a button press. Both the pre- and post-training scans showed activation in primary and secondary auditory cortices related to the perception of the auditory stimuli, but only the post-training scan showed significant bilateral activity in the pDFC, as well as ventrolateral frontal cortices. This indicates that pDFC is involved in associative learning of chord names. We infer that part of the chain of processing involved in AP, the conditional associative pairing of the perception of pitch with a note name, involves a similar process and hence recruits the pDFC. This work was funded by the International Foundation for Music Research (IFMR) and the Canadian Institutes of Health Research (CIHR). |
