Human Cognitive Neurophysiology Laboratory

Posters and Presentations

Posters may be downloaded in PDF format. You may view these files with the Adobe Acrobat reader available from www.adobe.com.

Z. E. Doss, E. W. Yund and D. L. Woods (2011): Perception of Consonants and Sentences by Young and Older Listeners
2011 Meeting of the America Auditory Society, Scottsdale, AZ, USA [Abstract] [PDF version] downloads.
D. L. Woods, A. Cate, T. J. Herron, E. W. Yund, and X. J. Kang (2009): Phonological Processing in Human Auditory Cortex
Neuroscience 2009, 39th Annual Meeting of the Society for Neuroscience, Chicago, IL, USA [Abstract] [PDF version] downloads.
X. J. Kang, T. J. Herron and D. L. Woods (2009): The Analysis of Peri-Cortical White and Gray Matter: A Comparison of FA and MTR
Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA [Abstract] [PDF version] downloads.
X. J. Kang, T. J. Herron and D. L. Woods (2009): Correlation Fiber tract characterization: Is the Ellipsoidal Area Ratio Better Than FA?
Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA [Abstract] [PDF version] downloads.
K. Alho, T. Rinne, T. J. Herron and D. L. Woods (2009): Meta-analysis of functional magnetic resonance imaging studies of human auditory cortex
Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA [Abstract] [PDF version] downloads.
T. J. Herron, X. J. Kang and D. L. Woods (2009): A toolbox for the visualization and meta-analysis of functional brain organization of the cortical surface using an anatomical database
Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA [Abstract] [PDF version] downloads.
A. D. Cate, T. J. Herron, X. J. Kang, E. W. Yund and D. L. Woods (2009): Sensory and attention-dependent regions in human extrastriate visual cortex
Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA [Abstract] [PDF version] downloads.
E. W. Yund, P. L. Divenyi, M. I. A. Ua Cruadhlaoich and D. L. Woods (2009): The Range of Signal-to-Noise Ratios Needed to Equalize Identification of Different Consonants
157th Meeting of the Acoustical Society of America, Portland, OR, USA [Abstract] [PDF version] downloads.
A. D. Cate, X. J. Kang, T. J. Herron and D. L. Woods (2009): Event-related fMRI of Integral and Separable Dimensions in Simple Visual Objects
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
T. J. Herron, X. J. Kang and D. L. Woods (2009): An improved toolbox for the visualization and meta-analysis of functional activations on the cortical surface
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
A. U. Turken, T. J. Herron, X. J. Kang and D. L. Woods (2008): Assessing Diffuse Axonal Injury in the Corpus Callosum Using Multimodal Imaging
Proceedings of 14th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Melbourne, Australia [Abstract] [PDF version] downloads.
X. J. Kang, T. J. Herron, A. Turken and D. L. Woods (2008): Anisotropic Diffusion Properties Near The Cortical Surface
Proceedings of 14th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Melbourne, Australia [Abstract] [PDF version] downloads.
K. Alho, T. Rinne, T. J. Herron and D. L. Woods (2008): The Functional Organization of Humna Auditory Cortex: Meta-Analysis of fMRI Studies
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
A. D. Cate, T. J. Herron, X. J. Kang, E. W. Yund, D. L. Woods (2008): Separating the Effects of Intermodal Selective Attention, Auditory Distraction and Stimulus Type on Visual fMRI Activations
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
T. J. Herron, X. J. Kang, K. Alho and D. L. Woods (2008): A Meta-Analysis Toolbox for Analyzing Regional Cortical Functional Organization
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
A. U. Turken, T. J. Herron, X. J. Kang, E. W. Yund and D. L. Woods (2008): Behavioral Correlates of Subtle Anatomical Abnormalities Following Traumatic Brain Injury
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
D. L. Woods, G. C. Stecker, T. J. Rinne, A. D. Cate, I. Liao, T. J. Herron, X. J. Kang and E. W. Yund (2008): Stimulus and Attention Effects on Sound Activations in Human Auditory Cortex
Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA [Abstract] [PDF version] downloads.
T. J. Herron, A. Turken, X. J. Kang and D. L. Woods (2007): Quantitative Automated Lesion Detection After Traumatic Brain Injury
Society for Neuroscience Annual Meeting, San Diego, CA, USA [Abstract] [PDF version] downloads.
X. J. Kang, T. Herron, E. W. Yund, and D. L. Woods (2007): Correlation Between Functional Activations And Fiber Directions in Human Auditory Cortex.
Proceedings of 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL, June 10-14, 2007. [Abstract] [PDF version] downloads.
A. Cate, T. Herron, X. J. Kang, E. W. Yund, and D. L. Woods (2007): Modulation of occipital cortex representing the peripheral visual field by auditory attention.
Proceedings of 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL, June 10-14, 2007. [Abstract] [PDF version] downloads.
E. W. Yund and D. L. Woods (2006): Learning in repeated measurements of speech reception thresholds for sentences in speech-spectrum noise.
IHCON, Lake Tahoe, CA [Abstract] [PDF version] downloads.
G. C. Stecker, T. Rinne, T. Herron, I. Liao, X. J. Kang, E. W. Yund, and D. L. Woods (2006): Assessing binaural interaction in functional magnetic resonance imaging of human inferior colliculus and auditory cortex.
ARO, Baltimore, MD [Abstract] [PDF version] downloads.
X. Kang, T Herron, E W Yund, and D L. Woods (2006): Interhemispheric Differences In Gyral Structure
Proceedings of 12th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Florence, Italy. June 11-15, 2006. [Abstract] [PDF version] downloads.
X Kang, E.W. Yund, T Herron, D Woods (2006): Improving the Resolution of Functional Maps on the Cortical Surface
Proceedings of The 14th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Seattle, Washington, USA, 6-12 May 2006 [Abstract] [PDF version] downloads.
C. Stecker, I. Liao, T. Rinne, T. Herron, X. J. Kang, E. W. Yund, and D. L. Woods (2005): FUNCTIONAL SPECIALIZATION OF HUMAN AUDITORY CORTICAL FIELDS
Society for Neuroscience Annual Meeting, Washington, DC, USA [Abstract] [PDF version] downloads.
Kang X. J., Herron T. J., Liao, I., Rinne T., Stecker C., Yund E. W. and Woods D. (2005): Visualizing Diffusion - Tensor Imaging Parameters On The Cortical Surface
Proceedings of 11th Annual Meeting of the Organization for Human Brain Mapping (OHBM). NeuroImage:26, Supplement 1. pps23, Toronto, Canada, June 12-16, 2005. [Abstract] [PDF version] downloads.
G. Christopher Stecker, Teemu Rinne, Isaac H. Liao, Xiaojian Kang, E. William Yund, Timothy J. Herron, David L. Woods (2005): Effects of Continuous Broadband Noise on Tone-Evoked Activations In Human Auditory Cortex
Association for Research in Otolaryngology 28th Midwinter Meeting, New Orleans LA, Feb 19-24 2005. [Abstract] [PDF version] downloads.
G. A. Bowman, G. C. Stecker, E. W. Yund, T. J. Herron, C. M. Roup and D. L. Woods (2005): At-Home Computer-Based Adaptive Training Improves Phoneme Processing In The Hearing Impaired
Association for Research in Otolaryngology 28th Midwinter Meeting, New Orleans LA, Feb 19-24 2005. [Abstract] [PDF version] downloads.
X Kang, C Stecker, T Rinne, I Liao, E.W. Yund, T Herron, D Woods (2004): Projection Maps of Inflated Cortical Surface of Human Brains
Proceedings of The 13th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, South Beach, Miami, Florida, USA, May 7-13, 2005 [Abstract] [PDF version] downloads.
I. Liao, T.J. Herron, X.J. Kang, E.W. Yund, C.I. Petkov, K.Alho and D.L. Woods (2004): Cortical Substrates of Intermodal Selective Attention
Proceedings of the Society for Neuroscience 34st Annual Meeting. San Diego, California, USA [Abstract] [PDF version] downloads.
X. Kang, B. Yund, D. Woods: Stochastic Co-Registration of Functional And Anatomical Data Improves the Spatial Resolution of fMRI
Proceedings of The 12th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Kyoto, Japan, May 15-21, 2003. [Abstract] [PDF version] downloads.
Kang X. J., Yund E. W., and Woods D: Cortical Thickness Measurements Of Human Auditory Cortex
Proceedings of The 11th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Toronto, Canada, July 10-16, 2003. [Abstract] [PDF version] downloads.
Kang X. J., Yund E. W., and Woods D: Improved Resolution of Functional Maps with Stochastic Co-Registration of Functional and Anatomical Images
Proceedings of Ninth Annual Meeting of the Organization for Human Brain Mapping (OHBM), NeuroImage: 19(2), Supplement 1. ppe 1274, New York City, NY, USA, June 18-20, 2003. [Abstract] [PDF version] downloads.
D. L. Woods, E. W. Yund and X. J. Kang: Unified Functional/anatomical Maps Of Human Auditory Cortex
Cognitive Neurosciences, NYC, March 31, 2003 [Abstract] [PDF version] downloads.
Kang X. J., and Woods D: Local Functional Mapping of Human Auditory Cortex
Proceedings of The 10th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Honolulu, Hawaii, USA, May 18-24, 2002. [Abstract] [PDF version] downloads.
D. L. Woods, E. W. Yund, C. M. Roup, and X. J. Kang: Intersubject Variability in Human Auditory Cortex Organization
Soc. Neuro. Abs. 28: 2002. [Abstract] [PDF version] downloads.
D. L Woods, C. Alain, and R. T. Knight: THE EFFECTS OF BRAIN LESIONS ON TARGET PROCESSING: A COMPARISON OF ERPs AND EVENT-RELATED DIFFERENCE SPECTRA (ERDISPs)
Cognitive Neuroscience Society, Sixth Annual Meeting, Washington DC, April 11-13, 1999 [Abstract]
D. L. Woods, K. Hartikainen, K. Ogawa, S. J. Thomas and E. W. Yund: ELECTROPHYSIOLOGICAL STUDIES OF DYADIC SEARCH
Soc. Neurosci. Abstr., Vol. 24, Part 2, pg. 1680, 1998 [Abstract] [PDF version] downloads.

Perception of Consonants and Sentences by Young and Older Listeners

Z. E. Doss, E. W. Yund and D. L. Woods (2011)

2011 Meeting of the America Auditory Society, Scottsdale, AZ, USA

Two groups of 16 normal-hearing listeners (mean ages 64.3 and 24.2 yrs) identified twenty initial and final consonants in consonant-vowel-consonant syllables (CVCs) embedded in speech-spectrum noise using the California Syllable Test (CaST). We also measured speech recognition thresholds (SRTs) in both groups using the HINT and Quick-SIN to discover the relationship between consonant-threshold signal-to-noise ratios (SNRs) in CVCs and SRTs. The measured SRTs indicated that one third of the consonants (Group A: /sh/, /ch/, /t/, /s/, /z/, /j/, and /r/) were presented in sentences well above the SNRs required for their identification in CVCs. Another third (Group C: /b/, /v/, /TH/, /ng/, /th/, /p/, and /h/) occurred in sentences well below the SNRs needed for their identification in CVCs. The remaining consonants (Group B: /d/, /g/, /l/, /m/, /n/, /f/, and /k/) occurred in sentences near their identification thresholds in CVCs. Overall consonant identification thresholds were elevated by 2.7 dB in older listeners. Threshold elevations were much larger for Group C consonants (4.3 dB), than for Group A (1.7 dB) or Group B (2.3 dB) consonants. Thresholds for Group C consonants correlated with the magnitude of high-frequency threshold elevation among older listeners. The lack of age-related changes in HINT and QuickSIN SRTs indicates that Group C consonants are not important in determining sentence SRTs as other consonants, vowels, and semantic and syntactic cues.

[PDF version] [back to title list]

Phonological Processing in Human Auditory Cortex

D. L. Woods, A. Cate, T. J. Herron, E. W. Yund, and X. J. Kang (2009)

Neuroscience 2009, 39th Annual Meeting of the Society for Neuroscience, Chicago, IL, USA

Brain mechanisms underlying the phonological analysis of speech sounds remain imprecisely localized and incompletely understood. Previous studies have identified lateral regions of auditory cortex that are activated by human speech, but it remains unclear whether these regions are hardwired to process speech-like sounds or whether they reflect the automatic engagement of auditory attention. We used fMRI and population-based cortical surface analysis to characterize the activations in human auditory cortex to consonant-vowel-consonant syllables (CVCs) and amplitude-modulated speech-spectrum noise bursts (SSNBs) with similar frequency spectra. As in our previous studies [1], subjects performed a challenging intermodal selective attention task that permitted the isolation of (1) exogenous stimulus dependent activations (SDAs) during visual-attention conditions, and (2) attention-related modulations (ARMs) during auditory attention. CVCs and SSNBs produced similar SDAs over medial auditory core and belt fields of both hemispheres. However, in comparison with SSNBs, CVCs disproportionately activated a restricted parabelt vocalization area (PVA, Figure 1) along the lateral-anterior superior temporal gyrus of both hemispheres. Attention to CVCs resulted in enhanced activations in both central and lateral regions of auditory cortex including the PVA, with ARMs extending into the planum temporale, the superior temporal sulcus, the middle temporal gyrus, the angular gyrus, and motor-speech regions of both hemispheres. The results provide evidence of a region in auditory cortex that is specialized to process speech sounds and an extensive network of brain regions that is optionally engaged by phonological attention.

[PDF version] [back to title list]

The Analysis of Peri-Cortical White and Gray Matter: A Comparison of FA and MTR

X. J. Kang, T. J. Herron and D. L. Woods (2009)

Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA

Fractional Anisotropy (FA) in Diffusion Tensor Imaging (DTI) and is sensitive to the integrity and organization of axons, while Magnetization Transfer Ratio (MTR) in Magnetization Transfer Imaging (MTI) is primarily sensitive to the integrity of cell membranes and myelin. We investigated the relation between FA and MTR measurements from peri-cortical white matter in healthy subjects, and assessed diffuse axonal injury in two traumatic brain injury (TBI) patients without frank cortical lesions.

[PDF version] [back to title list]

Correlation Fiber tract characterization: Is the Ellipsoidal Area Ratio Better Than FA?

X. J. Kang, T. J. Herron and D. L. Woods (2009)

Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA

A new diffusion anisotropy index, Ellipsoidal Area Ratio (EAR), was described recently and proved to be less noise-sensitive than fractional anisotropy (FA) by theory and simulation. Here we show that EAR has higher signal to noise ratios than FA in average DTI data from 40 normal subjects. EAR was also more sensitive than FA in detecting white matter abnormalities in a patient with widespread diffuse axonal injury. Monte Carlo simulation showed that EAR's mean values are more biased by noise than FA when anisotropy is small, both for single fiber tracts and when fiber tracts cross. However, the improved signal to noise ratio of EAR relative to FA suggests that EAR may be a superior measure of anisotropy both in quantifying both deep white matter with relatively uniform fiber tracts and pericortical white matter structure with relatively low anisotropy and fiber crossings.

[PDF version] [back to title list]

Meta-analysis of functional magnetic resonance imaging studies of human auditory cortex

K. Alho, T. Rinne, T. J. Herron and D. L. Woods (2009)

Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA

We reviewed functional magnetic resonance imaging (fMRI) studies on auditory processing. 102 studies reporting Talairach or MNI coordinates for activation peaks in the auditory cortex (AC) related to processing of a particular auditory feature (e.g., pitch or location) or particular sounds (e.g., speech or human voice) were included in the present meta-analyses. Our aim was to examine whether different auditory features or different kinds of sound are processed in different parts of AC. The AC activation loci reported in previous studies were analyzed with the Matlab toolbox VAMCA (www.ebire.org/hcnlab). First, the coordinates were transformed fromTalairach to MNI space when needed. Then, the MNI coordinates were transformed into cortical surface coordinates based on median cortical surface loci of MNI coordinates in 60 healthy right-handed adults. The resulting surface loci are shown on average cortical surface of these 60 adults produced with FreeSurfer software. Median cortical-surface and MNI coordinates were calculated for each group of activations. Permutation testing was used to evaluate statistical significance of distances between median coordinates of two groups of activations. The main findings of these meta-analyses are presented.

[PDF version] [back to title list]

A toolbox for the visualization and meta-analysis of functional brain organization of the cortical surface using an anatomical database

T. J. Herron, X. J. Kang and D. L. Woods (2009)

Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA

The recently introduced MatLab toolbox VAMCA (Visualization And Meta-analysis on Cortical Anatomy) [1] provides surface-based visualization of mean cortical functional activations that are published as stereotaxic 3D coordinates (www/ebire.org/hcnlab). VAMCA uses a database of cortices from 72 healthy subjects to locate activations on a standardized cortical surface by extending the technique of multi-fiducial mapping [2] in order to perform meta-analyses. We demonstrate that the multifiducial method should be able to reproduce cortical locations of functional activations. Non-parametric statistical tests are provided for determining (a) whether two groups of foci are in the same cortical location; (b) the extent of overlap of the two groups' foci; and (c) whether two groups of foci are differentially concentrated in any of the anatomically defined regions of interest (ROI) as defined using FreeSurfer. We explore extending VAMCA's functionality in two ways. First, we perform a proof-of-principle on the inclusion of studies in meta-analyses that do not report results in stereotaxic coordinates. E.g., electronic journal images can be captured, normalized to 3D space in a semi-automated fashion, and then projected to the cortical surface and used in meta-analyses. Second, a new distance metric, the "least axon distance (LAD)", is introduced. LAD reflects the minimum distance passing through brain matter between any pair of cortical surface points and provides a functionally relevant alternative to the 3D and 2D distance metrics currently used in VAMCA's statistical tests.

[PDF version] [back to title list]

Sensory and attention-dependent regions in human extrastriate visual cortex

A. D. Cate, T. J. Herron, X. J. Kang, E. W. Yund and D. L. Woods (2009)

Proceedings of 15th Annual Meeting of the Organization for Human Brain Mapping (OHBM), San Francisco, CA, USA

Introduction: High-level regions in visual cortex are known to respond specifically to complex aspects of attended visual stimuli. For example, multiple subregions in ventral occipitotemporal (VOT) are robustly activated while viewing faces, body parts, etc. Numerous studies have shown that attending selectively to a stimulus can modulate fMRI activation, usually by having subjects direct attention among different visual stimuli. In the current experiment, we examined the attentional modulation of visual cortex in an intermodal selective attention paradigm. Since the visual stimulus was constant regardless of the engagement of attention, this study could examine effects across the entire cortical surface to identify regions whose activity reflected either purely sensory responses or processes supporting selective attention.
Methods: Nine healthy, young subjects each completed 6 hour-long fMRI sessions. Subjects performed a demanding one-back task using either visual stimuli (faces or words) or auditory tone triplets in a blocked design that included four conditions. In unimodal visual (UV) and auditory (UA) blocks subjects attended to visual or auditory stimuli, while in bimodal visual and auditory blocks (BV and BA) they attended to the cued modality while ignoring the other. Structural scans of the cortical hemispheres were segmented to isolate gray matter, inflated to a sphere and aligned to a common coordinate system for group analysis using FreeSurfer. Functional results were displayed on equal-area 2D projections of these spherical maps. Visual stimulus-dependent activations (SDAs) in response to unattended visual stimuli were identified by the BA-UA subtraction. Visual attention-related modulations (ARMs) were identified by the BV-BA subtraction; both of these conditions presented the same stimuli, and differed only by cued modality.
Results: Baseline visually responsive regions (defined by UV-UA subtraction) encompassed ventral occipitotemporal and posterior parietal cortex, in addition to retinotopic regions. Figures show that: 1) the apparently uniform baseline visual activation in the VOT region was actually divided into discrete subregions by the ARM-to-SDA ratio. While the bulk of the fusiform gyrus was purely stimulus-driven, the posterior occipitotemporal sulcus responded hardly at all to unattended stimuli in spite of being strongly modulated by attention. 2) Similarly, intraparietal sulcus activation depended largely on visual attention, but anteromedial subregions were purely stimulus-driven.
Conclusions: This study examined the entire cortical surface to isolate two sources of visual fMRI activation: passive sensory processing of stimuli and active attention to these sensory representations during task performance. The results reveal that extensive anterior VOT visual areas can be effectively activated by unattended visual stimuli, and are largely unmodulated by attention. In contrast, we also identified a VOT region that is predominantly attention-related, even though it closely abuts stimulus-driven voxels. This intermodal attention study complements the results of studies employing intramodal (purely visual) attention manipulations.

[PDF version] [back to title list]

The Range of Signal-to-Noise Ratios Needed to Equalize Identification of Different Consonants

E. W. Yund, P. L. Divenyi, M. I. A. Ua Cruadhlaoich and D. L. Woods (2009)

157th Meeting of the Acoustical Society of America, Portland, OR, USA

Individual consonants vary in intelligibility when speech is presented in constant-intensity speech-spectrum noise. This variation in consonant intelligibility is explained at various levels, ranging from bottom-up energetic masking of different-intensity auditory cues needed to identify different consonants, to the top-down use of context information. This study focuses on bottom-up auditory cues by measuring the differences in signal-to-noise ratio (SNR) needed to equate consonant identification in consonant-vowel-consonant (CVC) syllables. The goal is to develop a CVC test with high sensitivity to changes in consonant perception because each consonant is presented at SNRs above the perceptual floor and below the perceptual ceiling. In order to present the initial and final consonants of each CVC at independently controlled SNRs, the intensity of the noise was changed during 100 ms centered in the vowel. In 16 young normal-hearing listeners, the SNR range required to equate identification across the 20 leading and trailing consonants was greater than 40 dB. In spite of these large SNR variations occurring during the test (and sometimes even within a single CVC) the expected pattern of consonant confusions was obtained, suggesting that the mid-vowel adjustment of noise intensity did not interfere with normal consonant-perception processes.

[PDF version] [back to title list]

Event-related fMRI of Integral and Separable Dimensions in Simple Visual Objects

A. D. Cate, X. J. Kang, T. J. Herron and D. L. Woods (2009)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

This study tested the hypothesis that different regions in both ventral and dorsal extrastriate visual cortex encode information about the size and shape of simple visual objects. Event-related fMRI responses were measured to pairs of briefly-presented shapes taken from a set of 5 simple rectangles that varied parametrically in width and height. Stimuli were presented in a counterbalanced sequence that included baseline fixation trials. Trials were binned according to the pattern of dimensional change in the stimulus pair. After participants' cerebral hemispheres were inflated and normalized to a cortical surface curvature template using FreeSurfer, analyses distinguished voxels sensitive to changes in the integral object dimensions of aspect ratio and surface area, as well as to changes in width and height. Contrasts were performed to identify visual regions that showed greater BOLD responses to changes in aspect ratio (shape) than to changes in area and vice versa, and regions were characterized in terms of the degree to which they responded to changes in width and height as integral or separable dimensions. Voxels sensitive to aspect ratio changes were identified in the medial fusiform gyrus, and voxels sensitive to surface area changes were found in the posterior parahippocampal gyrus. Subregions of the posterior parietal cortex were also found to be sensitive to both criteria as well. This study shows that the functional specificity of extrastriate visual areas, including regions that are typically selective for images of complex objects, can be characterized in terms of basic modes information-processing using simple shapes.

[PDF version] [back to title list]

An improved toolbox for the visualization and meta-analysis of functional activations on the cortical surface

T. J. Herron, X. J. Kang and D. L. Woods (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

The recently introduced MatLab toolbox VAMCA (Visualization And Meta-analysis on Cortical Anatomy) [1] provides surface-based visualization of mean cortical functional activations that are published as stereotaxic 3D coordinates. VAMCA uses a database of cortices from 72 healthy subjects to locate activations on a standardized cortical surface by extending the technique of multi-fiducial mapping [2]. Non-parametric statistical tests are provided for determining (a) whether two groups of foci are in the same cortical location; (b) the extent of overlap of the two groups' foci; and (c) whether two groups of foci are differentially concentrated in any of the anatomically defined regions of interest (ROI). We extend VAMCA's functionality in three ways. First, we implement the mapping of anatomical areas to the cortical surface. This permits the inclusion of studies in meta-analyses that do not report results in stereotaxic coordinates. Second, we describe how normalized 3D functional activation images can be mapped onto the normalized cortical surface. E.g. journal articles providing stereotaxic coordinates with standard errors or volumes can be mapped as 3D foci ellipsoids. Further, electronic journal images can be captured, normalized to 3D space in a semi-automated fashion, and then projected to the cortical surface and used in meta-analyses. Third, we describe procedures for mapping an individual subject's activations to VAMCA's normalized cortical surface. This permits simple activation visualization on labeled cortical anatomy.

[PDF version] [back to title list]

Assessing Diffuse Axonal Injury in the Corpus Callosum Using Multimodal Imaging

A. U. Turken, T. J. Herron, X. J. Kang and D. L. Woods (2008)

Proceedings of 14th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Melbourne, Australia

Diffuse axonal injury (DAI) due to blunt head trauma is accentuated in the posterior section of the corpus callosum. In particular, the isthmus and the splenium are predilection sites for traumatic brain injury (TBI). In vivo assessment of TBI-related neuropathology is complicated by the diffuse and often subtle nature of the abnormalities which may not be visible on standard brain scans. An important objective for TBI research is the development of automated quantitative procedures for objective assessment of TBI-related brain abnormalities. To this end, we developed an approach that combines multimodal imaging, atlas-based image analysis, and a normative database to detect subtle abnormalities in the corpus callosum in individual TBI patients. Diffuse axonal injury in the corpus callosum could be detected in an individual TBI patient with no macroscopic lesions using automated quantitative multimodal imaging procedures. Loss of transcallosal fiber integrity was evident in reduced splenial volume and thinner isthmus, as well as reduced anisotropy and increased diffusivity in all compartments of the corpus callosum with respect to normal controls. Automated assessment of callosal abnormalities with multiple imaging modalities will provide a useful tool for diagnosis of TBI.

[PDF version] [back to title list]

Anisotropic Diffusion Properties Near The Cortical Surface

X. J. Kang, T. J. Herron, A. Turken and D. L. Woods (2008)

Proceedings of 14th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Melbourne, Australia

Microstructural properties of pericortical white matter fibers and cortical gray matter are important characteristics of normal tissue organization and tissue neuropathology. Here we report on the distributions of two tissue properties - mean diffusivity (MD) and fractional anisotropy (FA) of different cortical lobes near the cortical surface. The differences between the left and right hemispheres and between sexes for these measures are also discussed.

[PDF version] [back to title list]

The Functional Organization of Humna Auditory Cortex: Meta-Analysis of fMRI Studies.

K. Alho, T. Rinne, T. J. Herron and D. L. Woods (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

We reviewed 277 functional magnetic resonance imaging (fMRI) studies of human audition and found 81 studies reporting the MNI or Talairach coordinates for auditory-cortex activations related to the processing of pitch (28 studies reporting 124 auditory-cortex loci), timbre (6 studies, 17 loci), spatial location (12 studies, 46 loci), speech (26 studies, 140 loci), human voice (9 studies, 42 loci) or to selective attention (11 studies, 32 loci). We transformed these activation loci into 2D cortical surface coordinates and evaluated the statistical significance of the differences in median 2D and MNI coordinates using permutation testing. We found evidence for separate auditory "what" and "where" pathways, as in each hemisphere, pitch processing was associated with activations near the crossing of Heschl's gyrus (HG) and the superior temporal gyrus (STG), while location processing produced significantly posterior activations in the planum temporale. Speech processing, in turn, elicited activations similar to those produced by voice processing, with both foci located in STG lateral to HG and significantly lateral to activations related to processing of spectral features (pitch or timbre) of non-human sounds. Differences were also found in median locations of attention-related modulations (ARMs) of auditory-cortex activity between studies using speech as attended and unattended stimuli and studies using other sounds, ARMs for speech being located in STG anterior to ARMs for nonspeech. Our results indicate that the meta-analysis methods applied here are powerful tools for elucidating the functional organization of human auditory cortex.

[PDF version] [back to title list]

Separating the Effects of Intermodal Selective Attention, Auditory Distraction and Stimulus Type on Visual fMRI Activations.

A. D. Cate, T. J. Herron, X. J. Kang, E. W. Yund, D. L. Woods (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

Cortical surface mapping was used to distinguish the effects of intermodal selective attention, auditory distraction, visual stimulus type and their interactions on whole-brain fMRI visual activations. Nine subjects each completed 6 hour-long sessions employing both continuous (2.9s TR) and sparse image acquisition (10.4s TR), which reduced activations due to scanner acoustic noise. Subjects performed demanding one-back tasks with either visual (faces or words) or auditory (tone triplets) stimuli in a blocked factorial design that included four key conditions. In "bimodal visual" blocks subjects performed a visual task while ignoring auditory stimuli; in "unimodal visual" blocks no auditory distracters were present. Equivalent bimodal and unimodal auditory blocks were also included. Visual and auditory stimuli had randomly staggered onsets to reduce crossmodal integration effects. Subjects' cortical hemispheres were segmented to isolate gray matter, inflated to a sphere and aligned to a common coordinate system for group analysis using FreeSurfer. A bimodal visual vs. bimodal auditory contrast showed the effects of task modality-selective attention separate from stimulus-specific effects, since both conditions contained identical stimuli. These visual attention-related modulations (ARMs) comprised a specific subset of the stimulus-dependent activations present in occipitotemporal (OT) and intraparietal sulcus (IPS) regions. Additional contrasts isolated anterior visual ARM areas that were significantly more active when auditory stimuli were present. Interactions between the attentional and stimulus type factors of the design further specified well-defined subregions involved during face and word tasks, not only in OT but also in IPS. These results clarify how attention modulates neural responses to visual stimuli.

[PDF version] [back to title list]

A Meta-Analysis Toolbox for Analyzing Regional Cortical Functional Organization.

T. J. Herron, X. J. Kang, K. Alho and D. L. Woods (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

Meta-analytic studies of the localization of cortical functions are faced with two problems: (1) Visualizing activations reported as 3D coordinates (MNI or Talairach space) in relation to cortical surface anatomy; (2) Evaluating the statistical significance of differences in cortical activation foci in different categories of experiments. We describe a meta-analysis toolbox that permits 3D activation foci to be mapped to a standardized representation of the inflated cortical surface and that statistically analyzes differences in location on both the cortical surface and in 3D space. Three steps are involved. First, each 3D coordinate in MNI or Talairach space is transformed into Freesurfer spherical coordinates by identifying the cortical surface voxel nearest to the 3D coordinate in each of 63 normal, right-handed brains and by computing the median location of those 63 cortical surface points on a standardized hemispheric atlas. Second, topographic maps showing cortical surface foci locations in relation to average gyral and sulcal structures are displayed on Mollweide whole hemisphere projections either for different groups of studies or for different subjects in a single experiment. 3D foci distributions and median locations can also be visualized on the Colin MNI brain. Third, cortical surface and 3D locations from two different categories of experiments or from two different conditions in a single experiment are statistically analyzed for group location differences using non-parametric permutation testing. A MatLab toolbox that includes the display and statistical tools as well as the anatomical database will be available at www.ebire.org/hcnlab.

[PDF version] [back to title list]

Behavioral Correlates of Subtle Anatomical Abnormalities Following Traumatic Brain Injury

A. U. Turken, T. J. Herron, X. J. Kang, E. W. Yund and D. L. Woods (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

Unlike focal brain injury, the diffuse damage caused by traumatic brain injury (TBI) presents a special challenge for establishing the structural correlates of TBI-related cognitive impairments. We report the case of a TBI patient who has severe cognitive impairments, with apparently normal brain structure when evaluated with standard MRI scans. We used multi-modal imaging (high-resolution anatomical imaging and diffusion tensor MRI) and automated quantitative image analysis to detect statistically significant alterations in brain tissue properties with respect to a database of normal subjects. Neurocognitive assessment revealed reduced processing speed, impaired executive function, memory problems and poor performance on visual-perceptual tasks. Cortical surface-based analyses revealed thinning and increased diffusivity in the cortex and reduced pericortical white matter anisotropy. Regional abnormalities were pronounced in orbitofrontal, prefrontal and occipital regions bilaterally. Fiber integrity was reduced in major white matter tracts. The genu, isthmus and splenium of the corpus callosum had low anisotropy and high diffusivity, with significant thinning of the isthmus. These brain abnormalities were predictive of behavioral impairments. Perceptual-motor slowing is consistent with diffuse white matter injury, impaired executive control with frontal lobe dysfunction, and visual deficits with occipital lobe damage. Memory problems are likely to be associated with disruption of the prefrontal- hippocampal connections through the cingulum bundle. Consistent with posterior callosal damage, the patient was disproportionately impaired in comparing simultaneously presented visual stimuli. These findings highlight the importance of novel brain imaging and behavioral assessment procedures for assessing brain-behavior relationships in TBI research, an area receiving increasing attention from cognitive neuroscientists.

[PDF version] [back to title list]

Stimulus and Attention Effects on Sound Activations in Human Auditory Cortex.

D. L. Woods, G. C. Stecker, T. J. Rinne, A. D. Cate, I. Liao, T. J. Herron, X. J. Kang and E. W. Yund (2008)

Cognitive Neuroscience Society Annual Meeting, San Francisco, CA, USA

Previous fMRI studies have shown that activations of auditory cortex vary with the frequency, intensity, and location of sounds and with attention. We present a group analysis of auditory cortical activations in nine subjects who participated in a factorial experiment that varied these factors conjointly. Each subject was scanned on six separate sessions, three with sparse (TR=11.6s) and three with continuous (TR=2.9s) sampling. Activations were analyzed using cortical-surface mapping tools. Subjects attended to visual or auditory stimuli in separate blocks and performed a difficult one-back matching task in the attended modality. Unimodal stimulus sequences were included to isolate auditory stimulus-dependent activations (SDAs) by subtracting activations in visual-attention blocks without auditory stimuli from those in visual-attention blocks with concurrent sounds. Auditory SDAs were restricted to Heschl's gyrus (HG) and the adjacent superior temporal gyrus (STG) and were larger during sparse than continuous sampling. Low-frequency tones (~225 Hz) produced more extensive SDAs than loudness-matched highfrequency tones (~3600 Hz), particularly in mid-HG. Contralateral presentations enhanced SDAs throughout HG/STG, whereas increasing intensity (70 to 90 dB SPL) enhanced SDAs in medial HG. Attentionrelated modulations (ARMs) were isolated by subtracting activations in bimodal attend-visual blocks from those in bimodal attend-auditory blocks. ARMs were largest in lateral STG with little enhancement seen in medial HG. Additional analyses revealed further dissociations between ARMs and SDAs. For example, ARMs to monaural tones were enhanced in auditory cortex ipsilateral to stimulation. Auditory attention does not simply amplify auditory sensory responses but dynamically modulates processing in higher-order auditory cortical fields.

[PDF version] [back to title list]

Quantitative Automated Lesion Detection After Traumatic Brain Injury.

T. J. Herron, A. Turken, X. J. Kang and D. L. Woods (2007)

Society for Neuroscience Annual Meeting, San Diego, CA, USA

Patients with traumatic brain injury (TBI) and significant cognitive deficits often fail to show abnormalities on standard magnetic resonance images (MRI). We describe Quantitative Automated Lesion Detection procedures using multiple MRI modalities (T1, DTI and FLAIR) to identify and characterize structural abnormalities in individual TBI patients through statistical comparisons with a normative database. Cortical surface (CS) mapping techniques are used to obtain cortical thickness measurements supplemented with mean diffusivity (MD) and fractional anisotropy (FA) maps in pericortical white matter and deep gray matter. CS maps are used to evaluate abnormalities in cortical thickness and pericortical white matter. Reductions in subcortical white matter integrity are assessed by FA and measures of corpus callosum (CC) thickness using T1 images normalized to Witelson's segmentation. MD, FA, and FLAIR maps of other subcortical structures are aligned using automated normalization procedures. QALD detected cortical thinning and CC structural abnormalities in a 34 year old TBI patient with normal standard MRI scans.

[PDF version] [back to title list]

Correlation Between Functional Activations And Fiber Directions in Human Auditory Cortex.

X. J. Kang, T. Herron, E. W. Yund, and D. L. Woods (2007)

Proceedings of 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL, June 10-14, 2007.

The combination of fMRI and diffusion tensor imaging (DTI) offers an opportunity to study the relationship between brain connectivity and functional organization. Here we use cortical surface mapping techniques to examine the correlation between functional activations and fiber connectivity within auditory cortex.

[download PDF version] [back to title list]

Modulation of occipital cortex representing the peripheral visual field by auditory attention.

A. Cate, T. Herron, X. J. Kang, E. W. Yund, and D. L. Woods (2007)

Proceedings of 13th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Chicago, IL, June 10-14, 2007.

Introduction: Evidence has accumulated for functional and behavioral connections between primary auditory and visual cortices in humans, especially in the domains of crossmodal processing and cortical reorganization (Bavalier et al. 2006, TICS). Recent anatomical studies in macaques (Falchier et al. 2002, J. Neurosci.) have also revealed efferent projections from primary auditory cortex to occipital regions that represent the visual periphery.
Methods. High-resolution T1 anatomical images (matrix size: 0.94 mm3) of 7 subjects were acquired´1.30´256, voxel size 0.94´212´256 on a 1.5 T Philips Eclipse scanner. The cortical surface of each hemisphere was segmented and inflated and the gray-white boundary was co-registered to a standard spherical coordinate system using FreeSurfer. An equal-area Mollweide projection was used to transform the coregistered spherical surfaces onto 2D maps. Right hemispheres were mirror-imaged and the two hemispheres were aligned to minimize differences in surface curvature.
fMRI studies using high-rate unimodal and bimodal stimulus sequences were performed repeatedly (3 times each with sparse and continuous sampling) in seven healthy participants. Task conditions were designed to analyze the effects of stimulus modality and attention independently: two bimodal conditions (attend-auditory, attend-visual) and two unimodal conditions (auditory-only, visual-only). During bimodal blocks both auditory and visual stimuli were presented concurrently but with asynchronous onsets. Participants performed a one-back recognition task in one modality while ignoring the other.
Results. Regions selectively active during auditory attention were identified by contrasting the attend-auditory and attend-visual conditions. In addition to auditory cortex, we identified occipital regions in the lingual gyrus and cuneus, areas representing the visual periphery. In contrast, foveal visual regions were activated only during visual attention.
ROI analysis of these anteromedial occipital regions revealed that they responded selectively to attended (vs. unattended) sounds, and that they were more active during difficult auditory conditions that required greater attention. ROIs were defined using data from sparse image acquisition scans (to minimize scanner acoustic noise) and investigated with an independent data set using the same participants and task.
Conclusions. Systematic and robust activation of peripheral, but not foveal, occipital cortex occurs during selective attention to simple auditory stimuli. These activations were stronger during difficult discriminations that required increased auditory attention. These results support the tight coupling of auditory attention and the visual periphery, as suggested by previous neuroimaging research (Jack et al. 2006, Neuron). The current results suggest a key role for visual cortex in attending to purely auditory stimuli, contributing to the coordination of attention between auditory and visual modalities.

[download PDF version] [back to title list]

Learning in repeated measurements of speech reception thresholds for sentences in speech-spectrum noise.

E. W. Yund, and D. L. Woods (2006)

IHCON, Lake Tahoe, CA (2006)

Stable measurements of the speech perception in the same individuals are critical for studying the objective benefits of hearing aids, acclimatization and auditory rehabilitation training in the hearing impaired. In a previous study [Wilson, Bell & Koslowski (2003), J. Rehabil. Res. Dev. 40, 329-336], sentence thresholds in quiet improved across five sessions, showing large improvements even for sentences that were never repeated. Here, we investigate the measurement stability of the Hearing In Noise Test (HINT) in normal-hearing individuals over test sessions on five different days within a period of two weeks. For each subject, some of the HINT sentence lists were repeated in all five sessions, while others were presented uniquely in either the first or last session. Although repeated HINT lists showed consistent performance improvements over the five sessions, performance on unique HINT lists was the same in the first and last sessions. Factors that may account for differences in results from the previous study include (1) higher-intensity speech, (2) testing in noise, and (3) test procedures. Independent of the explanation for the differences, the present results indicate that the HINT provides stable measures of speech perception in noise as long as the same HINT lists are not repeated for the same subjects.

[download PDF version] [back to title list]

Assessing binaural interaction in functional magnetic resonance imaging of human inferior colliculus and auditory cortex.

G. C. Stecker, T. Rinne, T. Herron, I. Liao, X. J. Kang, E. W. Yund, and D. L. Woods (2006)

ARO, Baltimore, MD (2006)

We investigated the binaural sensitivities of fMRI BOLD responses in human inferior colliculus (IC) and auditory cortex (AC). Stimuli consisted of three sinusoidally amplitude modulated iterated-rippled noise bursts (16 iterations at 10 ms delay, AM: 35 Hz @ 90%), arranged as a 600-ms pedestal-signal-pedestal stimulus. The 100-ms “signal” portion was presented 2-dB above the level of the variable-duration (100-300 ms) pedestals, for consistency with a separate experiment manipulation auditory attention. Stimuli were presented 1/s at 85 dB SPL using Stax MRI-005 electrostatic earphones housed inside Bilsom circumaural muffs, which attenuated outside noise by ~35 dB. Echo-planar imaging (1.5 T Phillips Eclipse) was used to acquire a single 10mm slice (1.9x1.9x4 mm resolution) oriented obliquely through IC and AC in nine subjects who were each scanned twice. Slices were acquired every 4 s, a “semi-sparse” imaging procedure that reduced masking of stimuli by scanner noise. During imaging, subjects performed a difficult visual oddball detection task (to limit overt attending to auditory stimuli) while presented with auditory stimulation in 20-s “sound” blocks alternated with 12-s “silent” blocks. Each sound block contained signals delivered to the left ear, right ear, or binaurally... Differences between sound and silent blocks showed bilateral activation of AC in all conditions, which was strongest in conditions of contralateral stimulation. IC activations appeared strictly contralateral in monotic conditions, but bilateral in the diotic case. That is, they exhibited “E0-type” binaural interactions (activation by contralateral stimulation regardless of the presence or absence of ipsilateral input). In contrast, AC activations were dominated by EE-type interaction, as evidenced by extensive bilateral cortical activation with monotic stimulation. Analysis of individual subjects’ data showed regional subsets of AC activation E0-type interactions along with EI-type interactions, where AC activations to monotic sound were reduced by binaural stimulation.. The results suggest that distinct mechanisms of spatial hearing are engaged at cortical and collicular levels of the human auditory system

[download PDF version] [back to title list]

Interhemispheric Differences In Gyral Structure

X. Kang, T Herron, E W Yund, and D L. Woods (2006)

Proceedings of 12th Annual Meeting of the Organization for Human Brain Mapping (OHBM). Florence, Italy. June 11-15, 2006. (2006)

Averaging anatomical images provides objective information about mean anatomical structure and is critical for creating a representative anatomical space for locating activations across subjects in functional magnetic resonance (fMRI) studies. Surface-based alignment provides a more accurate alignment of the gyral anatomy of individual subjects than does traditional averaging in 3D space, but usually results in separate maps of the left and right hemispheres that are difficult to compare or combine. Here, we propose a method that generates 2D Mollweide projection maps of left and right hemispheres in a common coordinate space so that anatomical features can be objectively compared and/or combined across hemispheres as well as across genders. We demonstrate the method by characterizing inter-hemispheric differences in gyral structure and cortical thickness in a population of 60 subjects thereby elucidating human cortical cytoarchitectonics.

[download PDF version] [back to title list]

Improving the Resolution of Functional Maps on the Cortical Surface

X Kang, E.W. Yund, T Herron, D Woods (2006)

Proceedings of The 14th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Seattle, Washington, USA, 6-12 May 2006 (2006)

Accurate mapping of functional magnetic resonance imaging (fMRI) data to human cortical anatomy is critical for understanding human cortical function. In the widely used functional space analysis (FSA) method, functional images are realigned and resampled to a functional reference image and processed in functional space. Both simulated and experimental data show that the activation maps from FSA have low resolution and poor reproducibility when displayed on the maps of cortical surface. Here, we describe a new method, anatomical space analysis (ASA), whereby low-resolution functional images are co-registered and resampled directly into high-resolution anatomical space with subsequent data processing performed in high-resolution space. One major advantage of ASA is that minor head movements, which are generally assumed to reduce spatial resolution, can actually increase spatial resolution by providing multiple samples of the relationship between functional and anatomical space. Both simulations and analyses of real fMRI data show that ASA improves the precision, objectivity and reproducibility of functional brain image mapping.

[download PDF version] [back to title list]

FUNCTIONAL SPECIALIZATION OF HUMAN AUDITORY CORTICAL FIELDS

C. Stecker, I. Liao, T. Rinne, T. Herron, X. J. Kang, E. W. Yund, and D. L. Woods (2005)

Society for Neuroscience Annual Meeting, Washington, DC, USA (2005)

Human auditory cortical fields (ACFs) exhibit differential sensitivity to different features of auditory signals, and are differentially modulated by attention. We mapped activations on the auditory cortical surface using functional magnetic resonance (fMRI) at 1.5T with both sparse and continuous sampling. During 20s blocks, subjects attended to either auditory or visual stimuli in balanced bimodal stimulus streams. Auditory stimuli were randomly ordered sequences of three tones (pattern duration = 750 ms, pattern onset asynchrony = 1.4s) presented over a narrow frequency range. The center tone frequency was 225 Hz, 900 Hz or 3200 Hz in different blocks, and the other two tones differed by +/- 6 semitones. Tone patterns varied factorially in center frequency, intensity (65 or 85 dB SPL), and lateralization (left, right or diotic) across blocks. Sounds were absent on 25% of the blocks. Visual stimuli were words or faces. Subjects performed a one-back matching task in the cued modality. Auditory stimulus-dependent activations (SDAs, the difference between sound-present and sound-absent blocks) were isolated during visual attention conditions. Attention-related modulations (ARMs) were isolated by subtracting activations to sounds during visual attention conditions from activations to the same sounds during auditory attention. Individual subjects showed highly replicable activation maps in auditory cortex over different days of testing. Discrete foci showed distinct patterns of modulation to changes in sound frequency, intensity, lateralization, and selective attention that may reflect the characteristic functional properties of different human ACFs.

[download PDF version] [back to title list]

Visualizing Diffusion - Tensor Imaging Parameters On The Cortical Surface

X. Kang, T. Herron, I. Liao, T. Rinne, C. Stecker, E. W. Yund and D. L. Woods (2005)

Proceedings of Ninth Annual Meeting of the Organization for Human Brain Mapping (OHBM), Toronto, Canada, June 12-16, 2005 (2005)

A new method is proposed to visualize the DTI parameters onto the mean 2D projection maps of the cortical surface. The cortical surfaces of 40 subjects were inflated and co-registered to a spherical coordinate system using FreeSurfer. An equal-area Mollweide projection was used to project spherical surfaces onto 2D maps. The average 2D map of left and right hemispheres provides a common coordinate space for comparing DTI data across subjects and hemispheres. Mean diffusivity (MD) and fractional anisotropy (FA) values were examined on the mean 2D map of whole cortical surface. The differences between the left and right hemispheres for these measures were also discussed.

[download PDF version] [back to title list]

Effects of Continuous Broadband Noise on Tone-Evoked Activations In Human Auditory Cortex

G. Christopher Stecker, Teemu Rinne, Isaac H. Liao, Xiaojian Kang, E. William Yund, Timothy J. Herron, David L. Woods (2005)

Association for Research in Otolaryngology 28th Midwinter Meeting, New Orleans LA, Feb 19-24 2005. (2005)

A significant challenge for auditory functional magnetic resonance imaging (fMRI) is the control of spectrotemporally complex noise present in the scanning enviroment, e.g., that generated by the MRI coolant pump (“pump noise”). One potential means of controlling the effects of this noise is the addition of low-level continuous broadband “masking noise.” Here we examine the effects of such masking noise on tone-evoked BOLD fMRI activations in human auditory cortex. Full-brain images (24 slices, 2 x 2 x 5 mm resolution) were acquired using sparse echo-planar imaging (TR=10.8 s) in a Philips Eclipse 1.5 T scanner. Functional data were projected onto flat maps of the cortical surface based on each subject's high-resolution anatomical data. Each 22-min scanning run consisted of 21.6-s blocks of auditory stimulation or silence, presented in quiet or accompanied by continuous broadband masking noise while subjects performed a 1-back visual matching task. Each block presented sequences of 120-ms pure tones at a mean frequency of 225, 900, or 3600 Hz, with individual tone frequencies varying by +/- 5 semitones. Masking noise levels were adjusted psychophysically to provide equal detectability of high, medium, and low-frequency tone sequences embedded in pump noise. Tone sequences were presented approximately 15 dB above masked threshold. In all subjects, tonal stimulation in quiet produced widespread activation of auditory cortex (Heschl's gyrus and adjacent regions of middle superior temporal gyrus). In general, a subset of lateral activations was maintained when tonal stimulation was combined with continuous masking noise. Medial activations were reduced by the presence of masking noise. Despite similar detectability of tones across subjects, some showed more widespread reductions. The results suggest differences in the sensitivity of auditory cortical regions to continuous noise and corresponding differences in regional contributions to auditory stimulus processing. These results also have important implications for the use of masking noise to control the effects of acoustic noise in fMRI experiments.

[download PDF version] [back to title list]

At-Home Computer-Based Adaptive Training Improves Phoneme Processing In The Hearing Impaired

G. A. Bowman, G. C. Stecker, E. W. Yund, T. J. Herron, C. M. Roup and D. L. Woods (2005)

Association for Research in Otolaryngology 28th Midwinter Meeting, New Orleans LA, Feb 19-24 2005. (2005)

The present study evaluated the effects of PC-based at-home training on the phoneme identification ability of hearing-impaired patients. Patients were randomly assigned to training or control groups and given phoneme identification tests before and after hearing aid fitting. After preliminary testing, a personal computer was installed in the home of patients in the training group, along with software for presenting sounds, recording responses, and automatically reporting performance data via modem. Subjects trained for 40-60 minutes per day (5 days/wk) in a single-interval, nine-alternative forced-choice phoneme classification task. Training involved 27 consonant-vowel and 27 vowel-consonant phoneme pairs spoken by a male and female talker, presented in continuous speech-spectrum noise. During training, the signal-to-noise ratio was adjusted on a trial to trial basis, using a 1-up, 1-down adaptive procedure. Phoneme-classification performance, assessed after one, two, four, and eight weeks, showed greater improvement in the trained groups at all test intervals, with little overlap between groups. Training-related improvements were similar in magnitude to the improvements associated with initial hearing aid fitting. Improved phoneme detection performance was also seen for phonemes spoken by novel male and female voices not used during training. Supported by DC005814 and the VA Research Service.

[download PDF version] [back to title list]

Projection Maps of Inflated Cortical Surface of Human Brains

X Kang, C Stecker, T Rinne, I Liao, E.W. Yund, T Herron, D Woods (2004)

Proceedings of The 13th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, South Beach, Miami, Florida, USA, May 7-13, 2005 (2004)

Averaging anatomical images provides objective information about mean anatomical structure and is critical for creating a representative anatomical space for across-subjects statistics of activations visualized in functional magnetic resonance (fMRI) studies. Surface-based alignment provides a more accurate alignment of the gyral anatomy of individual subjects than does traditional averaging in 3D space, but usually results in separate maps of the left and right hemispheres that are difficult to compare or combine. Here, we propose a method that generates symmetrical maps of left and right hemispheres in a common coordinate space so that both anatomical and functional data can be objectively compared and/or combined across hemispheres. We demonstrate the method by characterizing interhemispheric differences in gyral structure and cortical thickness in a population of 28 subjects.

[download PDF version] [back to title list]

Cortical Substrates of Intermodal Selective Attention

I. Liao, T.J. Herron, X.J. Kang, E.W. Yund, C.I. Petkov, K.Alho and D.L. Woods (2004)

Proceedings of the Society for Neuroscience 34st Annual Meeting. San Diego, California, USA (2004)

Intermodal selective attention is known to enhance fMRI-BOLD activations in sensory cortex of the attended modality, but less is known about how intermodal attention modulates processing in association cortex. Here, subjects performed a one-back matching task to monaural sounds (FM sweeps, tones or noise bursts) or concurrently presented visual stimuli (shapes or patterns of moving dots). Two similar experiments were performed with stimulation rate, target density and task difficulty equated across modalities. Attention-related modulations (ARMs) were isolated by subtracting activations to nonattended stimuli from activations to the same stimuli when attended. Functional data from individual subjects were aligned by projecting activations on the surface of the cortex after it had been inflated to a sphere and coregistered with Freesurfer reference templates. The surface maps from each hemisphere were then projected onto grids using the Mollweide projection and averaged across subjects. The results show modality-specific ARMs in auditory and visual sensory cortices. In addition, prominent visual ARMs were seen in the intraparietal sulcus and the frontal eye fields. Complete poster: http://www.ebire.org/hcnlab/papers/Intermodal.pdf.

[download PDF version] [back to title list]

Stochastic Co-Registration of Functional And Anatomical Data Improves the Spatial Resolution of fMRI

X. Kang, B. Yund, D. Woods

Proceedings of The 12th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Kyoto, Japan, May 15-21, 2003.

Accurate co-registration of functional and anatomical data from individual subjects is critical for the precise mapping of function on the cortical surface using functional magnetic resonance imaging (fMRI). Because of minor head movements a different spatial sampling of functional activations is inevitably obtained from successive functional images gathered during an experiment. It is generally thought that minor movements interfere with the precision of functional mapping. Here a stochastic co-registration method is described wherein low-resolution functional images are re-sampled and co-registered individually with the reference brain in high-resolution anatomical brain space. Counter intuitively, small movements significantly improve spatial mapping precision when stochastic co-registration is used, as demonstrated in both simulations and fMRI experimental results.

[download PDF version] [back to title list]

Cortical Thickness Measurements Of Human Auditory Cortex

Kang X. J., Yund E. W., and Woods D

Proceedings of The 11th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Toronto, Canada, July 10-16, 2003.

A new method is described to perform the thickness measurement of auditory cortex based on the anatomically normalized population maps of auditory cortex. The new method can measure the mean thickness to sub-millimeter precision. Two correlations are also studied, that between the mean thickness and gray/white matter surface curvature and that between the mean thickness and functional activations in different regions of auditory cortex.

[download PDF version] [back to title list]

Local Functional Mapping of Human Auditory Cortex

Kang X. J., and Woods D

Proceedings of The 10th Scientific Meeting & Exhibition of the International Society for Magnetic Resonance in Medicine, Honolulu, Hawaii, USA, May 18-24, 2002.

Abstract. Human auditory cortex is functionally partitioned into small cortical fields (ACFs) specialized for processing different features of acoustic stimuli. Average fMRI activations in groups of subjects performing different auditory tasks can reveal ACF specialization. However, traditional across-subject average mapping techniques have a spatial resolution that is too coarse to differentiate activations in nearby ACFs. We developed a new method using local anatomical and functional landmarks to create average functional maps. The accuracy of the method was compared with that of traditional 3D brain image normalization, and averaging based on spherical coordinates. Landmark-based mapping provided improved spatial resolution of ACF functional activations.

[download PDF version] [back to title list]

Improved Resolution of Functional Maps with Stochastic Co-Registration of Functional and Anatomical Images

Kang X. J., Yund E. W., and Woods D

Proceedings of Ninth Annual Meeting of the Organization for Human Brain Mapping (OHBM), New York City, NY, USA, June 18-20, 2003.

Accurate co-registration of functional and anatomical data is critical for the precise mapping of cortical functions using functional magnetic resonance imaging (fMRI). It is generally assumed that even minor head movements interfere with the precision of this co-registration process. Here, we introduce a stochastic co-registration method where low-resolution functional images are re-sampled and co-registered in high-resolution anatomical brain space. Paradoxically, small head movements provide independent perspectives on the relationship between functional and anatomical space and hence improve the precision of co-registration using stochastic co-registration in comparison with traditional methods. Stochastic co-registration improves the precision and reproducibility of functional activation maps.

[download PDF version] [back to title list]

Unified Functional/anatomical Maps Of Human Auditory Cortex

D. L. Woods, E. W. Yund and X. J. Kang

Cognitive Neurosciences, NYC, March 31, 2003

ABSTRACT. Functional magnetic resonance imaging (fMRI) was used to characterize the organization of human auditory cortical fields (ACFs) in four subjects performing a difficult intermodal attention task. In different blocks subjects either (1) detected targets of reduced intensity among high-rate tones (10/sec) that varied in location (left, right or binaural) and frequency (250, 1000 or 4000 Hz) in different blocks; or (2) detected repetitions in random sequences of visual stimuli presented at fixation. Tones were omitted in some visual attention blocks. Differences between visual attention blocks with and without tones were used to isolate stimulus-dependent auditory activations (SDAAs). Differences between attend-auditory and attend-visual blocks were used to isolate attention-related activations. Individual maps of auditory cortex were made by inflating the cortex with Freesurfer, and sectioning and flattening auditory cortical patches. For each subject, within-subject activations replicated precisely across three separate experimental sessions. In most subjects, SDAAs formed parallel stripes 40-60 mm in length and lying anterior and posterior to Heschl’s gyrus. Five anatomical fiducial points were identified as in our previous studies. Local peaks in functional activations were also used to identify ten functional fiducial points. Then, average maps of auditory cortex were created by warping individual patches based either on (A) functional or (B) anatomical fiducial points. Functionally normalized maps showed somewhat sharpened activation foci in comparison with anatomically normalized data, but the differences were slight. The results suggest that the spatial relationship between functionally defined auditory cortical fields (ACFs) may be more precise than the relationship between ACFs and anatomical landmarks. However, additional data are needed to characterize individual ACFs with more precision if functional normalization is to more fully elucidate human ACF organization. Supported by DC05814 and the VA Research Service.

[download PDF version] [back to title list]

Intersubject Variability in Human Auditory Cortex Organization

D. L. Woods, E. W. Yund, C. M. Roup, and X. J. Kang

Soc. Neuro. Abs. 28: (2002).

In order to evaluate the variability in location and functional organization of human auditory cortical fields, fMRI-BOLD activations were obtained from individual subjects in a demanding, intermodal selective attention task. Subjects shifted attention between high-rate auditory sequences and concurrent visual tasks. During auditory attention, subjects detected occasional omissions of a tone from continuous 10/sec sequences, with tone frequency (250, 1000, and 4000 Hz) and location (left, right, binaural) varying randomly in 20-sec blocks. Subjects were repeatedly tested (up to eight experiments) and the results averaged over experiments to obtain high signal/noise ratios sufficient to image individual auditory fields. SensoryExogenous sensory activations associated with the processing of tone frequency and location were analyzed during visual attention blocks. Endogenous attention-related activations were derived by subtracting responses to nonattended signals from activations to the same signals when attended. Functional activations were projected onto maps of individual auditory cortices obtained by inflating the brains with FreeSurfer and sectioning and flattening the superior temporal plane, superior and middle temporal gyri, insula and operculum. Anatomically normalized average maps were obtained by warping maps to mean population values based on local anatomical structures (Woods et al, SFN, 2001). The results show considerable variability in the functional activation patterns of different subjects, and even from the same subject in different hemispheres. Supported by the VA Research Service and the NIDCD

[download PDF version] [back to title list]

THE EFFECTS OF BRAIN LESIONS ON TARGET PROCESSING: A COMPARISON OF ERPs AND EVENT-RELATED DIFFERENCE SPECTRA (ERDISPs)

D. L Woods, C. Alain, and R. T. Knight

Cognitive Neuroscience Society, Sixth Annual Meeting, Washington DC, April 11-13 1999

Target stimuli elicit event-related brain potentials (ERPs) and alter the background EEG. Time-domain signal averaging isolates time-locked ERPs, whereas frequency domain averaging isolates event-related difference spectra (ERDISPs) that include ERP activity as well as phase-varying evoked activity and modulations in background EEG rhythms. We compared target-evoked ERPs and ERDISPs in a visual oddball task in control subjects (N=16) and patients with unilateral lesions of the frontal lobe (N=10), temporal/parietal junction (N=7), and mesial temporal lobe including the hippocampus (N=7). In control subjects, ERPs showed N2 and P3 components. ERDISPs showed target-related increases in delta, theta and gamma power, and target-related reductions in alpha and beta power. Different patterns of lesion-induced abnormalities were seen in ERPs and EEG spectra. For example, hippocampal lesions did not affect parietal P3b amplitudes, but markedly diminished target-related theta enhancements. Post-target alpha desynchronization was reduced in hippocampal patients and selectively diminished over the lesioned hemisphere in temporal/parietal and frontal patients. The results suggest that ERDISP and ERPs index the functioning of distinct neural networks involved in target detection.

[back to title list]

ELECTROPHYSIOLOGICAL STUDIES OF DYADIC SEARCH

D. L. Woods, K. Hartikainen, K. Ogawa, S. J. Thomas and E. W. Yund

Soc. Neurosci. Abstr., Vol. 24, Part 2, pg. 1680, 1998

SUMMARY. Mechanisms of visual search were examined in 24 subjects performing a feature conjunction task. Either one or two stimuli were randomly presented on each trial. Stimuli varied in shape and color. They were randomly presented in the different visual quadrants at a high rate (mean 2/sec). In each block, one of the 16 stimuli was designated the target. Reaction times, accuracy and false alarm rates suggested that subjects analyzed stimuli in parallel on dyadic trials. Event-related brain potentials (ERPs) were extracted following all stimuli, and ERPs to dyads were compared to the sum of corresponding singlet ERPs. SENSORY RESPONSES. ERPs to non-attended stimuli (markedly different in both color and shape from the target) showed occlusive interactions beginning at the later portions of the P1 (110 ms). These interactions occurred regardless of whether the stimuli were presented in the same or different hemifields. The results suggest that neurons in V2 (thought to be the generator of the P1) are subject to short latency influences from large regions of the visual field. Later exogenous components (the frontal N1) showed occlusive interactions that appeared to reflect processes of figure-to-ground extraction or color-coding. These were enhanced for stimulus pairs differing in color. ATTENTION-EFFECTS. The shortest latency attention effects were found over the central scalp (130-150 ms) and over contralateral parietal regions. They reflected preferential processing of stimuli of target color. The parietal effects were enhanced for lower hemifield stimuli, providing evidence of spatiotopically organized generators. Conjunction-specific effects occurred at longer latencies. A comparison of attention effects on singlet and dyad trials showed suppressive interactions at posterior sites, and enhancements frontally. Increasing occlusion was seen at longer latencies. For example, dyad N2s (220 ms) were influenced by total number of target features in both stimuli, whereas dyad P3s (450 ms) had similar amplitudes whether one or two targets were present. The results suggest short-latency parallel processing of stimulus features, followed by an increasing focus on the most target-like stimulus. Supported by NS32893, and the VA Research Service.

[download PDF version] [back to title list]