special tutorial: principles of functional magnetic ... · pdf fileprinciples of functional...

J Am Acad Audiol 11 : 239-272 (2000)

SPECIAL TUTORIAL

Principles of Functional Magnetic Resonance Imaging: Application to Auditory Neuroscience Anthony T. Cacace*' Talin Tasciyan= Joseph P. Cousins°

Abstract

Functional imaging based on magnetic resonance methods is a new research frontier for exploring a wide range of central nervous system (CNS) functions, including information pro-cessing in sensory, motor, cognitive, and linguistic systems . Being able to localize and study human brain function in vivo, in relatively high resolution and in a noninvasive manner, makes this a technique of unparalleled importance . In order to appreciate and fully understand this area of investigation, a tutorial covering basic aspects of this methodology is presented . We introduce functional magnetic resonance imaging (fMRI) by providing an overview of the studies of different sensory systems in response to modality-specific stimuli, followed by an outline of other areas that have potential clinical relevance to the medical, cognitive, and communicative sciences . The discussion then focuses on the basic principles of magnetic resonance methods including magnetic resonance imaging, MR spectroscopy, fMRI, and the potential role that MR technology may play in understanding a wide range of auditory functions within the CNS, including tinnitus-related activity. Because the content of the mate-rial found herein might be unfamiliar to some, we provide a broad range of background and review articles to serve as a technical resource .

Key Words: Auditory system, functional magnetic resonance imaging, hearing, magnetic resonance spectroscopy, perception, phantom perception, tinnitus

Abbreviations : AL = anterior lateral, BOLD = blood oxygen level dependent, Cho = choline, CL = caudal lateral, CNS = central nervous system, CR = creatine, EEG = electro-encephalography, EPI = echo planar imaging, ER = event related, ETS = echo-time shifting, fMRI = functional magnetic resonance imaging, FOV = field of view, G = Gauss, GET = gaze-evoked tinnitus, GRE = gradient-recalled echo, MEG = magnetoencephalography, ml = myo-inositol, ML = medial lateral, MRI = magnetic resonance imaging, MRS = magnetic resonance spectroscopy, NAA = N-acetylaspartate, NMR = nuclear magnetic resonance, PET = positron emission tomography, PRESTO = principles of echo shifting with a train of observations, RF = radio frequency, ROI = regions of interest, SE = spin echo, STEAM = short-TE stimulated echo acquisition mode, T = Tesla, TE = echo time, TR = repetition time

W ithin the last decade, major strides have been made in understanding various forms of information pro-

cessing in vivo in the human brain. Advance-

*Departments of Surgery, tNeurology, and §Radiology, Albany Medical College, Albany, New York; $Sensor

Systems, Inc., Sterling, Virginia Reprint requests : Anthony T . Cacace, Department of

Surgery, Division of Otolaryngology, Albany Medical Col-lege, 47 New Scotland Avenue, Albany, NY 12208

ments in this area have been due in large part to the rapid growth of functional magnetic res-onance imaging (fMRI), which evaluates spa-tially sensitive hemodynamic changes in the central nervous system (CNS) in response to physiologic activations (for general reviews see Shulman et al, 1993; Cohen and Bookheimer, 1994 ; Ellerman et al, 1994 ; LeBihan and Karni, 1995 ; Cohen, 1996 ; Moseley et al, 1996 ; Ban-dettini and Wong, 1997 ; Kim and Ugurbil, 1997 ; Forster et al, 1998; Ogawa et al, 1998 ; Turner et al, 1998 ; Howseman and Bowtell, 1999). The

239

Journal of the American Academy of Audiology/Volume 11, Number 5, May 2000

N N

N

cc3

ms sec min

Relative time

Brain

Map

Column Layer

Neuron

Dendrite

Synapse

Non-invasive Invasive

temporal and spatial coupling between hemo-dynamic events (blood flow and metabolism) and synaptic activation links these two processes together and sets the stage for a discussion of functional imaging methods. Figure 1 provides a schematic diagram that compares fMRI with other experimental procedures used to study brain function in the context of their spa-tiotemporal properties and the degree of inva-siveness of the procedures . The benefits of using /MRI in the study of human neurofunction are grounded on the facts that it is noninvasive, does not require exposure to ionizing radiation or radiopharmaceuticals, has good temporal and spatial resolution, is well suited for single-sub-ject repeated-measurement designs, and the results can be superimposed on the individ-ual's own brain structure (MRI) without the need for image transformation or warping (Moo-nen et al, 1990 ; Cohen and Bookheimer, 1994; Toga and Mazziotta, 1996 ; Frackowiak et al, 1997). In the domain of information processing, studies using /MRI have successfully investi-gated visual (Blamire et al, 1992 ; Menon et al, 1993 ; Serano et al, 1995 ; Fransson et al, 1998), auditory (Melcher et al, 1999a), vestibular (Bucher et al, 1998; Lobel et al, 1998), olfactory (Koizuka et al, 1994 ; Levy et al, 1997 ; Sobel et al, 2000), gustatory (Faurion et al, 1998), motor (van Gelderen et al, 1995; Rao et al, 1996; Mat-tay and Weinberger, 1999), and cognitive/lin-guistic systems (Binder, 1997, 1999 ; Bullmore

Figure 1 Schematic diagram rep-resenting to a first order of approxi-mation the temporal (x-axis) versus spatial (y-axis) properties of various neuroscience methods used to study brain function . The y-axis encom-passes a spatial representation, rang-ing from macro (whole brain level) to the micro (synaptic level) . The x-axis encompasses a temporal representa-tion, ranging from milliseconds to days . The parameter of the graph is the degree of invasiveness of the method . Optical imaging of intrinsic signals, although not depicted on the diagram, shares similar spatiotem-

hrs days poral properties with that of /MRI . Functional MRI estimates include event-related and blocked-stimulus designs. Abbreviations: MEG = mag-netoencephalography, ERP = event-related electric potentials, /MRI = functional magnetic resonance imag-ing, PET = positron emission tomog-raphy. Adapted from Cohen and Bookheimer (1994) .

et al, 1996). Moreover, as basic science studies continue to explore new areas, clinical-research applications are also being realized. Most promi-nent in this area is the role that functional imaging may play in planning neurosurgical procedures (Holodny et al, 1999 ; Lee, 1999), particularly as image-guided technology is being applied with greater frequency in the operating room (Grimson et al, 1999). In this context, /MRI has the potential to replace more invasive procedures (i .e ., the Wada test) for establishing the language-dominant hemisphere, in help-ing to determine whether language is bilater-ally represented and/or whether language dominance is atypical (Binder et al, 1995a, b; Cuenod et al, 1995 ; Desmond et al, 1995; Shay-witz et al, 1995 ; Benson et al, 1996, 1999 ; Binder, 1997; Gaillard et al, 1997 ; Baciu et al, 1999 ; Frost et al, 1999 ; Lee et al, 1999 ; Springer et al, 1999). Evidence is also accumulating that sug-gests that good concordance exists between /MRI maps of language function and those obtained by more direct cortical stimulation methods in individuals with CNS lesions (Schlosser et al, 1999). When used alone or in combination with magnetic resonance spec-troscopy (MRS), magneto encephalography (MEG), and/or electroencephalography (EEG), /MRI can also provide spatially localizing infor-mation about seizure foci in epilepsy (Jackson et al, 1994 ; Morris et al, 1994; Detre et al, 1995; Warrach et al, 1996). If surgery becomes nec-

240

Functional Magnetic Resonance Imaging/Cacace et al

essary for seizure control, and/or in cases when operable CNS lesions (tumors, arteriovenous malformations, etc.) are in or adjacent to sen-sory, motor, and/or language areas of the brain, then delineating diseased from normal tissue takes on additional significance for reducing morbidity and improving quality of life after surgery (Fried et al, 1995 ; Latchaw et al, 1995; Atlas et al, 1996 ; Bookheimer, 1996; Maldjian et al, 1996 ; Righini et al, 1996 ; Achten et al, 1999).

Other clinical areas where /MRI can play an important role include the study of neu-rodegenerative disease (Moseley et al, 1996), headache (Cao et al, 1999b), in evaluating brain activation patterns in individuals with implanted sensory devices (cochlear implants) (Eddington et al, 1999 ; Melcher et al, 1999a; Truy, 1999), or from electrical stimulation in the ear canal (Hofman et al, 1999); the study of the neurobiology of various learning disabilities, such as dyslexia (Richardson et al, 1997 ; Eden and Zeffiro, 1998 ; Rae et al, 1998; Shaywitz et al, 1998; Richards et al, 1999), where standard imaging procedures have not been conclusive in identifying abnormalities (Beaton, 1997); and the study of plasticity and procedural learning in normals and in instances of CNS reorgani-zation secondary to physical injury, surgery, or as a result of intrinsic neurologic insult such as stroke (Karni et al, 1995 ; Krings et al, 1997; Graveline et al, 1998; Rossini et al, 1998 ; Toni et al, 1998; Cao et al, 1999 ; Thulborn et al, 1999).

Thus, in a relatively short period of time, f MRI research has provided insight into sensory, motor, and cognitive/linguistic processing sys-tems and is evolving in a variety of areas that have potential clinical value . This paper reviews the underlying physics and pertinent technical considerations necessary for understanding research in this area and focuses on the imag-ing of auditory-related activity using MRI, /MRI, and MRS . Whereas the majority of studies to date have focused on auditory system activation secondary to the use of external stimuli (tones, music, speech, etc .), this is not the only appli-cation for this technology. A challenging but less straightforward area of investigation concerns

the use of functional imaging methods in the study of phantom perceptions like tinnitus and/or other perceptual experiences that occur in the absence of overt (external) sensory stimulation (hallucinations [auditory and visual], pain, synes-thesias, mental imagery, etc .) (Cacace, 1999) . To date, the exciting area of /MRI research has

had an impact on theoretical, experimental, and clinical issues and is poised to play an important role in auditory neuroscience research .

MAGNETIC RESONANCE

Background and Historical Perspective

In the mid 1940s, two independent research groups, one at Harvard (Purcell et al, 1946) and the other at Stanford University (Bloch et al, 1946) demonstrated that nuclear magnetic res-onance transitions between two different spin states of an atom could be experimentally mea-sured . This hallmark discovery earned Purcell and Bloch Nobel prizes in chemistry and set the stage for future developments in MRS and MRI. Prior to 1983, the scientific designation of the Bloch/Purcell discovery was termed "nuclear magnetic resonance" (NMR). However, the com-mercial manufacturing of whole-body NMR imaging systems in the early 1980s brought about the removal of the letter "N" for market-ing reasons . The less accurate but more common uses of MRI and MRS were subsequently codi-fied in late 1983 by the American Journal of Roentgenology .

The initial application of MR was in the field of chemistry, where MRS was used to study the underlying structure of composite materials and biologic tissues in vitro. However, it would take another 30 years of refinements in various scientific disciplines to reach the current state of MR where in vivo spatial imaging applications are possible . Critical in this evolution were notable advancements in the areas of super-conducting materials, affordable computers, and mathematics. The mathematical compo-nent led to the development of the projection reconstruction method of image formation (Lauterbur, 1973). Image reconstruction based on the two-dimensional (2D) Fourier transfor-mation method, which forms the basis for cur-rent techniques, was developed thereafter (Kumar et al, 1975). Of course, many other individuals made unique and fundamental con-tributions to MR development. A more com-plete biography of these individuals and their contributions can be found elsewhere (Mattson and Simon, 1996). Thus, with this brief histor-ical overview as a background, we now con-sider basic principles of MR, since a working knowledge of this area is necessary to fully appreciate the current state of the art, where physiologic as well as exquisite neuroanatomic

241


information of the living human body and brain can be attained with available technology.

Magnetic Resonance: Basic Principles

From basic physics, it is well known that the nuclei of certain atoms, such as hydrogen and phosphorus, possess a spin and thus rotate about their own axis . Fundamental to this concept is the knowledge that a magnetic dipole moment is associated with this nuclear spin . In the pres-ence of an external magnetic field (Bo), such as in the case of an individual placed within an MRI scanner, the hydrogen atoms from water mole-cules or other metabolites tend to become direc-tionally polarized and align in the direction of the external magnetic field. This is analogous to the behavior of a compass needle in the earth's magnetic field, where magnetic moments may become aligned with the north or south poles of the magnet in either parallel or antiparallel directions . Since a lower amount of energy is associated with the parallel orientation, the number of magnetic moments in this direction slightly exceeds that of the antiparallel direction, resulting in a small net magnetization (M). This small net magnetization (M) is the source of the

Precession

Applied External Magnetic

Field (Bo)

Linear projection of the precession motion to the horizontal plane

ft

Figure 2 Top, An atom with nonzero nuclear spin is shown to precess (like a gyroscope) about an externally applied magnetic field (Bo). Bottom, The projection of the precession motion to a horizontal plane is repre-sented as a rotating vector. The position of the vector at any point in time can be represented in terms of its phase (360° of total cyclic motion).

MR signal. The alignment of M with Bo, however, is never complete, leaving room for M to move around or precess about the external magnetic field Bo like a gyroscope (Fig. 2), with a preces-sion frequency that is dependent on the nucleus and strength of the magnet . The precession or Larmor frequency (w) is represented by the equation :

w = -yBO (Equation 1)

where y, the gyromagnetic ratio, refers to the nucleus specific constant of proportionality. The gyromagnetic ratio for the hydrogen atom is y = 42.57 kH/Gauss (kHz/G). For many magnets in the clinical setting, Bo = 1.5 Tesla (T) (1 .5 x 104 G), although even higher field strengths are desired for functional imaging applications (Kim et al, 1996). As a point of ref-erence and for comparison with other magnetic phenomena, the average magnetic field strength at the surface of the earth is approximately 0.00005 T (0.5 G) .

The near alignment of the magnetization with the external field (i .e ., the longitudinal state), however, is not conducive to MR signal measurements . This results from the fact that when the magnetization is in the longitudinal direction, the area intercepted by the preces-sional motion is so narrow that the change of motion cannot be measured by the detection device (a receiver coil) used by the MR instru-mentation. Thus, the detection of the MR signal requires that the magnetization be tilted or nutated away from the longitudinal axis . This is accomplished by the use of circularly polarized radio frequency (RF) pulses (henceforth referred to as B). RF pulses are applied orthogonal (per-

Figure 3 A, The motion of the tip of the magnetic moment in the presence of magnetic fields B, and Bo, as observed by a viewer at a distance. B, The motion of the same magnetic moment as observed by a viewer pre-cessing about B,, with the same precession frequency as that of the magnetic moment . In this rotating frame of reference, only the precession about B, is observed at a distance .

242


pendicular) to Bo at the same Larmor frequency as that of the magnetic moments. The magnetic moments, which perceive the B1 field to be sta-tionary due to its rotation, start precessing about B1 and Bo, and in doing so are tilted into the transverse plane. Figures 3A and 3B illustrate how the tilting procedure is viewed from both the stationary and rotating frames of reference . Upon removal of the 131 field, the tilted magne-tization continues to precess about Bo in the transverse plane, intercepting a much larger area and thereby generating a current in the receiver coil, which then becomes the MR signal . The matching of the oscillation frequency of B1 to the Larmor frequency can also be viewed as a resonance phenomenon, which allows trans-fer of energy from the RF pulse to the magnetic moments. As soon as the RF pulse is switched off, the system relaxes and starts to return the energy to its original equilibrium state . System relaxation occurs via two mechanisms, which have different time constants: the longitudinal relaxation time (T1) and the transverse relax-ation time (T2) . Longitudinal relaxation is a mechanism whereby the magnetization trans-fers energy back to the environment, the so-called lattice, and represents the regrowth of the magnetization M along the axis of the external magnetic field Bo . The transverse or spin-spin relaxation (T2) is associated with the loss of sig-nal, which is manifested as spins getting out of phase in the transverse plane. This so-called dephasing phenomenon results from the exchange of energy between the spins. Both relaxation processes (recovery and decay) have been modeled by exponential equations and are depicted in Figure 4. It can be seen that the loss of magnetization in the transverse plane is

Mo ------------

M

Longitudinal (T1 recovery)

1

Transverse (T2 decay)

Time

Figure 4 The transverse T2 decay and the longitudi-nal T1 recovery of the magnetization depicted pursuant to a 90° RF pulse. Note that the transverse decay is much faster than the longitudinal recovery (T2 < TI).

quicker than the longitudinal regrowth, leading to T2 < T1 . Apart from spin-spin energy exchange, magnetic field inhomogeneities can also contribute to the loss of phase coherence, making the effective transverse decay time T2* shorter than T2 :

1/T2* = 1/T2 + 1/ T2' (Equation 2)

Here, T2' represents the contribution of magnetic field inhomogeneities .

In standard MR applications, the contrast in a neuroanatomic image (i .e ., the segmented light and dark areas) is provided by the proton density and tissue-specific T1 and T2 mecha-nisms. In fMRI applications, the T2 effects are the primary contributors whereby changes in the local field uniformity, as determined by the blood oxygenation level, provide the signal difference (contrast) between activated and nonactivated states . The T1 effects only play an indirect role mediated by the increase in blood flow during activation . These issues will be discussed in greater detail in the next subsection .

To summarize, the contrast in MR images is due to the tissue-specific response of nuclei that have a nonzero nuclear spin, such as hydro-gen (1H), which is usually the nucleus of choice due to its natural abundance, larger gyromag-netic ratio than other nuclei, and high concen-tration in vivo . To observe the MR phenomenon, an externally applied magnetic field is essential to orient the spins in a common direction. For the measurement process, it is required that the orientation of the spins be tilted away from this external field. A much weaker RF pulse, B1, is applied in a direction orthogonal to Bo in order to tilt the spins away from Bo . Once in the trans-verse plane, the spins continue to precess about Bo after the removal of B1, thus inducing a net voltage in a receiver coil placed anywhere within this transverse plane. The removal of the B1 field also marks the beginning of the relaxation process, leading to the loss of transverse mag-netization due to a combination of longitudinal regrowth (T1) and transverse dephasing (T2) . The tissue-specific T1 and T2 relaxation rates, combined with the concentration of the probing nucleus, impart the contrast observed in MR images .

Magnetic Resonance Imaging

The essence of MRI is the spatial localization of the MR signal . This is achieved using the field dependence of the MR phenomenon, whereby

243


A B C

Figure 5 In the presence of magnetic field gradients, the strength of the external magnetic field experienced by a spin depends on its location as illustrated for gra-dients in A, the x-direction ; B, the y-direction ; and C, the z-direction. Note that irrespective of the direction of the magnetic field gradient, the magnetic field is in the direc-tion of the externally applied field Bo .

the signal is spatially localized to individual vol-ume elements (voxels) within the field of view (FOV) or area of interest. Let us now consider how this might occur. If we assume that each voxel is associated with a slightly different precession frequency, and if as many signal values as there are voxels could be recorded (although the recorded signals would be from the entire volume and not from individual voxels), then it would be possible to determine the signal associated with each voxel. The precession frequency of each voxel is altered through the application of mag-netic field gradients (i .e ., spatial variations of an external magnetic field) . As depicted in Figure 5, the net magnetic field is different at each voxel depending on the direction of the gradient . The signal is recorded following a delay (echo time [TED subsequent to the application of the RF pulse in the presence of a gradient applied along one of the directions, commonly referred to as the readout or frequency-encoding direction. In the direction perpendicular to the readout, the so-called phase encoding direction, a unique fre-quency is assigned to voxels by applying a gradient, this time prior to recording the signal . The time elapsed between two consecutive RF pulses is known as the repetition time (TR) . One can think of voxel localization as analogous to the organization of an organ or piano keyboard, where each key is associated with a unique fre-quency (a musical note). Likewise, a different pre-cessional frequency is associated with each voxel.

In summary, the spatial localization of the MR signal is accomplished by using pulse sequences, which prescribe the currents to be applied for the Bl field, the slice selection, read-out, and phase encoding gradients (Fig. 6) . When a gradient is present, the precession frequency of the spins is different at each voxel and thus spatial localization becomes possible . The num-ber of applied gradients equals the number of

RF

Slice Selection Direction

Frequency Encoding Direction

Phase Encoding Direction

Signal

#

E '1'g ~-

T

Figure 6 The MR acquisition process is illustrated with a simple gradient-recalled echo pulse sequence . After slice selection, which is accomplished by applying an RF pulse during the presence of a gradient, the posi-tion along one of the axes of the imaging plane is encoded by the PE gradient (GpE). The position along the per-pendicular axis is encoded while recording the signal in the presence of the readout gradient (G~Qed..d'

voxels in one of the directions of the imaging plane, known as the phase encoding direction. For each phase encoding gradient, the signal is then recorded multiple times in the presence of a perpendicularly applied gradient in the read-out direction. With conventional MRI techniques, when the magnetization is tilted into the trans-verse plane prior to each phase encoding, the acquisition time is directly proportional to the number of phase encodings. In fMRI applica-tions, where temporal resolution (i .e ., the neces-sity for rapid and repeated sampling of the hemodynamic response of the brain) is an impor-tant consideration, the choice of the pulse sequence becomes crucial.

As we will emphasize in subsequent sec-tions, whether one is detecting hydrogen atoms from water molecules (MRI) or biochemical com-pounds in the body (MRS), spatial localization is achieved through magnetic field gradients . The difference between the two applications is in the timing of the applied gradients . MRI uses gradients during the readout or acquisition por-tion of the pulse sequence to enable spatial local-ization, whereas in MRS, no gradients are applied during acquisition to preserve frequency information (i .e ., chemical-shift resolution) .

Different MR1 Techniques

MRI techniques, alternatively referred to as pulse sequences, are distinguished primarily by their transverse magnetization refocusing

244


approach (i .e ., restoration of phase coherence by application of magnetic gradients or RF pulses), their k-space coverage design, and echo-shifting properties . In any discussion of MRI, however, it is important to consider the concept of k-space . K-space is a mathematical construct used to characterize the spatial properties of an image in two dimensions . It is derived by regrouping the variables in the mathematical for-mula for the Fourier transformation (Bradley et al, 1999) and where the x-y coordinates are expressed in units of phase.' The underlying premise of Fourier analysis is that any complex signal can be broken down into their respective component frequencies (sine waves) . In MRI, since sine waves are used to characterize an object in the x-y plane, the axes have units of spa-tial frequency (cycles per millimeter) rather than units of temporal frequency (cycles per second) . In the domain of spatial frequency rep-resentation, low spatial frequencies describe the bulk of the image, whereas high spatial fre-quencies fill in the details . Figure 7 depicts a rec-tangular grid in k-space, which is a prerequisite for 2D Fourier transformation (Bracewell, 1986) . The signal recorded at each k-space point has information pertaining to the entire image . The center of k-space gives information about the average gray scale level of the image . Points that are off from the center by one provide infor-mation about very large structures that can be modeled by one sinusoid across the image. Sim-ilarly, k-space points that are off from the cen-ter by two are for structures that can be modeled by two periods of the sinusoid across the image . To be accurate, no structure can be modeled by just one sinusoid ; thus, a weighted combination of many sinusoids is required . In terms of the imaging pulse sequence, with each application of the phase encoding gradient, the position in k-space moves up by one line . At a particular k Y, loca-tion, the signal along k' is recorded in the presence of the readout gradient . The distance traversed in k-space is, in fact, angular distance, which is equivalent to the phase that the spin acquires while precessing about Bo. Thus, depending on the phase accumulated, it is possible to have trajec-tories in k-space that do not conform to a rec-tangular grid, as is the case for spiral imaging.

'Note that in k-space, although one of the axes is referred to as phase encoding and the other as frequency encoding, both axes represent units of phase (e .g ., as spins move around at a certain speed [angular frequency], they traverse a certain angular distance [phase]) .

phase encodin g

kx

2 . . o . . . . .

-1 -2

. . . . . . .

.

. . .

. a .

. . . .- .

ky

U- yxuxpt 2n

Frequency encoding

2n

Figure 7 The MR signal is recorded in k-space such that movement along the k and k axes is accomplished through the application of magnetic field gradients. The phase accrued during the application of a phase encod-ing gradient determines the k location, assuming phase encoding is in the y-direction. In the readout direction, movement along k. originates from the phase accumulated in between consecutive sampling points .

In such cases, the acquisition points are regrid-ded onto rectangular coordinates (Jackson et al, 1991) to benefit from the convenience of the Fourier transform.

In this section, to orient the reader to pulse sequence terminology and protocols, examples of some common imaging techniques based on rectangular k-space trajectories are briefly discussed and compared with respect to func-tional imaging applications . They include spin echo (SE), gradient-recalled echo (GRE), echo planar imaging (EPI), and principles of echo-shifting using a train of observations (PRESTO) . For auditory studies, an additional technical variant to enhance signal recovery during subcortical processing has been used, so-called cardiac gating . Initially applied in the mid-1960s by investigators in nuclear medicine to study images of the beating (moving) heart in vivo (Pickens, 1988), the use of cardiac gat-ing to the field of functional imaging has expanded research opportunities for auditory investigations (Guimaraes et al, 1998). Thus, as we proceed through the next sections, it should become evident that the MR physicists and engineers have been very creative in devel-oping the pulse sequences and the analysis techniques necessary to enhance various appli-cations, taking into consideration the limita-tions imposed by available hardware and complexities involved in studying the human brain in vivo .

245


90,

SPIN ECHO w,

GRADIENT ECHO (No 180°,)

4 gradient invert

Figure 8 The GRE and SE pulse sequences can be dif-ferentiated based on their method of refocusing . The SE pulse sequence employs a 180° pulse for refocusing, whereas the GRE pulse sequence employs gradients to reverse the movement of the spins in order to bring them into phase with each other.

CONVENTIONAL TECHNIQUES

Spin Echo and Gradient-Recalled Echo

In conventional techniques, a single line of k-space is acquired for each application of the RF pulse. Thus, during each TR, one readout line is recorded . At times coincident to TE, there is complete refocusing along the readout and all recorded points fall along the ky axis of Figure 7. The beginning and end of a k-space line cor-respond in turn to negative and positive accru-als of phase, respectively. Movement in the kI direction (i .e ., recording of successive horizontal lines) is accomplished through the application of phase encoding gradients with gradient mag-nitudes augmented for successive TRs.

The SE and GRE techniques constitute dif-ferent examples of conventional methods. In SE imaging, nuclear spins that are dephased in the transverse plane are refocused by a 180° pulse (Hahn, 1950), which flips the spins about the axis along which it is applied, as depicted in Figure 8. Identical timing between 90° to 180° and 180° to TE periods ensures that the spins travel the same angular distance during these two periods, enabling complete refocusing. In GRE, a single k-space line is acquired per RF pulse and no 180° pulses are applied. Consequently, both the TR and the overall acquisition times are con-siderably shorter due to the partial nutation or tilt into the transverse plane and rapid refocus-ing (Frahm et al, 1986). Initially, spins precess in the reverse direction in GRE under the influ-ence of a negative gradient and are subsequently refocused during the readout period (see Fig. 8) . This is in contrast to SE, in which spins main-tain the same precession direction. Although GRE refocuses only the effects of externally

applied gradients, SE can also refocus the effects of static field inhomogeneities, which act similarly to low-amplitude positive gradients. As a result, with SE, T2* = T2, since T2' inhomogeneity effects are compensated for, whereas in GRE, T2* has contributions from both T2 and T2' (Equa-tion 2) . In /MRI applications, temporal resolution considerations make the conventional imple-mentation of the SE pulse sequence unfavor-able. However, variants of SE, alternately referred to as rapid acquisition with relaxation enhancement (RARE) (Hennig et al, 1986) or fast spin echo, which phase encode multiple echoes per RF pulse, have been used in fMRI (Constable et al, 1994). The SE pulse sequence can also be used in fMRI when incorporated into faster imaging techniques such as EPI.

ECHO PLANAR IMAGING

n single-shot EPI, the entire k-space is acquired after a single application of the RF

pulse (Mansfield, 1977), considerably curtail-ing image acquisition time . To complete the acquisition prior to signal decay, fewer phase encodings and readout points are recorded at a higher sampling rate. Consequently, spatial res-olution is worse (2-4 mm in EPI vs ~1 mm for conventional imaging), and the images are nois-ier due to the increased bandwidth resulting from higher sampling rates. Nevertheless, despite the artifacts caused by T2* decay (Farzaneh et al, 1990), the depiction of the brain is of sufficient quality to make EPI one of the more coveted sequences, primarily due to the increased temporal resolution (Bandettini et al, 1992 ; Blamire et al, 1992 ; Kwong et al, 1992b; Turner et al, 1993 ; Menon et al, 1995). Whereas a single GRE slice can be acquired in 3 to 4 sec, multiple EPI acquisitions can be completed in 1 sec. Different variations of EPI include the SE-EPI and the GRE-EPI, named after the method of preparation analogous to their con-ventional counterparts. Nonetheless, data acqui-sition is accomplished in a single shot . Multishot versions of EPI also exist whereby the image is acquired during the course of several RF pulses (McKinnon, 1993). The gain in spatial resolution provided with these sequences, however, is at the expense of a reduction in temporal resolution .

PRINCIPLES OF ECHO SHIFTING WITH A TRAIN OF OBSERVATIONS

F unctional MRI has also been carried out in conjunction with three-dimensional (3D)

246

pulse sequences such as PRESTO, which improves the temporal resolution of the con-ventional GRE sequences while maintaining T2" sensitivity, specifically by introducing the echo-time shifting (ETS) concept (Moonen et al, 1992) . With PRESTO, multiple k-space lines are acquired within a single repetition period (van Gelderen et al, 1995), and TE is extended using the principle of ETS. Thus, PRESTO can be viewed as a type of multishot EPI method (Moonen and van Gelderen, 1997) . With ETS, the refocusing of the magnetization nutated into the transverse plane is deferred to a subsequent

repetition cycle leading to TE > TR. In the great majority of sequences where TE < TR, extension of TE to maximize T2* sensitivity inevitably leads to an extension of TR as well, causing an increase in the overall acquisition time and loss of temporal resolution . With ETS, T2* sensitiv-ity is obtained without a sacrifice in acquisition time . Volumetric acquisitions are rarely possi-ble in conjunction with fMRI due to lengthy acquisition times, yet they provide a special benefit in terms of minimizing arterial inflow

effects (Duyn et al, 1994a) .2 The PRESTO tech-nique refocuses all residual gradients within each repetition cycle and regulates the timing of echoes through additional gradients to enable the use of volumetric acquisitions in conjunction with f MRI .

MAGNETIC RESONANCE SPECTROSCOPY

W hereas in MRI, imaging hydrogen atoms from water molecules is of primary interest, MRS aims to detect metabolites in the mil-limolar range such as N-acetylaspartate (NAA), choline (Cho), creatine (Cr), myo-inositol (mI) and lactate, which can only be observed after

water suppression . Being able to quantify meta-bolic information from MR technology is anal-ogous to performing a noninvasive in vivo biopsy of body or brain tissue . The term neu-rospectroscopy, in particular, is synonymous

with proton (1H) MRS and relates to the diag-nosis made from MRS in evaluating the human

brain . In this context, Danielsen and Ross (1999) consider two main aims for the clinical

application of neurospectroscopy :

2 As a result of collaborations with Drs. Peter van Gelderen and Chrit T.W . Moonen, the PRESTO technique has been used in MR I studies at the Albany Medical Center (Cacace et al, 1996, 1999 ; Cacace, 1997, 1999).


1 . Clinically diagnostic in vivo biochemistry in patients (i .e ., the ability to make relevant biochemical measurements directly, thereby influencing medical management either by

diagnosis or by exclusion) ; 2. Defining new features in the pathophysiol-

ogy of disease that will lead to : a. New therapies based on MRS results, b. Disease progression or therapeutic

response monitoring by use of MRS, c. Refining disease prognosis by noninva-

sive MRS metabolic measurements in patients .

In the normal human brain, the 1H MR spectrum can be understood by the concept that each metabolite has a unique signature, and when added to other major metabolites, the combination results in a complex spectrum of overlapping peaks . Interpretation of MR spec-tra involves determining the presence of each cerebral metabolite and noting whether specific peaks are elevated, reduced, or unchanged in relation to a reference, usually Cr . In the eval-

uation of individuals within a clinical context, NAA is considered a marker of neuronal meta-bolic activity, Cho is considered a marker of

overall cellular density, Cr is considered a marker of cellular energetics, and ml is considered a marker of glial cell compartments . In MRI,

NAA

Cr

Cho M1 ~, Cr

I ` 1,J 'J n~1

~I ~ I _I Jv

NAA A Glu/Gln

y n n � N1

4 3 .5 3 2 .5 2 Figure 9 Magnetic resonance spectroscopy in white matter of the centrum semiovale showing metabolites (e.g., N-acetylaspartate [NAA], choline [Cho], creatine [Cr], glutamate/glutamine [Glu/Gln], myo-inositol [ml]). In this example, the short-TE stimulated echo acquisition mode (STEAM) pulse sequence was used .

247


Table 1

Metabolite Increased (Normal Cerebral Concentration)

Lactate (Lac) (-1 mmol/L ; not visible)

(Often) Hypoxia, anoxia, near drowning, stroke, hypoventilation, Canavan's, Alexander's, Hydrocephalus

N-acetylaspartate (NAA) (Rarely) (5, 10, or 15 mmol/L) Canavan's

Glutamate (Glu) and/or Glutamine (Gln) (Glu = ? 10 mmol/L ; Gln = ? 5 mmol/L)

Chronic hepatic encephalopathy (HE), acute HE, hypoxia, near drowning

myo-Inositol (ml) (5 mmol/L)

Total creatine (Cr) (8 mmol/L)

Glucose (Glc)

Neonate, Alzheimer's disease, diabetes mellitus, recovered hypoxia, hyperosmolar states

Trauma, hyperosmolar increasing with age

Diabetes mellitis, parenteral feeding

Decreased

(Unknown)

(Often) Developmental delay, infancy, hypoxia, anoxia, ischemia, intracranial hemorrhage, herpes II encephalitis, near drowning, hydrocephalus, Alexander's, epilepsy, neoplasm, multiple sclerosis, stroke, diabetes mellitus, closed head trauma

(Unknown) Possible Alzheimer's disease

Chronic HE, stroke, tumor

Hypoxia, stroke, tumor, infant

Not detectable (-1 mmol/L) HE

Choline (Cho) (1 .5 mmol/L)

Trauma, diabetes, white vs gray, neonates, after liver transplant, tumor chronic hypoxia, hyperosmolar, elderly normal (?Alzheimer's disease)

Asymptomatic liver disease, HE, stroke, nonspecific dementias

Acetoacetate, acetone, Detectable in specific ethanol, aomatic aminoacids, settings xenobiotics (propanediol, mannitol)

From Danielsen and Ross (1999) .

whereas signal intensities are displayed in the form of an image, in MRS, signal intensities are more commonly expressed as a function of frequency (a spectrum) (Cousins, 1995).

An MRS spectrum represents a series of peaks along the x-axis labeled in parts per mil-lion (ppm) or Hertz (Hz) to refer to the preces-sion frequency. For a proton 1H spectrum acquired at 1.5 T (63.5 MHz), 1 ppm corresponds to 63 .5 MHz. The ppm scale provides an easy comparison of data obtained at different field strengths since, irrespective of the strength of the magnetic field, a substance with a reso-nance at a chemical shift of 2.0 ppm still res-onates at 2.0 ppm at 0.5, 1.5, or 4.0 T. Atypical 1H spectrum is displayed in Figure 9, obtained

None

from white matter of the brain. Spectroscopic information can also be displayed in the form of an image, with each image representing a dif-ferent metabolite . In this context, the post-processing involves integrating the area under the relevant resonance peak for each voxel to derive metabolite-specific information. In this format, different colors are used to represent the spatial distribution of a particular metabolite over regions of interest (ROI). When many spec-tra are simultaneously obtained, rendering such information into an image format is a succinct method of presentation .

Magnetic resonance spectroscopy is becom-ing an important adjunct to standard MRI in large part because diseased and pathologic

248


states have identifiable neurobiochemistries (see Table 1 for differential diagnostic uses) . As a result, MRS is becoming routinely used for the clinical evaluation of primary CNS lesions, the evaluation of neurometabolic disorders, mitochondrial disorders (such as mitochondrial encephalomyopathic, lactic acidosis, and stroke-like episodes ; MELAS), the localization/later-alization of epileptic foci, acquired immunodeficiency syndrome (AIDS) related brain disorders, Alzheimer's disease evalua-tion, brain injury from hypoxia, near-drown-ing episodes, evaluation of postsurgical and postradiation tumor beds, etc . (Cousins, 1995 ; Frahm and Hanefeld, 1997 ; Danielsen and Ross, 1999; Wilichowski et al, 1999) .

Apart from its clinical value in assessing organic brain disorders, and given the robust-ness of these measures over time (Brooks et al, 1999 ; Hoshino et al, 1999), MRS has potential to play a role in understanding the neurobio-chemical correlates of human development . Some important aspects of this area include providing a better understanding of individual differences in cognition and intelligence in nor-mal brains (Hashimoto et al, 1995 ; Jung et al, 1999a, b) and in providing basic insights into the neurobiology of various learning disabilities

such as dyslexia (Richardson et al, 1997 ; Rae et al, 1998 ; Richards et al, 1999) . Moreover, stud-ies characterizing neurobiochemical properties of cortical auditory structures of normal indi-viduals and in individuals with sudden idio-pathic hearing loss (Richards et al, 1997 ; Syka et al, 1998) have also been initiated .

FUNCTIONAL MAGNETIC RESONANCE IMAGING

A historical review of functional imaging (Raichle, 1998) reveals that changes in cir-

culation related to mental activity were reported over 100 years ago in the late 19th century. Available techniques used for functional brain imaging are based on the elementary principle that neuronal activity requires energy and that increased energy demands, as reflected by local increases in blood flow and metabolism, can be visualized (see Magistretti and Pellerin [19991 for an overview of cerebral energy metabolism). However, it has been only with the advent of positron emission tomography (PET) in the 1970s that in vivo functional imaging experi-ments on humans have become possible (Phelps et al, 1981 ; Raichle et al, 1983). PET measure-

ments reveal that changes in blood flow and glucose metabolism far exceed that of oxygen metabolism during brain activation (Fox and Raichle, 1986 ; Fox et al, 1988), leading to increased arteriovenous oxygenation levels . The increased oxygenation levels also form the basis of the blood oxygen level dependent (BOLD) contrast, commonly employed in /MRI studies . The BOLD effect was initially demonstrated in the laboratory setting on rats (Ogawa et al, 1990) and was subsequently applied to in vivo functional activation studies of the human brain (Kwong et al, 1992a; Ogawa et al, 1992) . Other techniques based on external contrast agents (e.g., gadolinium diethylene-trimine-pentaacetic acid) have also been used in sensory activation studies to map the visual cortex by measuring changes in regional cerebral blood volume (Bel-liveau et al, 1991) . However, in BOLD imaging, it is the paramagnetic deoxyhemoglobin of blood that serves as an intrinsic or endogenous con-trast agent. Here, blood acts like an external T2 altering relaxation agent with the degree of paramagnetism determined by the oxygenation state imposed by the activation (Thulborn et al, 1982) . In human investigations, fMRI studies based on BOLD contrast are preferred over those based on the use of exogenous contrast agents because the former are noninvasive and are not limited by temporal considerations . The rapid passage of the external contrast agent through the brain restricts the number of repet-itive on/off stimulus cycles in a blocked-stimu-lus paradigm or the number of individual stimulus presentations in an event-related fMRI study. Note that blood flow can also serve as a T1 modulation mechanism due to the exchange of H2O across the blood-brain barrier during the traversal of blood through the capillaries (Detre et al, 1992 ; Williams et al, 1992) :

1/ T1' = 1/T1 + f/)` (Equation 3)

where f refers to flow, X to the blood-brain par-tition coefficient, and T1' to the modulated lon-gitudinal relaxation due to flow. However, note that for EPI sequences, T1 effects cannot play a substantial role, as the TRs are on the order of one second or more, leading to considerable longitudinal recovery of magnetization. In dis-cussions pertaining to f MRI, BOLD effects refer to the prolongation of T2* due to flow-based oxygenation level increases, as opposed to flow effects, which cover the contribution of incom-ing fresh spins and T1 changes.

249


Blocked-Stimulus Presentation and Postprocessing

The goal of functional imaging is to localize within the brain activated regions that respond to a stimulus by developing techniques that are sensitive to external manifestations of neuronal changes. Briefly, fMRI studies begin with a local-izing sequence, which is used to prescribe the slice or slices of interest in imaging applications and the ROI in spectroscopic applications . To date, by far the most common technique used for fMRI studies is based on the blocked-stimulus paradigm . With this technique, functional acqui-sitions are conducted in the presence of an inter-mittent external stimulus, where blocks of stimuli that represent activation or on states are alternately interleaved with rest or off states, where no stimulus is presented. For example, a motor paradigm can consist of alternating cycles of a finger opposition-tapping task with rest conditions . Finger tapping can be replaced by an illuminated reversing checkerboard pattern for a visual paradigm ; by tones, speech, or music pre-sentation for auditory paradigms; or by the pre-sentation of special words or sentence structures for language or working-memory paradigms . Additional complexities can be imposed by adding various attentional demands or memory loads to underlying experiments, making the technique well suited for studying modality-specific and supramodal processes. Alterna-tively, stimuli may be generated internally, as in the case of tinnitus or other phantom per-ceptual events . Irrespective of the nature of stimulation, it is required that data be sorted with respect to the presentation paradigm and analyzed in order to isolate the activation response . Upon completion of the stimulus pre-sentation and acquisition process, the functional data are registered to compensate for in-plane head motion and statistical tests are conducted to determine whether the data collected during the activation states are significantly different from those collected during the rest states . Although the type of statistical testing has been a topic of debate, various parametric and non-parametric techniques such as the t-test, z-score, correlation analyses, and Kolmogorov-Smirnov tests have all been applied (Bandettini et al, 1993 ; Cohen and Bookheimer, 1994 ; Friston et al, 1995a; Xiong et al, 1996 ; Kuppusamy et al, 1997 ; Rabe-Hesketh et al, 1997 ; Lange, 1999). For example, when the t-test is used in the blocked-stimulus paradigm, the average of the signal intensities recorded during the rest or off

state is subtracted from the average signal inten-sity during the activation or on state and divided by the pooled variance, which is a measure of the extent of signal fluctuations . Results of statis-tical testing are then thresholded, color coded, coregistered, and subsequently superimposed onto the anatomic reference images to allow for interpretation and display. Figure 10 shows sev-eral fMRI activation patterns in striate cortex based on a blocked-stimulus paradigm . The pur-pose of this graph is to illustrate the effects of magnetic field strength on the detection of the BOLD fMRI response .

Functional activation can be further assessed by plotting the signal intensities over time (pixel values from consecutive images at a given activated voxel) to visualize the pattern of the time series during the on and off sequences. In cases where brain activation may not neces-sarily follow precise on-off patterns, more sophis-ticated techniques such as principal component analysis (Friston et al, 1995b), fuzzy clustering algorithms (Baumgartner et al, 1997), and inde-pendent component analysis may be more appro-priate (Makeig et al, 1997).

Event-related tMRI

The alternative to the blocked-stimulus par-adigm is the recent development of event-related fMRI (ER-fMRI), which can detect activation without the need for repetitive on-off cycles, thus allowing for more transient aspects of the hemodynamic response to be studied (Rosen et al, 1998 ; Buckner, 1998 ; Buckner and Braver, 1999). ER-fMRI refers to the detection of corti-cal activation from single or selectively aver-aged trials of a particular task (Buckner et al, 1996; Dale and Buckner, 1997) and has already been applied with success within the auditory modality (Hickok et al, 1997 ; Robson et a1,1998; Belin et al, 1999). There are several aspects of ER-fMRI that make it advantageous and in some instances preferable to the use of the blocked-stimulus paradigm, particularly in appli-cations studying various cognitive processes. These advantages include the ability to ran-domize trial presentations, the flexibility of which allows for post hoc sorting of the stimu-lus response based on response time, type, or accuracy (Buckner and Braver, 1999), and the opportunity for testing the differences in the onset time of neural activity evoked by different trial types (D'Esposito et al, 1999). Other areas that are also being explored include the use of perfusion-based ER-fMRI (Liu and Gao, 1999),

250


microvasculature to appreciate the strengths and limitations of these techniques .

CAPILLARY CHARACTERIZATION

T he ultimate goal of any functional study is to identify the brain parenchymal contribution where neuronal activity resides. Thus, the capillaries and their surrounding gradients are of primary importance in f MRI. However, imag-ing of parenchyma as distinct from capillaries is not an attainable goal, given that the capil-lary diameter can be less than 5 [,m (Fulton, 1955) and the MR pixel resolution is at best -1 mm. Depending on the paradigm, the reso-lution can be even worse in f MRI by a factor of 3, due to temporal resolution trade-offs . Thus, the challenge is to restrict BOLD and flow-related effects to the capillary level. The so-called "brain versus vein" effects, referring to the cap-illary versus flow contribution, have been an ongoing concern within the fMRI community (Duyn et al, 1994b; Frahm et al, 1994). Arterial contributions are not a part of this discussion as they are an unlikely source of activation . Arter-ies are 95 percent oxygenated and do not allow exchange across the blood-brain barrier (Fulton, 1955); therefore, oxygenation level does not change with activation, making them immune to T2* and T1 effects, respectively. Neither are arterial inflow effects a consideration for long TR sequences, such as EPI, where there is total replenishment of flow within the slice. Arterial inflow effects become a concern with conven-tional GRE sequences depending on the TR, slice thickness, and magnitude of flow/velocity changes . Even gradient-recalled sequences become immune to arterial inflow effects when the nature of the analysis is 3D, as is the case for the PRESTO technique (van Gelderen et al, 1995).

The primary concern then becomes to dif-ferentiate the capillary contribution from those of larger veins. It has been shown that at the most commonly used field strength (1 .5 T), the main contributors to activation are the larger veins. The microvasculature gains prominence only at the higher magnetic field strengths, that is, 4T (Menon et al, 1993). It is of interest to note that at higher field strengths, preceding the increase in signal enhancement, a short-lived sig-nal decrease is observed (Ernst and Hennig, 1994 ; Menon et al, 1995). This initial dip, initially discovered by optical studies, is claimed to be more localized to the capillary activation sites (Malonek and Grinvald, 1996) and can also serve

as a mechanism by which to differentiate cap-illaries from veins . However, at most clinical centers, higher magnetic field strengths are not available, making the choice of pulse sequence and the accompanying parameters important considerations (Kennan et al, 1994 ; Weisskoff et al, 1994) . Another method of localizing capil-lary regions is by using the delay in the activa-tion response . The response to activation is not instantaneous, and shorter delays are associated with the capillaries (4-8 sec) and longer delays with the larger veins (8-14 sec) (Lee et al, 1995) . For example, one can sensitize GRE sequences to the microvasculature by lengthening the TE. The T2* weighting of GRE sequences is affected by the characteristic diffusion time (T,), which, in turn, is proportional to the square of the ves-sel radius . Vessels with T < TE exhibit qua-dratic decay, whereas those with r > TE exhibit linear decay (Yablonsky and Haacke, 1994) . Thus, at short TEs, the capillary contribution decays more readily. By extending TE, a qua-dratic decay can be imposed on large vessels as well, making the relative capillary contribution more pronounced. In gradient-recalled acquisi-tions, ETS, as used in PRESTO, is one way of extending TE to increase contributions from the microvasculature .

IMAGING AUDITORY PATHWAYS FROM COCHLEAR NUCLEUS TO CORTEX: USE OF EXOGENOUS STIMULI

S tudying auditory system activation using fMRI is an area that is in early stages of development. Melcher et al (1999a) have recently provided an overview of studies and discussion of pertinent issues that have arisen in studying this modality. Herein, our main intent is to illus-trate by various examples the range of potential applications and methodology that may enhance signal recovery and help to formulate and test various hypotheses. Then, we will provide a con-temporary update on imaging tinnitus-related activity with functional imaging methods.

Analysis of Subcortical Processing

Until recently, NRI studies in the auditory modality have focused on activation patterns in auditory cortex and related areas . However, the work of Guimaraes et al (1998) demonstrated that major nuclei in subcortical auditory path-ways can also be mapped with fMRI at a field strength of 1.5 T. The key to these initial stud-ies was the application of cardiac gating, which

252


Whereas much of the combined physiologic and neuroanatomic information describing cen-tral auditory system organization has been derived from experiments in macaque and rhe-

sus monkeys, there are obvious parallels in

human studies, although the full extent of these associations has yet to be realized . Contempo-rary neuroanatomic staining methods and var-

ious imaging-related activation studies suggest that the human auditory cortex can be subdi-

vided into at least eight different putative regions (Rivier and Clarke, 1997) . Other proposed schemata can also be taken into consideration when attempts are made to delineate various cor-tical areas (Rademacher et al, 1993). Thus, with this information as a background, we now con-sider some examples of sensory information pro-cessing in human auditory cortical areas being elucidated by f MRI.

Tonotopic Frequency Organization

Having a spatially sensitive method to inves-tigate the tonotopic organization of auditory cortex allows one to compare and contrast human data with single-unit mapping studies obtained from experimental animals. The fMRI work of Talavage et al (1996, 1997 ; also cited in Melcher et al, 1999a) provides direct evidence for multi-

ple, tonotopically organized areas on the medial

and lateral aspects of Heschl's gyrus in humans . Bilecen et al (1998) studied frequency repre-sentation in auditory cortex using binaurally presented low- (500-Hz) and high-frequency (4000-Hz) pure-tone stimuli. The low- and high-frequency tonal stimuli showed properties of

different frequency organization in the ML, frontal-occipital, and cranio-caudal orientations in both hemispheres. In the majority of indi-viduals, the 4000-Hz tonal stimulus had a more frontal and medial orientation than the 500-Hz tone . Strainer et al (1997) evaluated normal-hearing subjects and found that pure tones

activated medial and lateral aspects of the trans-verse temporal gyrus, whereas complex stimuli produced activation in association areas. Whereas microelectrode studies have been the gold standard for mapping tonotopic areas in experimental animals, they have rarely been performed on humans for obvious reasons. How-ever, Howard et al (1996) reported such a study in an individual with normal hearing being mon-itored for medically intractable epilepsy. When recordings were made from a multicontact depth microprobe steoreotaxically placed within the long axis of Heschl's gyrus on the right side, a

tonotopic frequency representation was found with higher best frequencies localized at more posterior-medial recording sites and lower fre-quencies along more AL sites. Whereas this type

of data is of particular importance in validating

the concept of tonotopicity in humans, the inher-ent limitations of using just a single-depth probe

study cannot address the issue of multiple tono-

topic fields . Thus, the impressive work demon-strating multiple tonotopic fields in human

studies by f MRI takes on additional importance

when compared to recent work in monkeys (e.g., Kaas et al, 1999 ; Romanski et al, 1999b) .

Intensity Coding

Of fundamental importance to auditory sig-nal processing is the area of intensity coding. In a parametric study using cardiac gating and the single-slice acquisition paradigm, Sigalovsky et al (1999) evaluated changes in stimulus intensity at the level of inferior colliculus . At this subcortical location, these authors showed a significant monotonic increase in fMRI activation with increases in stimulus level over a 40-dB range. In contrast, other experiments that have studied effects of stimulus intensity

on auditory cortical areas have produced more variable results (Millen et al, 1995 ; Strainer et

al, 1997 ; Mohr et al, 1999). For example, Millen

et al used pulsed pure-tone stimuli at 1000 Hz and 4000 Hz and continuous speech presented

at 20 and 50 dB SL to investigate auditory acti-

vation with fMRI . Whereas both pure tones

and speech stimuli activated areas within the superior temporal gyrus, neither the intensity

level nor the side of stimulus presentation pro-duced significant alterations in the number of pixels activated. Mohr et al (1999) studied stim-ulus intensity of binaurally presented mono-syllabic speech stimuli. Whereas intensity effects were reported and activation was localized to

planum temporale, Heschl's gyrus, and superior

temporal sulcus, large individual differences were observed . Jdncke et al (1998) used non-verbal pure-tone stimuli and various conso-nant-vowel-consonant speech tokens presented binaurally at three different levels (75, 85, or

95 dB SPL) to study auditory cortical activity. Although participants were not engaged in a dis-crimination task, they were required to detect

a particular target stimulus as a means to main-

tain vigilance. Their analysis showed greater activation in the left versus right hemisphere

in Brodman's area 22 for the verbal stimuli, particularly at higher stimulus levels (85 and

255

(2000) investigated the role that the human voice might play in activating auditory cortical areas . They found that voice-selective regions exist bilaterally in the area of the superior tem-poral sulcus . These authors also point out sim-ilar organizational features with the visual cortex, where face-selective regions have also been identified .

Because use of complex binaural stimuli may sequentially activate different areas of auditory cortex over time, the method used for the analysis of blocked fMRI data should be chosen carefully. Gaschler-Markefski et al (1997) demonstrate that both stationary and nonsta-tionary time series can result from different tasks in a blocked-stimulus paradigm, as shown in Figure 13 . Huckins et al (1998) used non-verbal tonal complexes and verbal stimuli pre-sented monaurally to the right ear to study cortical auditory responses with fMRI . The unique feature of this investigation was the use of a small (5-in) surface receiver coil placed over the left temporal lobe . Use of a surface coil, in contrast to a much larger whole-head (bird-cage) type receiver coil, was applied to maximize sensitivity and optimize signal-to-noise ratio during imaging. With this method, Huckins et al (1998) demonstrated highly localized responses in regions of the superior temporal lobe and good reproducibility within and between test sessions . However, no difference in activa-tion patterns was observed between the two classes of stimuli. Whereas the novel use of a sur-face receiver coil has the benefit of providing increased sensitivity and signal-to-noise ratio, their design only provides information about one hemisphere and therefore limits information on laterality of function .

EFFECTS OF ATTENTION

S everal studies have documented the influ-ence that attention may have on several auditory tasks. Pugh et al (1996) found that when greater demands were placed on infor-mation processing by having individuals perform dichotic versus binaural pitch or speech token discrimination tasks, primary centers on the superior temporal plane and other areas in pos-terior parietal and inferior frontal systems were engaged. Jancke et al (1999) studied the effects of attention on consonant vowel stimuli by hav-ing subjects listen passively (condition 1), actively attend to stimuli (condition 2), and detect a tar-get syllable (condition 3) . The authors evalu-ated the effects of attention, hemisphere, and


primary versus secondary auditory cortex . They

found a significant hemisphere and cortex effect but no interactions . The left hemisphere and primary auditory cortex showed the greatest

effects, documenting greater activation in audi-tory cortex during active versus passive listening .

Seidman et al (1998) used an auditory con-

tinuous performance task made difficult by

increasing working memory demands and inter-

ference . They found significantly poorer perfor-

mance on working memory versus a simple vigilance task . In contrast to a vigilance task,

working memory showed activation within a

complex network including the lateral and medial prefrontal cortex, precentral cortex,

mesial temporal lobe (insula and hippocampus), parietal-occipital cortex, cingulate, thalamus,

and superior colliculus . Interestingly, perfor-mance and degree of activation was associated

with IQ . Other work emphasizes that attention

can act locally during auditory cognitive pro-

cessing and that modulation of auditory and

visual sensory cortex attentional demands can be modality dependent (Woodruff et al, 1996) .

Larisch et al (1999) evaluated the effects of motivation on a dichotic listening task in a sam-ple of five participants . Individuals included were those classified as success oriented, based on responses to a compendium of questions of motivational behavior (Schmalt's Questionaire). Two tasks were compared: a competition task (high motivation) for monetary compensation and a calibration task (neutral motivation). Whereas all participants showed significant bilateral activation in superior temporal gyrus during auditory stimulation and in the right anterior cingulate, in the high motivation con-dition, three of five participants showed activa-tion in the cingulate and right prefrontal cortex . The authors attributed these activation pat-terns to nonperceptual motivational aspects of the experiment. Although the small sample lim-its the generalizability of results, the authors suggest that similar questionnaires be used when recruiting volunteers in fMRI experiments to help ensure homogeneity of results.

Yoshiura et al (1999) used /MRI to investi-gate brain activation patterns during auditory and visual oddball tasks by focusing on rare target stimulus detection and making compar-ison between target and nontarget stimuli . Whereas modality-specific activations were found (auditory tasks activated transverse temporal gyri and transverse temporal sulci bilaterally ; visual tasks activated occipital lobes and the occipitotemporal junctions bilaterally), a broad

257


overlapping range of nonspecific activation pat-terns was also noted for each task . The authors interpret their results as demonstrating that target stimuli in an oddball task represent activ-ity from a complex combination of multiple generators rather than from a single neural population . Opitz et al (1999) used surface-recorded event-related potentials and fMRI to study stimulus deviancy and target stimulus detection within an oddball paradigm to evalu-ate neuroanatomic constraints for dipole mod-eling . In instances of dipole modeling of surface-recorded EEG or MEG signals, assump-tions or constraints regarding dipole number, ori-entation, and locations are necessary in order to find unique inverse solutions. Their results showed that when unattended deviants demon-strated a mismatch negativity by the event-related potential (100- to 160-msec range), fMRI activation was mapped to the superior-trans-verse temporal gyri bilaterally. When attended targets generated a mismatch negativity fol-lowed by P3oo response (320- to 280-msec range), f MRI activation was mapped bilaterally to supe-rior temporal gyri and neostriatum. Their results show that inverse solutions for dipole modeling of mismatch negativity localized activity near Heschl's gyrus during stimulus deviance pro-cessing, whereas the P3oo complex was localized to superior temporal gyrus during target detec-tion. Clearly, further work is needed in this area to solidify many of the important unanswered questions that remain.

Advancements in speech reading research may also benefit from an improved under-standing of the neural mechanisms involved in integrating auditory and visual stimuli. For example, when semantically congruent speech is presented simultaneously in the auditory and visual modalities versus conditions that pre-sent auditory or visual stimuli alone, responses were enhanced in modality-specific auditory and visual areas (Brodman's area 41, 42; visual area V5) (Calvert et al, 1999).

PRE- AND POSTNATAL DEVELOPMENTAL ASPECTS

F unctional MRI investigations are not limited to investigations on adults, and considerable information about human CNS development might be derived by studying children of all ages . Probably, the most serious constraints in apply-ing this methodology to infants and young

children are issues related to cooperation and movement-related artifacts, which would, no doubt, negate interpretable data in many instances. However, like any complex area, where substantial challenges exist, successful imple-mentation can also reap high rewards.

Hykin et al (1999) reported an intriguing study designed to evaluate the feasibility of mapping central auditory system function in utero using fMRI . Using auditory stimulation from a loudspeaker (recorded nursery rhyme) and presenting acoustic stimuli to the abdomen region of pregnant women, Hykin et al (1999) demonstrated that significant temporal lobe activation could be detected in the human fetus. The authors suggest that ".. .fetal fMRI has potential to become a powerful tool in investi-gating brain development." Children of school age appear to be a viable population to investi-gate, and Ulualp et al (1998) have obtained use-ful data in response to acoustic stimuli in individuals as young as 6 years of age (range 6-10 years) in normal-hearing healthy children . Based on passively listening to a prerecorded sample of ongoing speech, Ulualp et al report bilateral activation in superior and transverse temporal gyri, planum temporale, frontal, pari-etal, and cingulate regions of the brain. In a recent review of central auditory processing dis-orders in school-aged children, Cacace and McFarland (1998) note the paucity of research in the area of cortical language organization and representation in children (Bates et al, 1992; Duchowny et al, 1996 ; Hertz-Pannier et al, 1997; Nobre and Plunkett, 1997). Clearly, the lack of normative developmental data on the neural representation of language and lan-guage development complicates full under-standing of childhood language disorders. Indeed, efforts toward this end will be of considerable importance in helping to explain altered sensory information processing and other neurobehav-ioral limitations in children with various types of learning disabilities, such as dyslexia (Cacace et al, in press) . Furthermore, in specific groups of children at risk for neurodevelopmental delay, such as those with epilepsy or other types of brain lesions, fMRI evaluation may potentially play a role in studying cognition, in monitoring disease progression, and in the context of lan-guage localization, particularly if surgery becomes necessary (Bookheimer et al, 1999 ; Booth et al, 1999).

To date, significant advances have been made that highlight the importance of fMRI as applied to a wide range of sensory processing

258


experiments . The idea that using fMRI com-bined with various auditory psychoacoustic tasks will help understand the underlying basis of various auditory perceptual and cognitive/lin-guistic processes (Elliot, 1994) is beginning to materialize . However, the influence of imager-related background noise on task performance, the ability to precisely control stimulus pre-

sentation, and numerous safety concerns (Savoy et al, 1999) are some of the technical issues that have not yet been fully resolved . Furthermore, because fMRI is sensitive to all types of brain activity, factors such as motivation, memory, motor skills, novelty of the environment, men-tal imagery, and word rehearsal need to be rec-ognized and closely controlled. Attention to these concerns will help to reduce spurious results, eliminate alternative explanations, and minimize experimental confounds . Many of these impor-tant experimental design concerns have been addressed by others (see Aguirre and D'Espos-ito, 1999) .

Below, we provide an overview of how func-tional imaging is beginning to play a role in the study of phantom perceptions like tinnitus . This is another area, which is in the early stages of development, where creative and novel experi-mental designs can have an impact .

ical enigma . Because the area of functional imag-ing in studying tinnitus has not been limited to MRI, and since the literature at present is not voluminous, we will digress slightly here and dis-cuss both PET and fMRI studies . These approaches, reviewed recently by Cacace (1999), include :

1. Evaluation of glucose metabolism in indi-viduals with constant chronic tinnitus ;

2. Evaluation of individuals that can mod-ify/modulate a constant background tinni-tus by performing some type of overt behavior in another sensory, motor, or sen-sory motor modality, such as oral-facial maneuvers (jaw clinching) and change in eye position (eye gaze) from a neutral head-referenced condition;

3. Evaluation of individuals that can acti-vate/trigger their tinnitus (turn it on and off) by performing an overt behavior in another sensory, motor, or sensory motor modality, that is, change in eye position from a neu-tral head-referenced condition or cutaneous stimulation of the hand or fingertip region ;

4. Evaluation of stimulus-induced modifica-tion of lateralized tinnitus activity; and

5 . Evaluation of pharmacologic-induced tinnitus-related activity.

IMAGING TINNITUS-RELATED ACTIVITY ENDOGENOUSLY

GENERATED STIMULI

E ither implicitly or explicitly, most models and/or theories of tinnitus generation

assume altered spontaneous firing rates or pat-terns of neuronal activity within the central auditory system (Eggermont and Komiya, 1999). These hypotheses are driving a concerted research effort to validate the tinnitus percep-tion behaviorally and to localize and better understand the underlying neural substrate mediating such activity in the nervous system of experimental animals (e.g ., Jastreboff et al, 1988 ; Chen and Jastreboff, 1995 ; Kaltenbach et al, 1996, 1999 ; Eggermont and Kenmochi, 1998 ; Bauer et al, 1999). Moreover, at the end of the 20th century, innovative research designs using fMRI and PET have been applied to the task of studying tinnitus-related neural activity. From a basic science and clinical standpoint, func-tional imaging investigations in humans are of considerable importance to tinnitus research since they have potential value in providing fundamental insight into the underlying phys-iology and pathophysiology of this complex clin-

The idea that tinnitus could potentially be imaged was initially proposed by Sasaki et al (1980) and Kauer et al (1982), based on an ani-mal model of peripheral auditory deafferenta-tion using autoradiography with a glucose tracer, [14C]2-deoxyglucose. Although innovative and novel, there were interpretive issues that lim-ited the significance of this work (Meyerhoff et al, 1982). Similarly, with functional imaging applications in humans, the effects of hearing loss and the need to secure appropriate con-trols also remain as focal issues that can affect data interpretation. PET, which is based on tracer technology, is analogous to autoradiog-raphy (Lockwood et al, 1999b) . However, what distinguishes it from autoradiography is that both endogenous and/or exogenously generated activities can be explored in the human brain in vivo . Along these lines, Arnold et al (1996) stud-ied adult patients with chronic tinnitus and hearing loss using PET with [18F] deoxyglucose as a probe and compared them to controls with-out tinnitus or hearing loss . They assumed that if tinnitus was associated with synaptic hyper-activity, then the effects of such activity could be detected by increased glucose metabolism,

259


measurable by PET. In individuals with tinni-tus but not in controls, asymmetric metabolic hyperactivity (predominantly localized to the left hemisphere) was detected in the region of Heschl's gyrus in the primary auditory cortex. Additional work along these lines has been car-ried out by Oestreicher et al (1999) . They com-pared 25 tinnitus patients who were heterogeneous with respect to tinnitus and hear-ing loss to 20 controls who were audiometri-cally normal . The majority of tinnitus subjects (24 of 25, 96%) had significant sensorineural hearing loss . Of these, 15 of 25 (60%) had mild to severe sensorineural loss, 9 of 25 (36%) had profound unilateral deafness, and 1 had normal sensitivity. In this sample, 20 of 25 (80%) char-acterized their tinnitus as a "high-pitched ring-ing," which was localized to one ear, and the remaining 5 of 25 (20%) described their tinni-tus as bilateral or localized "in the head." Inter-estingly, and in contrast to Arnold et al (1996), Oestreicher et al (1999b) failed to show any sig-nificant activation patterns in primary cortical auditory areas. However, significant deactiva-tions (i .e ., reductions in blood flow) were observed in several nonauditory regions that included posterior parietal cortex, anterior insula, and posterior portion of the anterior cingulate cor-tex, which constitute portions of the limbic sys-tem. Whereas the interpretation of deactivations in PET studies is controversial (Mirz et al, 1999a), these authors suggest that their findings may represent a distributed neural processing network associated with the psychological com-ponent of the tinnitus perception.

In addition to studying glucose metabolism with PET, several other approaches have been proposed that would satisfy the conditions nec-essary for imaging tinnitus with fMRI or PET. That is, if the tinnitus percept could somehow be turned on and off, and/or if a constant back-ground tinnitus could be consciously modulated (i .e ., changed in loudness), then the conditions necessary for assessing a differential response, and thereby obtaining a difference image, would be satisfied (similar to the blocked-stimulus subtraction paradigm described previously using exogenous stimuli) . It was initially suggested that individuals with gaze-evoked tinnitus could meet these criteria by internally generating on and off states during data acquisition (Cacace et al, 1994a) . In its purest form, gaze-evoked tin-nitus is absent in a particular setting (i.e ., a cen-tral eye-gaze location) and present when a static change in eye gaze exceeds a certain spatial location and this eye position is maintained

(e.g., left/right horizontal, up/down vertical) (see Wall et al [1987] ; Cacace et al [1994b] for details of the auditory and visual psychophysics) . Attempts at localizing brain sites correlated with gaze-evoked tinnitus was reported by Cacace et al (1996a) and Cacace (1997) using fMRI, and several other groups have followed this lead using PET (Giraud et al, 1999; Lock-wood et al, 1999a) . For example, Lockwood et al (1999a) reported a group of individuals who could modulate (increase) the pitch and loudness of a constant with sustained lateral gaze (gaze-evoked tinnitus). Six of seven had pro-found hearing loss, presumably on the involved side, varying degrees of facial nerve injury, and mild abnormalities of ocular motility. Single-sub-ject repeated-measures analysis produced sig-nificant results in all individuals, but substantial variations among individuals were also noted. Activation sites included areas of brainstem (left vestibular nucleus), primary auditory, or auditory association cortex . After between-subject averaging and normalization to Talairach coordinates, activation foci were also noted in supramarginal gyrus and cerebellar vermis . In another study, Giraud et al (1999) found bilateral activations in auditory associ-ation areas in individuals with gaze-evoked tin-nitus and also provided detailed psychometric data correlating visual spatial coordinates with pitch and loudness data .

In addition to eye gaze, tinnitus loudness can also be modulated in some individuals by per-forming oral-facial movements (jaw clinching), and Lockwood et al (1998) have documented several brain activation sites in this select group of individuals. In this experiment, Lockwood et al (1998) used PET to evaluate if changes in loudness of tinnitus perceptions produced by oral-facial movements could be detected with this methodology. Other nontinnitus-related condi-tions were also assessed . Two groups were scanned separately during oral-facial move-ments (tinnitus group) and jaw clinching (nor-mal control group without tinnitus or hearing loss) and at rest (no activity). Normal controls showed bilateral activation of sensory-motor cortex and supplemental motor area in response to jaw clinching. In two patients where oral-facial movements increased tinnitus loudness (e.g ., where tinnitus was localized to the right ear in one patient and to the left ear in the other), increases in cerebral blood flow (CBF) were observed in sensory-motor cortex, in pri-mary auditory cortex in the left superior tem-poral gyrus, and in a region near the medial

260


geniculate nuclei . To separate changes in CBF

due to increases in tinnitus loudness, group sub-tractions were performed between PET results obtained during jaw clinching in controls and oral-facial movement in tinnitus patients . The subtraction procedure showed residual activa-tion in the left thalamic region (left medial genic-

ulate nuclei) in the tinnitus group . This was interpreted as indicating that the postsubtrac-tion increase in neural activity was due to the

increase in tinnitus loudness . In two patients where oral-facial movements decreased tinnitus

loudness (e .g ., where tinnitus was localized to the right ear in both individuals), a decrease in CBF was observed in the posterior and mid portion of the left middle temporal gyrus . Here the sub-traction procedure showed a region of reduced CBF in the temporal lobe and hippocampus of the left hemisphere .

It has also been shown that cutaneous-evoked tinnitus, a previously unrecognized phe-

nomenon, can also be imaged with f MRI and the PRESTO method (Cacace et al, 1999a, b) . In a single case, phantom auditory perceptions trig-gered by touching the fingertip region of the right hand produced localized activation in the

temporal-parietal junction . This represented activation in auditory centers of the brain on the superior aspect of the temporal and inferior aspect of the parietal lobes . The finger opposi-tion-tapping task also produced activity in the right (ipsilateral) caudate area, slight activation in the contralateral orbital frontal region, and prominent activation in the contralateral motor, premotor and pre-rolandic sulcus regions . A con-trol finger opposition-tapping task with the other hand produced activation limited to the contralateral motor cortex, premotor cortex, and pre-rolandic sulcus regions . These data clearly dissociated cutaneous-evoked tinnitus-related

activity from activation produced by the finger opposition-tapping task using the opposite hand . Recent evidence also suggests that somatosen-sory system interactions with auditory path-ways can generate tinnitus perceptions in individuals without overt otologic disturbances (Levine, 1999) . These observations, in con-junction with various other studies noted above, further expand the biologic basis of tinnitus

generation . Although in early stages of devel-opment, modulation of tinnitus loudness by eye gaze, oral-facial movements, cutaneous stimu-lation, and craniocervical manipulation are some of the ways in which to demonstrate the poten-tial for imaging phantom auditory perceptions in human subjects .

Other important paradigms have been reported in studying tinnitus-related activity in the CNS. For example, Mirz et al (1999b)

used PET to study conditions of tinnitus sup-pression/inhibition using narrow-band acoustic masking, pharmacologic tinnitus reduction using intravenous lidocaine, and in combined condi-

tions of acoustic masking plus lidocaine admin-istration . In this study, tinnitus-related activity was localized to the right prefrontal (middle

and superior) and right posterior (middle tem-poral and precuneous) gyri . Deactivations were observed in the left transverse temporal gyri . Recently, Melcher et al (1999b) demonstrated that tinnitus modified by lidocaine administra-tion can also be detected with f MRI using meth-ods described below.

A series of systematic experiments designed to study individuals with lateralized tinnitus based on an acoustic stimulation/masking par-adigm and the single-slice fMRI method (Guimaraes et al, 1998) has been used by a group of auditory neuroscientists (Levine et al, 1998; Melcher et al, 1999b, 2000; Sigalovsky et al, 1999). In these experiments, both the method and selection of individual subjects are impor-tant considerations . To reiterate briefly, the sin-gle-slice fMRI method, which captures neuroanatomic segments of primary auditory cortex and inferior colliculi, uses a rapid EPI sequence with cardiac gating . In this context, imager-related background noise is reduced by studying a single slice of anatomy, and the car-diac gating method serves to minimize vascular-related movement of subcortical structures and subject-related head movement. The selection of individuals with normal hearing sensitivity and lateralized tinnitus also tends to reduce alter-native explanations in comparison to using het-erogeneous populations with both hearing loss and more complex tinnitus perceptions. In a group of adults with normal hearing sensitivity and tinnitus lateralized to one ear, they found that presentation of binaural noise induced acti-vation in the inferior colliculi that was always more asymmetric in individuals with lateral-ized tinnitus than in controls . Interestingly, in comparison to normal controls, all individuals with unilateral tinnitus had abnormally low fMRI activations in response to binaural noise in the inferior colliculus contralateral to the tin-nitus percept. Two models, the saturation model and the physiologic masking model, have been proposed to allow for systematic hypothesis test-ing. Both models tacitly assume elevated base-line neuronal activity from the continuous

261


background tinnitus, which is thought to be pre-sent at the level of the inferior colliculus .

Although advancements are being made, it should be recognized that the various methods described above are not without limitations . For example, whereas both PET and f MRI allow for tinnitus localization and assessment of potential connectivity patterns underlying the phantom perceptions, they do not delineate the temporal structure of the underlying neural activity that may be responsible for the tinnitus perception (i .e ., single-unit spike trains, synchronized spon-taneous activity, ensemble activity, etc.) . Func-tional MRI studies of the auditory modality can also pose additional concerns due to imager-related background noise, as noted below.

LIMITATIONS OF STUDIES IN THE AUDITORY MODALITY

t should be recognized that fMRI studies in the auditory modality pose additional

concerns because of inherent imager-related background noise, due in large part to system hardware-related switching gradients during data acquisition . For paradigms that require active discriminations, this background inter-ference can be problematic. System-related noise is a complex issue because of level effects and sound/vibration transmission routes (air and bone conduction). As a result, passive attenua-tion devices (e.g., insert ear plugs, headphones, etc.) can only be partially successful at noise reduction (see Ravicz and Melcher [1998], Melcher et al [1999a], and Savoy et al [1999], for a more comprehensive discussion of these issues). Besides passive attenuation or even active noise cancellation/reduction devices (Goldman et al, 1989 ; McJury et al, 1997), other options include alterations in pulse sequences such as the pre-sentation of the stimulus during quiet periods of the pulse sequence (Belin et al, 1999; Eden et al, 1999 ; Edmister et al, 1999 ; Talavage et al, 1999). With respect to the issue of scanner noise and in contrast to f MRI, PET may have certain advantages . For example, Salvi et al (1999) have reported that when hearing threshold testing is conducted in the PET scanner which insert ear-phones covered by earmuffs with employ active noise reduction, thresholds are comparable to those obtained from an audiometric test booth over a bandwidth from approximately 0.5 to 8.0 kHz. This indicates that auditory paradigms using low-level auditory stimuli are possible to study without noise interference under these conditions using PET.

RELATIONS TO OTHER FORMS OF PHANTOM PERCEPTIONS

n addition to tinnitus, studying other types of phantom perceptions may also be of value

in order to gain insight into common properties of various forms of phantom events . Like gaze-evoked and cutaneous-evoked tinnitus, synesthesia is a condition where stimulation in one modality can evoke a conscious perceptual experience in another modality (cross-modal phantom perceptions) . For example, in color-word synesthesia, words can serve to activate specific colors in the absence of visible-luminous stimuli. In a PET study of color-word synes-thesia, Paulesu et al (1995) found that specific words but not tones evoked cross-modal synes-thetic perceptions. Whereas words activated perisylvian language areas and nonprimary visual association areas (posterior inferior tem-poral cortex and the parietal-occipital junc-tions), it was also found that prefrontal cortex, the insula, and superior temporal gyrus were also activated. Thus, this study indicates that brain areas concerned with language and visual-feature integration may underlie these particular cross-modal synesthetic perceptions without requiring overt activation of primary visual cortex .

Capturing neural activity associated with the experience of other phantom experiences such as auditory and visual hallucinations (Siber-sweig et al, 1995, 1996; Howard et al, 1997; Woodruff et al, 1997 ; Fytche et al, 1998 ; Szecht-man et al, 1998 ; David, 1999 ; Lennox et al, 1999) represents another line of investigation using contemporary imaging modalities that may have parallels to understanding certain forms of tinnitus (also see Cacace et al [1999b] for a discussion). Consider, for example, an experiment designed to determine common brain areas involved in hearing, imaginal hearing, and hypnosis-induced auditory hallucinations in attempts to establish linkage between these dif-ferent states . The premise is that auditory hal-lucinations share with imaginal hearing the property of being self-generated and with real hearing the experience of the stimulus as being an external one. In this study, Szechtman et al (1998) compared changes in regional blood flow using PET under conditions of hearing (auditory presentation of repetitions of a taped verbal sentence), imagining, hypnosis-induced audi-tory hallucinations, and a baseline nonactivated state. Data were obtained from university stu-dents who were divided into two groups : hallu-

262


cinators and nonhallucinators . In hallucinators,

but not in nonhallucinators, it was found that a brain site in the right anterior cingulate region (Brodmann area 32) was activated by both real hearing and hallucinations but not by imaginal hearing . It was suggested that the right ante-

rior cingulate region may be the site in the brain where an auditory event is tagged as being

external . It is speculated that spontaneous audi-

tory hallucinations could result from inappro-priate activation of the cingulate cortex. This interpretation is consistent with other experi-

ments implicating cingulate cortices in atten-tional/intentional processes and views that suggest that aberrant activity in these regions may result in inappropriate allocation of atten-

tional resources or signal amplification in the modality of the hallucination (Vogt et al, 1992 ; Posner, 1994; Silbersweig and Stern, 1996) .

Likewise, parallels between tinnitus and pain have also been suggested (Tonndorf, 1987; Moller, 1997, 1999) . Like tinnitus, pain percep-tion is complex and poorly understood, although functional imaging studies are beginning to yield new and important information on this topic as well (Berman, 1995 ; Davis et al, 1998a, b; Derbyshire et al, 1998 ; Disbrow et al, 1998 ; Oshiro et al, 1998 ; Porro et al, 1998 ; Coghill et al, 1999 ; Ingvar, 1999, Treede et al, 1999). Based on neurophysiologically based models of tinni-tus, affective emotional reactions to the tinnitus perception and limbic system linkage have been hypothesized . Functional MRI studies may also provide additional details about state and trait relationships and interactions with emotion (Davidson and Irwin, 1999).

CONCLUSIONS

have limited advancements in this area . Although presently underused in the auditory modality, MRS is another powerful tool to fur-

ther our understanding of a wide range of otopathologic conditions and potential treat-ments. One point, however, that cannot be

emphasized enough is the fact that successful implementation of this technique requires con-structive and close collaborations from a variety

of scientific disciplines and individuals (MR physicists, engineers, organic and biochemists, statisticians, radiologists, sensory and cogni-

tive scientists, etc .) . Advancements in this area would not have occurred so rapidly without these important interactions .

Clearly, the future of fMRI looks extremely bright . This optimism is not understated because there is reason to believe that real-time func-

tional imaging and analysis is on the horizon

(Voyvodic, 1999 ; Yoo et al, 1999) . Furthermore,

if this particular application materializes, then clinical uses will no doubt expand . Although

experimental in nature, and many issues remain to be fully resolved, this fascinating but complex area of/MRI opens up a whole new research fron-

tier for auditory neuroscience to explore in the 21st century.

Acknowledgments. We thank Dr. Gary Jacobson for initially inviting us to write an article dealing with prin-

ciples of /MRI and its application to tinnitus research .

Whereas the original manuscript was intended for inclu-

sion in a special issue of JAAA on tinnitus, after editorial review and as the article evolved, it became apparent

that it would be better suited as a general and more inclu-

sive tutorial on a wider range of auditory processing . We

thank Dr. James Jerger for guiding us in the direction

of a tutorial presentation and in making this option available.

F unctional MRI is a research area that is pro-viding insight into fundamental properties

of human sensory, motor, and cognitive func-tion in both normal and pathologic states . This tutorial provides a snapshot of this powerful method, which is rapidly evolving and improv-ing over time . The appeal of this technique is based on its noninvasive nature, temporal and spatial resolution, suitability for studying humans, and potentially wide-range future avail-ability. It is also gaining prominence as an exper-imental tool to study and quantify various phantom sensory events such as tinnitus . The application of f MRI to tinnitus research is par-ticularly noteworthy since the lack of objective, noninvasive methods for studying this enig-matic phenomenon is one of many reasons that

REFERENCES

Achten E, Jackson GD, Cameron JA, Abbott DF, Stella DL, Fabini GC . (1999) . Presurgical evaluation of the motor hand area with functional MR imaging in patients with tumors and dysplastic lesions. Radiology 210: 529-538.

Aguirre GK, D'Esposito M. (1999) . Experimental design for brain /MRI . In: Moonen CTW Bandettini P, eds. Func-tional MRI. Berlin : Springer-Verlag, 369-380.

Arnold W, Bartenstein P, Oestreicher E, Romer W, Schwaiger M. (1996) . Focal metabolic activation in the predominant left auditory cortex in patients suffering from tinnitus : a PET study with ["Fl deoxyglucose . ORL J Oto rhinolaryngol Relat Spec 58:195-199 .

Atlas SW, Howard RS, Maldjian J, Alsop D, Detre JA, Listerud J, D'Esposito M, Judy KD, Zager E, Stecker M. (1996) . Functional magnetic resonance imaging of regional

263


brain activity in patients with intracerebral gliomas: findings and implications for clinical management. Neu-rosurgery 38:329-338 .

Bilecen D, Scheffler K, Schmid N, Tschopp K, Seelig J . (1998) . Tonotopic organization of the human auditory cortex as detected by BOLD-FMRI. Hear Res 126:19-27 .

Baciu MV, Rubin C, Decorps MA, Segbarth CM. (1999). FMRI assessment of hemispheric language dominance using a simple inner speech paradigm . NMR Biomed 12:293-298 .

Bandettini PA, Wong EC . (1997) . Magnetic resonance imaging of human brain function . Principles, practicali-ties, and possibilities . Neurosurg Clin North Am 8:345-371 .

Bandettini PA, Wong EC, Hinks RS, Tikofsky RS, Hyde JS . (1992) . Time course EPI of human brain function during task activation. Magn Reson Med 25:390-397 .

Bandettini PA, JesmanowiczA, Wong EC, Hyde JS . (1993) . Processing strategies for time-course data sets in func-tional MRI of the human brain . Magn Reson Med 30:161-173 .

Bates EA, Thal D, Janowsky J. (1992) . Early language development and its neural correlates. In: Segalowitz SJ, Rapin I, eds. Handbook of Neuropsychology. Vol 7: Child Neuropsychology. Amsterdam: Elsevier Science, 69-110 .

Bauer CA, Brozoski TJ, Rojas R, Boley J, Wyder M. (1999). Behavioral model of chronic tinnitus in rats . Otolaryn-gol Head Neck Surg 121:457-462 .

Baumgart F, Gaschler-Markefski B, Woldorf MG, Heinze HJ, Scheich H. (1999) . A movement sensitive area in audi-tory cortex . Nature 400:724-726 .

Baumgartner R, Scarth G, Terchtmeister C, Somorjai R, Moser E. (1997) . Fuzzy clustering of gradient-echo func-tional MRI in the human visual cortex . Part I : Reproducibility. J Magn Reson Imaging 7:1094-1101 .

Beaton AA . (1997) . The relation of planum temporale asymmetry and morphology of the corpus callosum to handedness, gender and dyslexia : a review of the evi-dence. Brain Lang 60:255-322 .

Belin P, Zattore RJ, Hoge R, Evans AC, Pike B . (1999) . Event-related /MRI of the auditory cortex . Neuroimage 10:417-429 .

Belin P, Zattore RJ, LaFaille P, Ahad P, Pike B . (2000) . Voice-selective areas in human auditory cortex . Nature 403:309-312 .

Belliveau JW, Kennedy DN, McKinstry RC, Buchbinder BR, Weisskoff RM, Cohen MS, Vevea JM, Brady TJ, Rosen BR . (1991) . Functional mapping of the human visual cortex by magnetic resonance imaging . Science 254:716-719 .

Benson RR, Logan WJ, Cosgrove GR, Cole AJ, Jing H, LeSueur LL, Buchbinder BR, Rosen BR, Caviness VS Jr. (1996) . Functional MRI localization of language in a 9-year-old child. Can J Neurosci 23:213-219 .

Benson RR, Fitzgerald DB, LeSueur LL, Kennedy DN, Kwong KK, Buchbinder BR, Davis TL, Weisskoff RM, Talavage TM, Logan WJ, Cosgrove GR, Belliveau JW Rosen BR. (1999) . Language dominance determined by whole brain functional MRI in patients with brain lesions. Neurology 52:798-809 .

Berman J . (1995) . Imaging pain in humans . Br JAnaesth 75:209-216 .

Binder J. (1997) . Functional magnetic resonance imag-ing. Language mapping. Neurosurg Clin North Am 8:383-392 .

Binder J. (1999). Functional MRI of the language system . In: Moonen CTW, Bandettini PA, eds. Functional MRI. New York : Springer-Verlag, 407-419.

Binder J, Rao SM, Hammeke TA, Frost JA, Bandettini PA, Jesmanowicz A, Hyde JS . (1995a). Laterahzed human brain language systems demonstrated by task subtrac-tion functional magnetic resonance imaging. Arch Neurol 52:593-601 .

Binder JR, Swanson SJ, Hammeke TA, Morris GL, Mueller WM, Fischer M, Benbadis S, Frost JA, Rao SM, Haughton VM. (1995b). Determination of language dom-inance using functional MRI: comparison with the Wada test . Neurology 46:978-984 .

Blamire AM, Ogawa S, Ugurbil K, Rothman D, McCarthy G, Ellermann JM, Hyder F, Rattner Z, Shul-man RG. (1992) . Dynamic mapping of the visual cortex by high-speed magnetic resonance imaging . Proc Natl Acad Sci U S A 89 :11069-11073 .

Bloch F, Hansen WW, Packard M. (1946) . Nuclear induc-tion . Phys Rev 69:127 .

Bookheimer SY (1996) . Functional MRI applications in clinical epilepsy. Neuroimage 4:S139-5141 .

Bookheimer SY, Dapretto M, Karmarker U. (1999) . Func-tional MRI in children with epilepsy. Dev Neurosci 21:191-199 .

Booth JR, Macwhinney B, Thulborn KR, Sacco K, Voyvodic J, Feldman HM. (1999) . Functional organization of acti-vation patterns in children : whole brain /MRI imaging during three different cognitive tasks. Prog Neuropsycho-pharmacol Biol Psychiatry 23:669-682 .

Bracewell RN . (1986) . The Fourier Transform and Its Applications . San Francisco: McGraw-Hill.

Bradley WG, Chen D-Y, Atkinson DJ, Edelman RE . (1999) . Fast spin-echo and echo planar imaging In : Stark DD, Bradley WG, eds . Magnetic Resonance Imaging . Volume 3, 3rd Ed . St . Louis : Mosby, 125-157 .

Brooks WM, Friedman SD, Stidley CA. (1999) . Repro-ducibility of 1H-MRS in vivo . Magn Reson Med 41:193-197 .

Bucher SF, Diederich M, Wiesmann M, Weiss A, Zink R, Yousry TA, Brandt T. (1998) . Cerebral functional mag-netic resonance imaging of vestibular, auditory, and nociceptive areas during galvanic stimulation. Ann Neurol 44:120-125 .

Buckner RL . (1998) . Event-related /MRI and the hemo-dynamic response . Hum Brain Map 6:373-377 .

Buckner RL, Bandettini PA, O'Craven KM, Savoy RL, Petersen SE, Raichle ME, Rosen BR (1996) . Detection of cortical activation during averaged single trials of a cog-nitive task using functional magnetic resonance imaging . Proc Natl Acad Sci U S A 93:14878-14883 .

264


Buckner RL, Braver TS. (1999) . Event related functional

MRI. In : Moonen CTW, Bandettini PA, eds. Functional

MRI. New York : Springer-Verlag, 441-452.

Buckner RL, Kelley WM, Petersen SE . (1999) . Frontal

cortex contributes to human memory formation. Nat Neu-

rosci 2:311-314 .

Bullmore ET, Rabe-Hesketh S, Morris RG, Williams SCR,

Gregory WL, Gray JA, Brammer MI . (1996) . Functional

magnetic resonance image analysis of a large-scale neu-

rocognitive network . Neuroimage 4:16-33 .

Cacace AT. (1997) . Imaging tinnitus with fMRI, Presi-

dential Symposium. Assoc Res Otolaryngol 20 :7 .

Cao Y, Vikingstad EM, George KP, Johnson AF, Welch

KM. (1999a) . Cortical language activation in stroke

patients recovering from aphasia with functional MRI.

Stroke 30:2331-2340 .

Cao Y, Welch KMA, Aurora S, Vikingstad EM. (1999b).

Functional MRI-BOLD of visually triggered headache in

patients with migraine . Arch Neural 56:548-554 .

Chen G, Jastreboff PJ. (1995) . Salicylate-induced abnor-

mal activity in the inferior colliculus of rats . Hear Res

82:158-178.

Clare S, Humberstone M, Hykin J, Blumhardt L, Bowtell

R, Morris P (1999) . Detecting activations in event-related

fMRI using analysis of variance . Magn Reson Med

42:1117-1122 .

Cacace AT . (1999) . Delineating tinnitus-related activity

in the nervous system : application of functional imaging

at the fin de siecle . In : Hazell J, ed . Proceedings of the

Sixth International nnnitus Seminar. London : The Tin-nitus and Hyperacusis Centre, 39-44.

Cacace AT, Lovely TJ, McFarland DJ, Parries SM, Winter

DF. (1994a). Anomalous cross-modal plasticity following

posterior fossa surgery: some speculations on gaze-evoked

tinnitus . Hear Res 81:22-32 .

Cacace AT, Lovely TJ, Winter DF, Parries SM, McFar-

land DJ . (1994b). Auditory perceptual and visual spatial

characteristics of gaze-evoked tinnitus . Audiology

33:291-303 .

Cacace AT, McFarland DJ . (1998) . Central auditory pro-

cessing disorders in school-age children : a critical review.

J Speech Lang Hear Res 41 :355-373 .

Cacace AT, Cousins J, Moonen CWT, Van Gelderen P,

Parnes Lovely TJ . (1996a). In-viva localization of phan-

tom auditory perceptions during functional magnetic

resonance imaging of the human brain. In : Reich G,

Vernon J, eds. Proceedings of the Fifth International 7ln-

nitus Seminar. Portland : American Tinnitus Association,

397-401.

Cacace AT, Parries SM, Lovely TJ, Winter DF, McFar-

land DJ . (1996b). Gaze-evoked tinnitus following

unilateral peripheral auditory deafferentation : a case for

anomalous cross-modal plasticity. In : Salvi R, Hender-

son D, Fiorino F, Colletti V, eds: Auditory System Plasticity

and Regeneration . New York : Thieme Medical, 354-358.

Cacace AT, Cousins JP, Parries SM, Semenoff D, Holmes

T, McFarland DJ, Davenport C, Stegbauer K, Lovely TJ .

(1999a). Cutaneous-evoked tinnitus : I. Phenomenology,

psychophysics and functional imaging. Audiol Neurootol

4:247-257 .

Cacace AT, Cousins JP, Parnes SM, McFarland DJ,

Semenoff D, Holmes T, Davenport C, Stegbauer K, Lovely

TJ . (1999b). Cutaneous-evoked tinnitus . II . Review of

neuroanatomical, physiological and functional imaging

studies. Audiol Neurootol 4:258-268 .

Cacace AT, McFarland DJ, Ouimet JR, Schrieber EJ,

Marro P. (2000) . Temporal processing deficits in remedi-ation-resistant reading-impaired children . Audiol Neurootol 5:83-97 .

Calvert GA, Brammer MJ, Bullmore ET, Campbell R,

Iversen SD, David AS . (1999) . Response amplification in

sensory-specific cortices during crossmodal binding. Neu-

roreport 20:2619-2623 .

Coghill RC, Sang CN, Maisog JM, Iadarola MJ . (1999) .

Pain intensity processing within the human brain: a bilat-

eral, distributed mechanism. J Neurophysiol 82:1934-1943 .

Cohen MS . (1996) . Rapid MRI and functional applica-

tions. In : AW Toga, Mazziotta JC, eds. Brain Mapping:

The Methods. San Diego: Academic Press, 223-255.

Cohen MS, Bookheimer SY. (1994) . Localization of brain

function using magnetic resonance imaging. Trends Neu-

rosci17 :268-277 .

Constable RT, Kennan RP, Puce A, McCarthy G, Gore

JC . (1994) . Functional NMR imaging using fast spin echo

at 1.5T. Magn Reson Med 31:686-690 .

Cousins JP. (1995) Clinical MR spectroscopy : funda-

mentals, current applications, and future potential . AJR

Am J Roentgenol164:1337-1347 .

Cuenod CA, Bookheimer SY, Hertz-Pannier L, Zeffiro

TA, Theodore WH, LeBihan D. (1995) . Functional MRI

during word generation using conventional equipment:

a potential tool for language localization in clinical envi-

ronment. Neurology 45:1821-1827 .

Dale AM, Buckner RL . (1997) . Selective averaging of

rapidly presented individual trials using fMRI . Hum

Brain Map 5:329-340 .

Danielsen ER, Ross BD . (1999) . Neurospectroscopy. In :

Stark DD, Bradley WG, eds. Magnetic Resonance Imag-

ing. Vol. 3, 3rd Ed . St. Louis: Mosby, 1595-1635.

David AS . (1999) . Auditory hallucinations : phenomenol-ogy, neuropsychology and neuroimaging update . Acta Psychiatr Scand Suppl 395 :95-104 .

Davidson RJ, Irwin W. (1999). Functional MRI in the

study of emotion. In : Moonen CTW, Bandettini PA, eds.

Functional MRI. New York: Springer-Verlag, 487-499.

Davis KD, Kwan CL, Crawley AP, Mikulis DJ . (1998a).

Functional MRI study of thalamic and cortical activa-

tions evoked by cutaneous heat, cold, and tactile stimuli.

J Neurophysiol80:1533-1546 .

Davis KD, Kwan CL, Crawley AP, Mikulis DJ . (1998b).

Event-related f MRI of pain : entering a new era in imag-

ing pain . Neuroreport 14:3019-3023 .

Derbyshire SW, Vogt BA, Jones AK . (1998) . Pain and

stroop interference tasks activate separate processing

modules in anterior cingulate cortex . Exp Brain Res

118:52-60 .

265


Desmond JE, Sum JM, Wagner AD, Demp JB, Shear PK, Glover GH, Gabrieli JD, Morrell MJ . (1995) . Functional MRI measurement of language lateralization in Wada-tested patients . Brain 118:1411-1419.

D'Esposito M, Zarahn E, Aguirre GK . (1999) . Event-related functional MRI : implications for cognitive psychology. Psychol Bull 125:155-164 .

Ellerman J, Garwood M, Henderich K, Hinke R, Hu X, Ugurbil K. (1994) . Functional imaging of the brain by nuclear magnetic resonance. In : Gilles R, ed . NMR in Physiology and Biomedicine. San Diego: Academic, 137-150.

Elliot LL. (1994). Functional brain imaging and hearing. J Acoust Soc Am 96:1397-1407.

Detre JA, Leigh JS, Williams DS, KoretskyAP. (1992) . Perfusion imaging. Magn Reson Med 23:37-45 .

Detre JA, Sirven JI, Alsop DC, O'Connor MJ, French JA . (1995). Localization of subcortical ictal activity by func-tional magnetic resonance imaging : correlation with invasive monitoring . Ann Neural 38:618-624 .

Detre JA, Alsop DC, Aguirre GK, Sperling MR. (1996) . Coupling of cortical and thalamic ictal activity in human partial epilepsy : demonstration by functional magnetic resonance imaging. Epilepsia 37:657-661 .

Disbrow E, Buonocore M, Antognini J, Carstens E, Rowley HA. (1998) . Somatosensory cortex : a comparison of the response to noxious thermal, mechanical, and electrical stimuli using functional magnetic resonance imaging. Hum Brain Map 6:150-159 .

Duchowny M, Jayakar P, Harvey AS, Resnick T, Alvarez L, Dean P, Levin B. (1996) . Language cortex represen-tation: effects of developmental versus acquired pathology. Ann Neural 40:31-38 .

Duyn JH, Mattay VS, Sexton RH, Sobering GS, Barrios FA, Liu G, Frank JA, Weinberger DR, Moonen CTW. (1994a). 3-dimensional functional imaging of human brain using echo-shifted FLASH MRI. Magn Reson Med 32:150-155 .

Duyn JH, Moonen CWT, Yperen GH, Boer RW, Luyten PR . (1994b) . Inflow versus deoxyhemoglobin effects in BOLD functional MRI using gradient echoes at 1 .5T. NMR Biomed 7:83-88 .

Eddington DK, Melcher JR, Garcia N, Qin M, Sroka J, WeisskoffRM. (1999) . Electrically evoked cortical activ-ity in cochlear implant subjects can be mapped using functional magnetic resonance imaging. Assoc Res Oto-laryngol 22:41.

Eden GF, Zeffiro TA . (1998) . Neural systems affected in developmental dyslexia revealed by functional neu-roimaging. Neuron 21:279-282 .

Eden GG, Joseph JE, Brown HE, Brown CP, Zeffiro TA . (1999) . Utilizing hemodynamic delay and dispersion to detect /MRI signal change without auditory interference . The behavior interleaved gradients technique. Magn Reson Med 41:13-20 .

Edmister WB, Talavage TM, Ledden PJ, Weisskoff RM . (1999) . Improved auditory cortex imaging using clustered volume acquisitions . Hum Brain Map 7:89-97.

Eggermont JJ, Kenmochi M. (1998) . Salicylate and qui-nine selectively increase spontaneous firing rates in secondary auditory cortex. Hear Res 117:149-160 .

Eggermont JJ, Komiya H. (1999) . Auditory cortex reor-ganization after noise trauma : relation to tinnitus? In : Hazell J, ed . Proceedings of the Sixth International Tin-nitus Seminar. London : Tinnitus and Hyperacusis Centre, 146-151.

Ernst T, Hennig J. (1994) . Observation of a first response in functional MR . Magn Reson Med 32:146-149 .

Farzaneh F, Riederer SJ, Pelc NJ . (1990) . Analysis of T2 limitations and off-resonance effects on spatial resolu-tion and artifacts in echo-planar imaging . Magn Reson Med 14:123-139 .

Faurion A, Cerf B, Le Bihan D, Pillas AM . (1998) . /MRI study of taste cortical areas in humans . Ann N YAcad Sci 855:535-545 .

Forster BB, MacKay AL, Whittall KP, Mehl KA, Smith AM, Hare RD, Liddle PF. (1998) . Functional magnetic resonance imaging: the basics of blood-oxygen-level depen-dent (BOLD) imaging . Can Assoc Radial 49:320-329 .

Fox PT, Raichle ME . (1986) . Focal physiologic uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects . Proc Natl Acad Sci U S A 83:1140-1146 .

Fox PT, Raichle ME, Mintun MA, Dence C . (1988) . Nonox-idative glucose consumption during focal physiologic neural activity. Science 241:462-464 .

Frackowiak RSJ, Friston KJ, Frith CD, Roland RJ, Mazz-iotta JC . (1997) . Human Brain Function . San Diego : Academic .

Frahm J, Hanefeld F. (1997). Localized proton magnetic resonance spectroscopy of brain disorders in childhood. In : Bachelard HS, ed. Magnetic Resonance Spectroscopy and Imaging in Neurochemistry . New York : Plenum, 329-402.

Frahm J, Haase A, Matthei D. (1986) . Rapid NMR imag-ing of dynamic processes using the FLASH technique. Magn Reson Med 3:321-327 .

Frahm J, Merboldt KD, Hanicke W, Kleinschmidt A, Boecker H. (1994) . Brain or vein-oxygenation or flow? On signal physiology in functional MRI of human brain activation . NMR Biomed 7:45-53 .

Fransson P, Kruger G, Merboldt KD, Frahm J. (1998) . Temporal characteristics of oxygenation-sensitive MRI responses to visual activation in humans. Magn Reson Med 39:912-919 .

Fried I, Nenov VI, Ojeman SG, Woods RP (1995) . Func-tional MR and PET imaging of rolandic and visual cortices for neurosurgical planning. JNeurosurg 83 :854-861 .

Friston KJ, Holmes AP, Worsley KJ, Poline JB, Frith CD, Frackowiak RSJ. (1995a). Statistical parametric maps in functional imaging. A general linear approach . Hum Brain Map 2:189-210 .

Friston KJ, Frith CD, Frackowiak RSJ, Turner R. (1995b). Characterizing dynamic brain response with /MRI . A multivariate approach. Neuroimage 2:166-172 .

266


Friston KJ, Williams S, Howard R, Frackowiak RSJ, Turner R. (1996) . Movement-related effects in /MRI time series . Magn Reson Med 35:346-355 .

Friston KJ, Zarahn E, Josephs O, Henson RN, Dale AM. (1999) . Stochastic designs in event-related /MRI . Neuro-image 10:607-619 .

Frost JA, Binder JR, Springer JA, Hammeke TA, Bel-lowan PSF, Rao SM, Cox RW (1999) . Language processing is strongly left lateralized in both sexes. Evidence from functional MRI. Brain 122:199-208.

Fulton JF. (1955) . Textbook of Physiology . Philadelphia : WB Saunders .

Fytche DH, Howard RJ, Brammer MJ, David A, Woodruff P, Williams S. (1998). The anatomy of conscious vision : an fMRI study of visual hallucinations . Nat Neurosci 1:738-742 .

Hashimoto T, Tayama M, Miyazaki M, Fujii E, Harada M, Miyoshi H, Tanouchi M, Kuroda Y (1995) . Develop-mental brain changes investigated with proton magnetic resonance spectroscopy. Dev Med Child Neurol 37:398-405 .

Hennig J, Nauerth A, Friedburg H. (1986) . RARE imag-ing: a fast imaging method for clinical MR . Magn Reson Med 3:823-833 .

Henson RN, Shallice T, Dolan RJ . (1999) . Right prefrontal cortex and episodic memory retrieval: a functional MRI test of the monitoring hypothesis . Brain 122:1367-1381 .

Hertz-Pannier L, Gaillard WD, Mott SH, Cuenod CA, Bookheimer SY, Weinstein S, Conry J, Papero PH, Schiff SJ, Le Bihan D, Theodore WH. (1997) . Noninvasive assessment of language dominance in children and ado-lescents with functional MRI : a preliminary study. Neurology 48:1003-1012 .

Gaillard WD, Bookheimer SY, Hertz-Pannier L, Blaxton TA. (1997) . The noninvasive identification of language function . Neuroimaging and rapid transcranial magnetic stimulation. Neurosurg Clin North Am 8: 321-335.

Gaschler-Markefski B, Baumgart F, Tempelman C, Schindler F, Stiller D, Heinze H-J, Scheich H. (1997) . Statistical methods in functional magnetic resonance imaging with respect to nonstationary time-series: audi-tory cortex activity. Magn Reson Med 38:811-820 .

Gati JS, Menon RS, Ugurbil K, Rutt BK. (1997) . Exper-imental determination of the BOLD field strength dependence in vessels and tissue . Magn Reson Med 38:296-302 .

Gevins A, Smith ME, McEvoy LK, Leong H, Le J. (1999) . Electroencephalographic imaging of higher brain func-tion. Philos Trans R Soc Load B Biol Sci 354:1125-1133 .

GiraudAL, Chery-froze S, Fischer G, Fischer C, Vighetto A, Gregoire M-C, Lavenne F, Collet L. (1999) . A selective imaging of tinnitus . Neuroreport 10 :1-5 .

Goldman AM, Gossman WE, Friedlander PC . (1989) . Reduction of sound levels with anti-noise in MR imag-ing. Radiology 173:549-550.

Graveline CJ, Mikulis DJ, Crawley AP, Hwang PA . (1998) . Regionalized sensorimotor plasticity after hemispherec-tomy /MRI evaluation . Pediatr Neural 19:337-342 .

Griffiths TD, Rees A, Witton C, Shakir RA, Henning GB, Green GG. (1996) . Evidence for a sound movement area in the human cerebral cortex. Nature 383:425-427 .

Griffiths TD, Green GG, Rees A, Rees G. (2000) . Human brain areas involved in the analysis of auditory move-ment. Hum Brain Mapp 9:72-80.

Grimson WEL, Kikinis R, Jolesz FA, Black PM . (1999). Image-guided surgery. Sci Am 280:62-69 .

Guimaraes AR, Melcher JR, Talavage TM, Baker JR, Ledden P, Rosen BR, Kiang NYS, Fullerton BC, Weis-skoff RM . (1998) . Imaging subcortical auditory activity in humans . Hum Brain Map 6:33-41 .

Hackett TA, Stepniewska I, Kass JH . (1999). Prefrontal connections of the parabelt auditory cortex in macaque monkeys. Brain Res 817 :45-58 .

Hahn E. (1950) . Spin echoes . Phys Rev 80:580-594.

Hickok G, Love T, Swinney D, Wong EC, Buxton RB . (1997) . Functional MR imaging during auditory word perception : a single-trial presentation paradigm . Brain Lang 58:197-201 .

Hofman E, Preibisch C, Knaus C, Muller J, Kremser C, Teissl C. (1999) . Noninvasive direct stimulation of the cochlear nerve for functional MR imaging of the auditory cortex . Am J Neuroradiol 20:1970-1972 .

Holodny AI, Schulder M, Liu WC, Maldjian JA, Kalnin AJ . (1999) . Decreased BOLD functional MR activation of the motor and sensory cortices adjacent to a glioblas-toma multiforme : implications for image-guided neurosurgery. Am J Neuroradiol20:609-612 .

Hoshino Y, Yoshikawa K, Inoue Y, Asai S, Nakamura T, Ogino T, Umeda M, Iwamoto A. (1999) . Reproducibility of short echo time proton magnetic resonance spectroscopy using point-resolved spatially localized spectroscopy sequence in normal human brains . Radial Med 17:115-120 .

Howard MA III, Volkov IO, Abbas PJ, Damasio H, Ollendiek MC, Granner MA. (1996) . A chronic micro-electrode investigation of the tonotopic organization of human auditory cortex. Brain Res 724:260-264 .

Howard R, David A, Woodruff P, Mellers I, Wright J, Brammer M, Bullmore E, Williams S. (1997) . Seeing visual hallucinations with functional magnetic resonance imaging. Dement Geriatr Cogn Disord 8:73-77 .

Howseman AM, Bowtell RW (1999) . Functional magnetic resonance imaging: imaging techniques and contrast mechanisms . Philos Trans R Soc Lond B Biol Sci 354:1179-1194 .

Huckins SC, Turner CW Doherty KA, Fonte MM, Szev-erenyi NM. (1998) . Functional magnetic resonance imaging measures of blood flow patterns in human audi-tory cortex in response to sound. J Speech Lang Hear Res 41:538-548 .

Hykin J, Moore R, Duncan K, Clare S, Baker P, Johnson I, Bowtell R, Mansfield P, Gowland P. (1999) . Fetal brain activity demonstrated by functional magnetic resonance imaging. Lancet 354: 645-646.

Ingvar M. (1999) . Pain and functional imaging. Philos Trans R Soc Lond B Biol Sci 354:1347-1358 .

267


Jackson GD, Connelly A, Cross JH, Gordon 1, Gadian DG . (1994) . Functional magnetic resonance imaging of focal seizures . Neurology 44:850-856 .

Jackson JJ, Meyer CH, Nishimura DG, Macovski A. (1991) . Selection of a convolution function for Fourier inversion using gridding . IEEE Trans Med Imag 10:473-478 .

Jancke L, Shah NJ, Posse S, Grosse-Ryuken M, Mtiller-Gartner H-W (1998). Intensity coding of auditory stimuli: an fMRI study. Neuropsychologia 36:875-883 .

Jancke L, Mirzazade S, Shah NJ . (1999) . Attention mod-ulates activity in the primary and the secondary auditory cortex : a functional magnetic resonance imaging study in human subjects . Neurosci Lett 266:125-128 .

and functional images at 4 Tesla. Magn Reson Med 35:895-902 .

Kim S-G, Ugurbil K. (1997). Functional magnetic reso-nance imaging of the human brain. J Neurosci Methods 74:229-243 .

Koizuka I, Yano H, Nagahara M, Mochizuki R, Seo R, Shimada K, Kubo T, Nogawa T. (1994) . Functional imag-ing of the human olfactory cortex by magnetic resonance imaging. ORL J Oto rhinolaryngol Relat Spec 56:273-275 .

Krings T, Buchbinder BR, Butler WE, Chiappa KH, Jiang HJ, Cosgrove GR, Rosen BR . (1997) . Functional mag-netic resonance imaging and transcranial magnetic stimulation : complementary approaches in the evalua-tion of cortical motor function . Neurology 48:1406-1416 .

Jastreboff PJ, Brennan JF, Coleman JK, Sasaki CT. (1988) . Phantom auditory sensation in rats : an animal model for tinnitus . Behav Neurosci 102:811-822 .

Jung RE, Brooks WM, Yeo RA, Chiulli SJ, Weers DC, Sibbitt WL Jr. (1999a). Biochemical markers of intelli-gence : a proton MR spectroscopy study of normal human brain . Proc R Soc Lond B Biol Sci 266:1375-1379 .

Jung RE, Yeo RA, Chiulli SJ, Sibbitt WL Jr, Weers DC, Hart BL, Brooks WM. (1999b). Biochemical markers of cognition: a proton MR spectroscopy study of normal human brain. Neuroreport 10 :3327-3331 .

Kaas JH, Hackett TA . (1999) . `What' and ̀ where' pro-cessing in auditory cortex . Nat Neurosci 2:1045-1047 .

Kaas JH, Hackett TA, Tramo MJ. (1999) . Auditory pro-cessing in primate cerebral cortex . Curr Opin Neurosci 9:164-170 .

Kaltenbach JA, Godfrey DA, McCaslin DL, Squire AB . (1996). Changes in spontaneous activity and chemistry of the cochlear nucleus following intense sound exposure . In : Reich GE, Vernon JE, eds. Proceedings of the Fifth International Tinnitus Seminar. Portland : American Tin-nitus Association, 457-466.

Kaltenbach JA, Heffner HE, Afman CE. (1999) . Effects of intense sound on spontaneous activity in the dorsal cochlear nucleus and its relation to tinnitus . In : Hazell J, ed . Proceedings of the Sixth International Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 133-138.

Karni A, Gundela M, Jezzard P, Adams MM, Turner R, Ungerleider LG . (1995) . Functional MRI evidence for adult motor cortex plasticity during motor skill learning. Nature 377:155-158 .

Kauer JS, Nemitz JW, Sasaki CT. (1982) . Tinnitus aurium : fact . . . or fancy? Laryngoscope 92:1401-1407 .

Kerman RP, Zhong J, Gore JC . (1994). Intravascular sus-ceptibility contrast mechanisms in tissues. Magn Reson Med 31:9-21 .

Khalaffa O, Bushara O, Weeks RA, Kenji I, Catalan MJ, Tian B, Rauschecker JP, Hallet M. (1999) . Modality-specific frontal and parietal areas for auditory and visual spatial localization in humans . Nat Neuroscience 2:759-766 .

Kim S, Hu X, Adriany G, Ugurbil K. (1996) . Fast inter-leaved echo-planar imaging with high resolution anatomic

KumarA, Welti D, Ernst RR . (1975) . NMR Fourier zeug-matography. J Magn Reson 18:69-83 .

Kuppusamy K, Lin W Haacke EM. (1997) . Statistical assessment of cross correlation and variance methods and the importance of electrocardiogram gating in func-tional magnetic resonance imaging. Magn Reson Imaging 15:169-181 .

Kwong KK, Belliveau JW Chesler DA, Goldberg JE, Wiesskoff RM, Poncelet BP, Kennedy KW, Hoppel BE, Cohen MS, Turner R, Cheng HM, Brady TJ, Rosen R. (1992a). Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc Natl Acad Sci U S A 89:5675-5679 .

Kwong KK, Wanke J, Donahue KM, Davis TL, Rosen BR. (1992b). EPI imaging of global increase of brain MR signal with breath-hold preceded by breathing Oz . Magn Reson Med 33:448-452 .

Lange N. (1999) . Statistical procedures for functional MRI. In : Moonen CTW Bandettini PA, eds. Functional MRI. New York : Springer-Verlag, 301-335.

Larisch R, KStter R, Kehren F, Tosch M, Shah NJ, Kalveram KT, Jancke L, Miiller-Gartner H-W (1999) . Motivation effects in a dichotic listening task as evident from functional magnetic resonance imaging in human subjects . Neurosci Lett 267:29-32 .

Latchaw RE, Hu X, Ugurbil K, Hall WA, Madison MT, Heros RC. (1995) . Functional magnetic resonance imag-ing as a management tool for cerebral arteriovenous malformations. Neurosurgery 37:619-625 .

Lauterbur PC. (1973) . Image formation by induced local interactions : examples employing nuclear magnetic res-onance . Nature 242:190-191 .

LeBihan D, Karni A. (1995) . Applications of magnetic resonance imaging to the study of human brain function . Curr Opin Neurobiol 5:231-237 .

Lee AT, Glover GH, Meyer CH . (1995) . Discrimination of large venous vessels in time-course spiral blood-oxygen-level-dependent magnetic resonance functional neuroimaging. Magn Reson Med 33:745-754 .

Lee BC, Kuppusamy K, Grueneich R, El-Ghazzawy O, Gordon RE, Lin W Haacke EM . (1999). Hemispheric lan-guage dominance in children demonstrated by functional magnetic resonance imaging. J Child Neural 14:78-82 .

Lee CC . (1999) . Assessment of functional MR imaging in neurosurgical planning . Am JNeuroradiol 20:1511-1519 .

268

Lennox BR, Bert S, Park G, Jones PB, Morris PG . (1999) . Spatial and temporal mapping of neural activity associ-ated with auditory hallucinations . Lancet 353:644.

Levine RA . (1999) . Somatic (craniocervical) tinnitus and the dorsal cochlear nucleus hypothesis . Am J Otolaryn-gol 20:351-362 .

Levine RA, Melcher JR, Sigalovsky I, Guinan JJ . (1998) . Abnormal inferior colliculus activation in subjects with lateralized tinnitus . Ann Neurol 44:441 .

Levy LM, Henkin RI, Hutter A, Lin CS, Martins D, Schellinger D. (1997) . Functional MRI of human olfac-tion . J Comput Assist Tomogr 21:849-856 .

Liu HL, Gao JH . (1999) . Perfusion-based event-related functional MRI. Magn Reson Med 42:1011-1013 .

Lobel E, Kleine JF LeBihan D, Leroywillig A, Berthoz A. (1998) . Functional MRI of galvanic vestibular stimu-lation . JNeurophysiol 80:2699-2709 .

Lockwood AH, Salvi RJ, Coad ML, Towsley MA, Wack DS, Murphy BW. (1998) . The functional neuroanatomy of tinnitus : evidence for limbic system links and neural plasticity. Neurology 50:114-120 .

Lockwood AH, Burkard RF, Salvi RJ, Arnold SA, Reyes SA, Coad ML, LincoffNS, Galanowicz PJ (1999a). Positron emission tomography (PET) studies of gaze-evoked tin-nitus. Assoc Res Otolaryngol 22:119 .

Lockwood AH, Salvi RJ, Burkard RF, Galantowicz PJ, Coad ML, Wack DS . (1999b). Neuroanatomy of tinnitus . Scand Audiol 28 : Suppl 51:47-52 .

Lu ZL, Sperling G. (1995) . The functional architecture of visual motion perception . Vision Res 35:2697-2722 .

Magistretti PJ, Pellerin L. (1999) . Regulation of cerebral energy metabolism . In : Moonen CTW, Bandettini PA, eds. Functional MRI. New York: Springer-Verlag, 25-34.

Makeig S, Jung TP, Bell AJ, Ghahremani D, Sejnowski TJ . (1997) . Blind separation of auditory event-related brain responses into independent components . Proc Natl Acad Sci U S A 94:10979-10984 .

Maldjian MJ, Atlas SW, Howard RW, Greenstein E, Alsop D, Detre JA, Listerud J, D'Esposito M, Flamm ES . (1996). Functional magnetic resonance imaging of regional brain activity in patients with intracerebral arteriovenous mal-formations before surgical or endovascular therapy. J Neurosurg 84:477-483 .

Malonek D, Grinvald A. (1996) . Interactions between electrical activity and cortical microcirculation revealed by imaging spectroscopy : implications for functional brain mapping. Science 27:351-354.

Mansfield P. (1977) . Multi-planar image formation using NMR spin-echoes. J Phys C 10:L55-L58 .

Mattay VS, Weinberger DR . (1999) . Organization of the human motor system as studied by functional magnetic resonance imaging. Eur J Radiol 30:105-114 .

Mattson J, Simon M. (1996) . The Pioneers of NMR and Magnetic Resonance in Medicine: The Story of MRI. New York: Dean .

McCarthy G. (1999) . Event related potentials and func-tional MRI : a comparison of localization in sensory,


perceptual and cognitive tasks . Electroencephalogr Clin Neurophysiol Suppl 49:3-12 .

McJury M, Stewart RW, Crawford D, Toma E. (1997) . The use of active noise control (ANC) to reduce acoustic noise generated during MRI scanning: some initial results. Magn Res Imaging 15:319-322 .

McKinnon GC . (1993) . Ultrafast interleaved gradient-echo-planar imaging on a standard scanner. Magn Reson Med 30:609-616 .

Melcher JR, Talavage TM, Harms MP. (1999a). Func-tional MRI of the auditory system. In : Moonen CTW, Bandettini P, eds. Functional MRI. Berlin : Springer-Verlag, 393-406.

Melcher JR, Sigalovsky I, Levine RA . (1999b). Tinnitus-related fMRI activation patterns in human auditory nuclei . In : Hazell J, ed . Proceedings of the Sixth Inter-national Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 166-170.

Melcher JR, Sigalovsky I, Guinan JJ Jr, Levine RA. (2000). Lateralized tinnitus studied with functional magnetic resonance imaging: abnormal inferior colliculus activa-tion . JNeurophysiol 83:1058-1072 .

Menon RS, Ogawa S, Tank DW Ugurbil K. (1993).4 Tesla gradient recalled echo characteristics of photic stimula-tion induced signal changes in the human primary visual cortex . Magn Reson Med 30:380-386 .

Menon RS, Ogawa S, Hu X, Strupp JP, Anderson P, Ugur-bil K. (1995) . BOLD functional MRI at 4 Tesla including a capillary bed contribution echo-planar imaging corre-lates with previous optical imaging using intrinsic signals. Magn Reson Med 33:453-459 .

Meyerhoff WL . (1982) . Discussion of: Kauer JS, Nemitz JW, Saski CT. (1982) . Tinnitus aurium : fact . . . or fancy? Laryngoscope 92:1407.

Millen SJ, Haughton VM, Yetkin S. (1995) . Functional magnetic resonance imaging of central auditory path-way following speech and pure tone stimuli. Laryngoscope 105:1305-1310 .

Mirz F, Ovesen T, Ishizu K, Johannsen P, Madsen S, Gjedde A, Pedersen CB . (1999a). Stimulus-dependent central processing of auditory stimuli. Scand Audiol 28 :161-169 .

Mirz F, Pedersen CB, Ovesen T, Madsen S, Gjedde A. (1999b). Positron emission tomography of cortical cen-ters of tinnitus . Hear Res 134:133-144 .

Mohr CM, King WM, Freeman AJ, Briggs RW Leonard CM. (1999) . Influence of speech stimuli intensity on the activation of auditory cortex investigated with functional magnetic resonance imaging . J Acoust Soc Am 105:2738-2745 .

MollerAR . (1997) . Similarities between chronic pain and tinnitus . Am J Otol 18:577-585 .

Moller AR . (1999) . Pathophysiology of severe tinnitus and chronic pain. In: Hazell J, ed. Proceedings of the Sixth International Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 26-31 .

Moonen CTW, Liu G, van Gelderen P, Sobering G. (1992) . A fast gradient-recalled MRI technique with increased

269


sensitivity to dynamic susceptibility effects . Magn Reson Med 26:184-189 .

Moonen CTW, van Gelderen P (1997). Optimal efficiency of 3D and 2D BOLD gradient echo fMRI methods. In : Moonen CTW Bandettini PA, eds. Functional MRI. New York : Springer-Verlag, 161-172.

Pickens DR . (1988). Gating: cardiac and respiratory. In : Partain CL, Price RR, Patton JA, Kulkarni MV, James AE Jr, eds. Magnetic Resonance Imaging. Vol. 2. 2nd Ed . Philadelphia : WB Saunders, 1399-1410.

Porro CA, Cettolo V, Francescato MP, Baraldi P. (1998) . Temporal and intensity coding of pain in human cortex . J Neurophysiol 80:3312-3320 .

Moonen CTW, van Zijl PC, Frank JA, Le Bihan D, Becker ED. (1990) . Functional magnetic resonance imaging in medicine and physiology. Science 250:53-61 .

Morris GL, Mueller WM, Yetkin FZ Haughton VM, Ham-meke TA, Swanson S, Rao SM, Jesmanowicz A, Estkowski LD, Bandettini PA, Wong C, Hyde JS . (1994) . Functional magnetic resonance imaging in partial epilepsy. Epilep-sia 35:1194-1198 .

Moseley ME, de CrespignyA, Spielman DM. (1996) . Mag-netic resonance imaging of human brain function . Surg Neurol 45:385-391 .

Nobre AC, Plunkett K. (1997) . The neural system of lan-guage : structure and development . Curr Opin Neurobiol 7:262-268 .

Ogawa S, Lee TM, KyAR, Tank DW (1990) . Brain mag-netic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 87 :9868-9872 .

Ogawa S, Tank DW Menon R, Ellerman J, Kim S, Merkle H, Ugurbil K. (1992). Intrinsic signal changes accompa-nying sensory stimulation: functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci U S A 89:5951-5955 .

Ogawa S, Menon RS, Kim S-G, Ugurbil K . (1998). On the characteristics of functional magnetic resonance imag-ing of the brain . Ann Rev Biophys Biomol Struct 27:447-474.

Opitz B, MecklingerA, von Cramon DY, Kruggel F. (1999) . Combining electrophysiological and hemodynamic mea-sures of the auditory oddball. Psychophysiol 36:142-147 .

Oshiro Y, Fuijita N, Tanaka H, Hirabuki N, Nakamura H, Yoshiya 1. (1998) . Functional mapping of pain-related activation with echo-planar MRI: significance of the SII-insular region. Neuroreport 13:2285-2289.

Oestreicher E, Willoch F, Lamm K, Arnold W Barten-stein P. (1999a). Changes in metabolic glucose rate in the central nervous system induced tinnitus . Assoc Res Oto-larygol 22:42.

Ostreicher E, Willoch F, Lamm K, Arnold W Bartenstein P. (1999b). Changes in metabolic glucose rate in the cen-tral nervous system induced by tinnitus. In: Hazell J, ed . Proceedings of the Sixth International Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 171-174.

Pandya DN. (1995) . Anatomy of the auditory cortex. Rev Neurol (Paris) 151:486-494.

Paulesu E, Harrison J, Baron-Cohen S, Watson JDG, Goldstein L, Heather RSJ, Frackowiak RSJ, Frith CD . (1995) . The physiology of coloured hearing: a PET activa-tion study of colour-word synesthesia . Brain 118:661-676 .

Phelps ME, Kuhl DE, Mazziotta JC . (1981) . Metabolic mapping of the brain's response to visual stimulation . Studies in humans . Science 211:1445-1448 .

Posner MI . (1994) . Attention : the mechanisms of con-sciousness . Proc Natl Acad Sci U S A 91:7398-7403 .

Pugh KR, Shaywitz BA, Shaywitz SE, Fulbright RK, Byrd D, Skudlarski O, Shankweiler DP, Katz L, Con-stable RT, Fletcher J, Lacadie C, Marchione K, Gore JC . (1996) . Auditory selective attention: an fMRI investiga-tion . Neuroimage 4:159-173 .

Purcell EM, Torrey HC, Pound RV. (1946) . Resonance absorption by nuclear magnetic moment in a solid. Phys-iol Rev 69:37-38 .

Rabe-Hesketh S, Bullmore ET, Brammer MJ. (1997) . The analysis of functional magnetic resonance images . Stat Methods Med Res 6 :215-237 .

Rademacher J, Caviness VS Jr, Steinmetz H, Galaburda AM. (1993). Topographical variation of the human pri-mary cortices : implications for neuroimaging, brain mapping, and neurobiology. Cereb Cortex 3:313-329 .

Rae C, Lee MA, Dixon RM, Blamire AM, Thompson CH, Styles P, Talcott J, Richardson AJ, Stein JF. (1998) . Meta-bolic abnormalities in developmental dyslexia detected by 1H magnetic resonance imaging. Lancet 351:1849-1852.

Raichle ME . (1998) . Behind the scenes of functional brain imaging. A historical and physiological perspective . Proc Natl Acad Sci U S A 95:765-772 .

Raichle ME, Martin WRW Herscovitch P, Mintun MA, Markham J. (1983) . Brain blood flow measured within intravenous H2150. 11 . Implementation and validation . J Nucl Med 24:790-798 .

Rao SM, Bandettini PA, Binder JR, Bobholz JA, Ham-meke TA, Stein EA, Hyde JS . (1996) . Relationship between finger movement rate and functional magnetic resonance signal change in human primary motor cortex. J Cereb Blood Flow Metab 16:1250-1254 .

Rauschecker JP. (1998a). Cortical processing of complex sounds . Curr Opin Neurobiol 8:516-521 .

Rauschecker JP (1998b). Parallel processing in the audi-tory cortex of primates. Audiol Neurootol 3:86-103.

Ravicz ME, Melcher JR . (1998) . Reducing imager-generated acoustic noise at the ear during functional magnetic resonance imaging (fMRI) : passive attenua-tion . Assoc Res Otolaryngol 21:208 .

Richards TL, Gates GA, Gardner JC, Merrill T, Hayes CE, Panagiotides H, Serafini S, Rubel EW (1997). Func-tional MR spectroscopy of the auditory cortex in healthy subjects and patients with sudden hearing loss . Am J Neuroradiol 18:611-620 .

Richards TL, Dager SR, Corina D, Sarafini S, Heide AC, Steury K, Strauss W, Hayes CE, Abbott RD, Craft S, Shaw D, Posse S, Beringer VW. (1999) . Dyslexic children have abnormal brain lactate response to reading-related language tasks. Am J Neuroradiol 20:1393-1398 .

270

Richardson AJ, Cox IJ, Sargentoni J, Puri BK . (1997) . Abnormal cerebral phospholipid metabolism in dyslexia indicated by phosphorus-31 magnetic resonance spec-troscopy. Nucl Magn Reson Biomed 10:309-314 .

Righini A, de Divitis 0, Prinster A, Spagnoli D, Appollo-nio 1, Bello L Scifo P, Tomei G, Villeni R, Fazio F, Leonardi M. (1996) . Functional MRI: primary motor cortex local-ization in patients with brain tumors . J Comput Assist Tomogr 20:702-708 .

Rivier F, Clarke S. (1997) . Cytochrome oxidase, acetyl-cholinesterase, and NADPH-Diaphorase staining for human supratemporal and insular cortex : evidence for multiple auditory areas. Neuroimage 6:288-304 .

Robson MD, Dorosz JL, Gore JC . (1998) . Measures of the temporal fMRI response of the human auditory cortex to trains of tones. Neuroimage 7:185-198 .

Romanski LM, Bates JF, Goldman-Rakic PS . (1999a). Auditory belt and parabelt projections to the prefrontal cortex in the rhesus monkey. J Comp Neurol 403:141-157.

Romanski LM, Tian B, Fritz J, Mishkin M, Goldman-Rakic PS, Rauschecker JP. (1999b). Dual streams of auditory afferents target multiple domains in the pri-mate prefrontal cortex . Nat Neurosci 2 :1131-1136 .

Rosen BR, Buckner RL, Dale AM. (1998) . Event-related functional MRI: past, present and future . Proc Natl Acad Sci US A 95 :773-780 .

Rossini PM, Caltagirone C Castriota-Scanderbeg A, Cicinelli P, Del Gratta C, Demartin M, Pizzella V, Tra-versa R, Romani GL . (1998) . Hand motor cortical area reorganization in stroke : a study with fMRI, MEG and TCS maps . Neuroreport 9 :2141-2146 .

Salvi RJ, Lockwood AH, Coad ML, Wack DS, Burkhard R, Arnold S, Galantowicz P. (1999) . Positron emission tomography identifies neuroanatomical sites associated with tinnitus modified by oral-facial and eye-movements. In: Hazell J, ed . Proceedings of the Sixth International Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 175-180.

Sasaki CT, Kauer JS, Babitz L. (1980) . Differential 11"C12-deoxyglucose uptake after deafferentation of the mammalian auditory pathway-a model for examining tinnitus . Brain Res 194:511-516 .

Savoy RL, Ravicz ME, Gollub R . (1999) . The psy-chophysiological laboratory in the magnet : stimulus delivery, response recording, and safety. In : Moonen CTW, Bandettini PA, eds. Functional MRI. New York : Springer-Verlag 347-365.

Schlosser MJ, Luby M, Spencer DD, Awad IA, McCarthy G. (1999) . Comparative localization of auditory compre-hension by using functional magnetic resonance imaging and cortical stimulation. JNeurosurg 91:626-635 .

Seidman LJ, Breiter HC Goodman JM, Goldstein JM, Woodruff PW O'Craven K, Savoy R, Tsuang MT, Rosen BR . (1998) . A functional magnetic resonance imaging study of auditory vigilance with low and high informa-tion processing demands. Neuropsychology 12:505-518 .

Serano MI, Dale AM, Reppas JB, Kwong KK, Belliveau JW, Brady TJ, Rosen BR, Tootell RB. (1995,) . Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. Science 268:889-892 .


Shaywitz BA, Shaywitz SA, Pugh KR, Constable RT, Skudlarski P, Fulbright RK, Bronen RA, Fletcher JM, Shankweiler DP, Katz L, Gore JC . (1995) . Sex differences in the functional organization of the brain for language . Nature 373:607-609 .

Shaywitz SE, Shaywitz BA, Pugh KR, Fulbright RK, Con-stable RT, Mencil WE, Shankweiler D, Liberman A, Skudlarski P, Fletcher JM, Katz L, Marchione KE, Lacadie C Gatenby C, Gore JC . (1998) . Functional disruption in the organization of the brain for reading in dyslexia . Proc Natl Acad Sci U S A 95 :2636-2641 .

Shulman RG, Blamire AM, Rothman DL, McCarthy G. (1993) . Nuclear magnetic resonance imaging and spec-troscopy of human brain function . Proc Natl Acad Sci USA 90 :3127-3133 .

Sigalovsky IS, Melcher JR, Levine RA . (1999) . Growth of f MRI activation with stimulus level in the inferior col-liculi : implications for understanding tinnitus-related abnormalities. In : Hazell J, ed . Proceedings of the Sixth International Tinnitus Seminar. London : Tinnitus and Hyperacusis Centre, 317-322.

Silbersweig DA, Stern E. (1996) . Functional neuroimaging of hallucinations in schizophrenia: toward an integra-tion of bottom-up and top-down approaches . Mol Psychiatry 1:367-375 .

Silbersweig DA, Stern E, Frith C, Cahill C, Holmes A, Grootooknk S, Seaward J, McKenna P, Chua SE, Schnorr L, Jones T, Frackowiak RST. (1995) . A functional neu-roanatomy of hallucinations in schizophrenia. Nature 378:176-179 .

Sobel N, Prabhakaran V, Zhao Z, Desmond JE, Glover GH, Sullivan EV, Gabrieli JDE. (2000) . Time course of odorant-induced activation in the human primary olfac-tory cortex . J Neurophysiol 83:537-551 .

Springer JA, Binder JR, Hammecke TA, Swanson SJ, Frost JA, Bellgowan PS, Brewer CC, Perry HM, Morris GL, Mueller WM. (1999) . Language dominance in neu-rologically normal and epilepsy subjects : a functional MRI study. Brain 122:2033-2046 .

Strainer JC, Ulmer JL, Yetkin FZ, Houghton VM, Daniels DL, Millen SJ . (1997) . Functional MR of the primary auditory cortex : an analysis of pure tone activation and tone discrimination. Am J Neuroradiol 18:601-610 .

Syka J, Hajek M, Dezortova M, Burian M, Jilek M. (1998) . In-vivo quantification of metabolites in the human audi-tory cortex with proton magnetic resonance spectroscopy. Assoc Res Otolaryngol 21:35.

Szechtman H, Woody E, Bowers KS, Nahmias C. (1998) . Where the imaginal appears real : a positron emission tomography study of auditory hallucinations . Proc Natl Acad Sci U S A 95 :1956-1960 .

Talavage TM, Benson RR, Galburda AM, Rosen BR . (1996) . Evidence for multiple tonotopic fields in human auditory cortex . Proc Int Soc Magn Reson Med 4:1842.

Talavage TM, Ledden PJ, Sereno MI, Rosen BR, Dale AM . (1997) . Multiple phase-encoded tonotopic maps in human auditory cortex . Neuroimage 5:88.

Talavage TM, Edmister WB, Ledden PJ, Weiskoff RM . (1999) . Quantitative assessment of auditory cortex responses induced by imager acoustic noise. Hum Brain Map 7:79-88 .

271


Thulborn KR, Waterton JC, Matthews PH, Radda GK. (1982) . Oxygenation dependence of the transverse relax-ation rate of water protons in whole-blood at high field . Biochim Biophys Acta 714:265-270 .

Thulborn KR, Carpenter PA, Just MA. (1999) . Plasticity of language-related brain function during functional recovery from stroke . Stroke 30:749-754 .

1bgaAW, Mazziotta JC . (1996) . Brain Mapping: The Meth-ods. San Diego: Academic Press.

Voyvodic JT. (1999) . Real-time /MRI paradigm control, physiology, and behavior combined with near real-time statistical analysis . Neuroimage 10:91-106 .

Wall M, Rosenberg M, Richardson D. (1987) . Gaze-evoked tinnitus . Neurology 37:1034-1036 .

Warrach S, Ives JR, Schlaug G, Patel MR, Darby DG, Thangoraj V, Edelman RR, Schomer DL . (1996) . EEG-triggered echo-planar functional MRI in epilepsy . Neurology 47:89-93 .

Toni I, Krams M, Turner R, Passingham RE . (1998) . The time course of changes during motor sequence learning : a whole-brain /MRI study. Neuroimage 8:50-61 .

Tonndorf J. (1987) . Analogy between tinnitus and pain : a suggestion for a physiological basis of chronic tinnitus . Hear Res 28:271-275 .

Treede RD, Kenshalo DR, Gracely RH, Jones AK. (1999) . The cortical representation of pain . Pain 79:105-111 .

Truy E. (1999) . Neuro-functional imaging and profound deafness . Int J Pediatr Otorhinolaryngol 47:131-136 .

Turner R, Jezzard P, Wen H, Kwong KK, LeBihan D, Zef-firo T, Balaban R. (1993) . Functional mapping of the human visual cortex at 4 and 1.5 Tesla using deoxy-genation contrast EPI. Magn Reson Med 29:277-279.

Turner R, Howseman A, Rees GE, Josephs O, Friston K. (1998) . Functional magnetic resonance imaging of the human brain : data acquisition and analysis . Exp Brain Res 123:5-12 .

Ulualp SO, Biswal BB, Yetkin FZ, Kidder TM . (1998) . Functional magnetic resonance imaging of auditory cortex in children . Laryngoscope 108:1782-1786.

van Gelderen P, Ramsey N, Liu G, Duyn J, Frank J, Wein-berger DR, Moonen CTW. (1995) . Three-dimensional functional magnetic resonance imaging of human brain on a clinical 1 .5T scanner. Proc Natl Acad Sci U S A 92:6906-6910 .

Vogt BA, Finch DM, Olson CR . (1992) . Functional het-erogeneity in cingulate cortex : the anterior executive and posterior evaluative regions . Cereb Cortex 2:435-443 .

WeisskoffRM, Zuo CS, Boxerman JL, Rosen BR. (1994). Microscopic susceptibility variation and transverse relax-ation : theory and experiment . Magn Reson Med 31 : 601-610 .

Wilichowski E, Pouwels PJW, Frahm J, Hanefeld F. (1999). Quantitative proton magnetic resonance spectroscopy of cerebral metabolic disturbances in patients with MELAS. Neuropediatrics 30:256-263 .

Williams DS, Detre JA, Leigh JS, Koretsky AP. (1992) . Magnetic resonance imaging of perfusion using spin inver-sion of arterial water. Proc Natl Acad Sci U S A 89 : 212-216.

Woodruff PW, Wright IC, Bullmore ET, Brammer M, Howard RJ, Williams SC, Shapleske J, Rossell S, David AS, McGuire PK, Murray RM. (1997) . Auditory halluci-nations and the temporal cortical response to speech in schizophrenia: a functional magnetic resonance imaging study. Am J Psychiatry 154:1676-1682 .

Xiong J, Gao JN, Lancaster JL, Fox PT. (1996) . Assess-ment and optimization of functional MRI analyses. Hum Brain Map 4:153-167.

Yablonsky DA, Haacke EM. (1994) . Theory of NMR signal behavior in magnetically inhomogeneous tissues : the static dephasing regime . Magn Reson Med 32:749-763 .

Yoo SS, Guttmann CR, Zhao L, Panych LP (1999) . Real-time adaptive functional MRI. Neuroimage 10:596-606 .

Yoshiura T, Zhong J, Shibata DK, Kwok WE, Shier DA, Numaguchi Y (1999) . Functional MRI study of auditory and visual oddball tasks. Neuroreport 10:1683-1688 .

special tutorial: principles of functional magnetic ... · pdf fileprinciples of functional...

Documents