Abstract
Even the healthiest older adults experience changes in cognitive and sensory function. Studies show that older adults have reduced neural responses to sensory information. However, it is well known that sensory systems do not act in isolation but function cooperatively to either enhance or suppress neural responses to individual environmental stimuli. Very little research has been dedicated to understanding how aging affects the interactions between sensory systems, especially cross‐modal deactivations or the ability of one sensory system (e.g., audition) to suppress the neural responses in another sensory system cortex (e.g., vision). Such cross‐modal interactions have been implicated in attentional shifts between sensory modalities and could account for increased distractibility in older adults. To assess age‐related changes in cross‐modal deactivations, functional MRI studies were performed in 61 adults between 18 and 80 years old during simple auditory and visual discrimination tasks. Results within visual cortex confirmed previous findings of decreased responses to visual stimuli for older adults. Age‐related changes in the visual cortical response to auditory stimuli were, however, much more complex and suggested an alteration with age in the functional interactions between the senses. Ventral visual cortical regions exhibited cross‐modal deactivations in younger but not older adults, whereas more dorsal aspects of visual cortex were suppressed in older but not younger adults. These differences in deactivation also remained after adjusting for age‐related reductions in brain volume of sensory cortex. Thus, functional differences in cortical activity between older and younger adults cannot solely be accounted for by differences in gray matter volume. Hum Brain Mapp 2009. © 2007 Wiley‐Liss, Inc.
Keywords: functional magnetic resonance imaging, cortical atrophy, vision, auditory, cross‐modal
INTRODUCTION
Both behavioral and physiological changes occur during the aging process, and the impact of these changes on daily activities can determine how successfully an individual ages. It is well known that unisensory ability declines with increasing age and that older adults typically experience increased perceptual thresholds and decreased sensory cortical activity responses [D'Esposito et al.,1999; Johnson et al.,2000; Justino et al.,2001; Levine et al.,2000; Ross et al.,1997]. Studies in the auditory and visual modalities, the most‐widely studied sensory domains, have indicated that aging changes occur not only at the level of the sensory organ (e.g., lens changes such as glaucoma and cataracts, loss of hair cells in the cochlea) but also within the neural substrates of the unisensory system. Specifically, neural substrate differences include anatomical changes, such as reduced gray matter and cortical thickness [Fjell et al.,2006; Good et al.,2001; Jernigan et al.,2001; Jernigan and Gamst,2005; Lemaitre et al.,2005; Sowell et al.,2003; Tisserand et al.,2004], as well as functional changes including decreased amplitude of electroretinograms [Li et al.,2000], visual and auditory evoked potentials [Justino et al.,2001], and differing activation patterns in positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) [D'Esposito et al.,1999; Levine et al.,2000; Ross et al.,1997]. For example, a PET study on processing visual form showed that older adults activate occipital and frontal regions, whereas young adults preferentially activate occipital‐temporal areas to solve the same task [Levine et al.,2000].
Other neural mechanisms underlying higher order cognitive processing may also play a part in sensory processing changes for older adults. For example, it has been suggested that older adults are more sensitive to distractions both within and across sensory modalities due to reduced attentional capacity. In young adults, visual stimuli increase activity in visual cortices while suppressing activity in other sensory modalities (e.g., auditory, somatosensory) [Alsius et al.,2005; Haxby et al.,1994; Kawashima et al.,1995; Laurienti et al.,2002; Macaluso et al.,2000; Shulman et al.,1997]. This cross‐modal inhibition is seen as task‐specific deactivations in fMRI that indicate a decrease in neural activity from the baseline condition. Behaviorally, this reduction in neural activity may provide a mechanism that limits distractions and interference from other sensory modalities. In fact, attending to a specific modality slows response times to stimuli presented in a nonattended sensory modality while speeding response times to stimuli presented in the attended modality [Spence and Driver,1997; Spence et al.,2001]. Slower response times could be due to the decreased neural activity within the cross‐modal sensory cortex that occurs when one does not expect a stimulus in that sensory domain. Studies have demonstrated that task‐independent deactivations associated with the default mode of brain function are reduced in older adults compared with younger adults [Lustig et al.,2003]. However, it remains unknown if there are age‐related changes in the magnitude of cross‐modal deactivations that may contribute to older adults increased distractibility.
A critical aging confound emerges when measuring functional neural activity with imaging methods, since the underlying anatomy responsible for generating the functional activity is also experiencing changes during the aging process. Aging is consistently associated with neuroanatomical differences in gray matter volume measured with voxel‐based morphometry (VBM) as well as with cortical thickness methods which indicate variable rates of thinning throughout the cortical mantle [Fjell et al.,2006; Good et al.,2001; Jernigan et al.,2001; Jernigan and Gamst,2005; Lemaitre et al.,2005; Tisserand et al.,2004]. Neither age‐related anatomical changes nor their functional impact are uniform throughout the cortex; rather, age‐related anatomical decreases in some areas are related to reduced performance, others are related to increases in performance, and still changes in other areas appear to not affect many performance measures [Colcombe et al.,2003,2004,2005; Fjell et al.,2006; Haier et al.,2005; Prvulovic et al.,2005]. PET experiments show that the application of a partial volume correction can adjust the quantitative PET signal for heterogeneous cortical atrophy [Meltzer et al.,1996; Muller‐Gartner et al.,1992]. This partial volume correction employs either a whole‐brain or a voxel‐by‐voxel division of the PET image by tissue volume (e.g., gray matter or CSF) and has successfully controlled for age‐related differences in quantitative PET measurements reducing the differences seen between older and younger adults [Alavi et al.,1993; Tanna et al.,1991; Yanase et al.,2005]. Methods to perform similar corrections for fMRI data have only recently been introduced [Casanova et al.,2007].
To evaluate the functional impact of aging on sensory processing, simple auditory and visual discrimination tasks were performed and imaged using fMRI in a group of 18–80‐year‐old participants. Additional analyses of the observed blood‐oxygenation level dependent (BOLD) activity for the auditory and visual perception tasks were performed to control for age‐related anatomical differences on a voxel to voxel basis. These analyses utilized the Biological Parametric Mapping (BPM) toolbox recently developed within our laboratory [Casanova et al.,2007]. The BPM analysis provides a means of correcting BOLD signal change within each voxel for structural differences (i.e., gray matter volume) that may impact the measured BOLD response within sensory cortices. We hypothesize that much like activation differences, deactivation to cross‐modal sensory stimuli will show age‐related differences. Specifically, older adults will show reduced amounts of cross‐modal deactivation in nontask cortex relative to younger adults for both the auditory and visual tasks. Further, when anatomical differences are controlled in the BPM analysis, functional differences will remain significant for both activations and deactivations.
METHODS
Subjects
This study included 61 volunteers between the ages of 18 and 80 years (mean age, 49; 34 females; 57 right handed), who were compensated for their time. All subjects were in normal health with normal hearing and normal or corrected to normal vision. Following information on the scanning procedure, which was approved by and in accordance with the Wake Forest University School of Medicine Institutional Review Board, subjects gave written informed consent in accordance with the Declaration of Helsinki. For certain analyses age was considered as a categorical value so that differences between younger and older adults could be more easily compared to existing literature that uses age as a categorical variable—young, 18–38 (n = 20); middle‐aged, 39–64 (n = 21); and older, 65–80 years (n = 20). Five additional subjects whose functional data were unusable due to technical difficulties (1 young, 1 middle‐aged, and 3 older) were included only within the VBM analysis.
To assess for normal health, all subjects received an extensive medical history and medication review. Subjects stating a history of or taking medications for epilepsy, stroke, Parkinson's disease, Alzheimer's disease, attention deficit‐hyperactivity disorder, diabetes, serious vision problems (other than refractive problems or astigmatism), hearing problems, head injury with loss of consciousness, brain surgery, and psychiatric disorders other than treated depression were excluded. In addition, participants were screened for alcoholism using the alcohol use disorders identification test (AUDIT), untreated depression using the Center for Epidemiological Studies Depression scale (CES‐D), hand preference using the Edinburgh Survey, and color blindness using the Concise Edition of Ishihara's Test for Colour‐Blindness (Kanehara, Tokyo, Japan). Subjects with moderate hearing loss were excluded from the study by threshold hearing tests at 1,000–2,000 Hz assessed with an audioscope (Welch Allyn, Skaneateles Falls, NY; 40 dB loss), a digital audiometer (Digital Recordings, Halifax, Nova Scotia; 50 dB loss), or audiologist assessment by the Department of Hearing and Speech at Wake Forest University School of Medicine (50 dB loss). A modified Snell visual acuity examination was performed for each eye (with corrective lenses if applicable), and subjects worse than 20/40 were excluded. The modified mini‐mental state examination was performed to assess cognitive function, and performance less than two standard deviations from the mean for age and education level was used to exclude participants [Crum et al.,1993]. Demographic and behavioral results for the age groups are summarized in Table I.
Table I.
Demographic and behavioral results
Age | Age range | Education | MMSE | Females | Right‐handed | Visual % response | Auditory % response | Visual RT | Auditory RT | |
---|---|---|---|---|---|---|---|---|---|---|
Young | 29.2 (1.33) | 18–37 | 15.4 (0.43) | 29.1 (0.46) | 12 | 17 | 76.30 (1.68) | 97.39 (0.60) | 476.1 (11.78) | 323.7 (10.34) |
Middle | 47.6 (1.66) | 39–63 | 14.8 (0.42) | 29.1 (0.45) | 12 | 20 | 74.53 (1.64) | 98.40 (0.58) | 476.4 (11.50) | 339.7 (10.09) |
Older | 71.3 (1.03) | 65–80 | 15.4 (0.43) | 27.8 (0.46) | 10 | 20 | 78.40 (1.68) | 97.60 (0.60) | 529.2 (11.78) | 335.1 (10.34) |
Effect of age | F 2,58 = 232.5, *P < 0.001 | F < 1, P = n.s. | F 2,58 = 2.70, P < 0.08 | F 2,58 = 1.35, P = n.s. | F < 1, P = n.s. | F 2,58 = 6.78, *P < 0.005 | F < 1, P = n.s. |
Because of presentation limits in the E‐prime program, the visual task had greater durations where responses were unable to be recorded, which resulted in lower response percentages for all age groups. Mean differences due to age on the MMSE (mini‐mental status exam) are consistent with published normative data (see Crum et al.,1993). RT: response time. Standard error of the mean is reported in parentheses.
Stimuli
Subjects performed two sets of visual and auditory block paradigms. Stimuli were delivered via MR‐compatible display goggles and headphones (Resonance Technology, Northridge, CA) with stimulus presentation and timing controlled by E‐prime software (Psychology Software Tools, Pittsburgh, PA). For both paradigms, subjects were asked to fixate on a gray cross in the center of the display and respond with a button press when they detected the target stimulus (23 occurrences per run). An infrared eye tracker within the display goggles was used to ensure that participants kept their eyes open during the entire run. Both paradigms consisted of four 30‐s task periods and four 30‐s rest or baseline periods (only fixation cross displayed). In the task portion of the auditory paradigm, 2‐Hz bursts of white noise alternated with silence. Subjects had to identify a 500‐Hz tone randomly embedded in the white noise. Prior to performing the task stimulus, volume was adjusted to ensure that subjects could hear the target tone above scanner noise. For the task portion of the Visual paradigm, a black and white alternating checkerboard flashed at 2 Hz with the randomly occurring target being a 250 ms blurred version of the checkerboard display. Paradigm runs lasting 260 s were repeated twice for a total of eight task periods and eight rest periods per auditory and visual paradigm. The first 20 s displayed the fixation cross and were discarded from further analysis.
Image Acquisition
Brain activity was assessed by examining BOLD signal change [Ogawa and Lee,1990] in a series of fMRI scans performed on a 1.5‐T General Electric twin‐speed LX scanner with a birdcage head coil (GE Medical Systems, Milwaukee, WI). Functional scans were performed using a gradient echo echo‐planar imaging (EPI) protocol [Mansfield,1977] with the following parameters: 24 cm field of view, 64 × 64 acquisition matrix, 28.5‐mm‐thick slices with no gap, 40‐ms TE, 2,500‐ms TR, and an in‐plane resolution of 3.75 × 3.75 mm2, frequency direction anterior to posterior. Anatomic T1 weighted images were acquired with a 128‐slice spoiled gradient inversion recovery (3DSPGR‐IR) protocol (TE/TI 1.9/600; flip angle 20°; 256 × 256 acquisition matrix, 1.5 mm slice thickness with no gap; and an in‐plane resolution of 0.9375 × 0.9375 mm2). All images are presented in neurological format with the right hemisphere on the right side of the image.
Structural Image Analysis: VBM
Optimized VBM analysis was performed with SPM2, as described previously [Good et al.,2001; Mechelli et al.,2005]. Briefly, a study‐specific template and prior probability maps from gray matter, white matter, and CSF were created. To create the study‐specific template, each subject's data were aligned to the MNI template using affine transformation and segmented into gray matter, white matter, and CSF tissue compartments based on SPM2 prior probability maps. This yielded three‐segmented tissue images in native space for each subject. For the template, each subject's segmented gray matter image was then normalized to the SPM2 gray matter prior probability map. Next, parameters from this normalization were applied back to the participant's T1 image in native space. Finally, these normalized T1 images were averaged across all 64 subjects to yield the study‐specific template and prior probability maps.
These study‐specific images were used in the optimized VBM analysis to perform a normalization procedure on each subject's data yielding segmented gray matter images. These gray matter images were multiplied by the Jacobian determinants derived from the normalization procedure to yield modulated gray matter volume maps. To exclude regions of very low probability of being gray matter, an absolute threshold of 0.15 was used. All images were resampled to 1 mm3 and smoothed using a Gaussian kernel of 8 mm3. Each subject's total intracranial volume (ICV) was calculated by summing the gray, white, and CSF segmentations of the modulated VBM data. Using total ICV as a covariate, two different random‐effects analyses were performed to compare age‐related changes in gray matter volume from our study population with those changes published in the literature—(1) regression of gray matter volume across age and (2) ANCOVA of gray matter volume between age groups (i.e., young/middle, young/older, middle/older). To control for multiple comparisons, a Family Wise Error (FWE) correction of P < 0.05 was applied based on the random field theory [Worsley et al.,1996].
Functional Image Analyses
Statistical parametric maps (SPMs) were generated using SPM99 (Wellcome Department of Imaging Neuroscience, London, UK). Data were spatially realigned using a six‐parameter “rigid‐body” transform to the first slice of the run correcting for within run motion. Functional images were spatially normalized using the 16‐parameter affine transformation in combination with the parameters from the normalization of the anatomic image to Montreal Neurological Institute (MNI) space [Maldjian et al.,1997]. Further, signal within the images was normalized using global scaling correction to remove confounds of whole‐brain signal changes in our regional analyses [Aguirre et al.,1998; Laurienti,2004; Laurienti et al.,2002]. Images from each run were then realigned to the first slice of the first run to correct for motion between runs. An additional normalization of the functional data sets to the SPM99 EPI template was then performed, and the data were resampled to a 4 × 4 × 5 mm3 using sinc interpolation. Finally, data sets were smoothed using an 8 × 8 × 10 mm3 full‐width half‐maximum (FWHM) Gaussian kernel.
A multiple regression design matrix with a box‐car design convolved with the hemodynamic response function and including an explicitly modeled baseline condition (rest block) was constructed for each run. Multiple paradigm runs were combined using a fixed‐effects analysis for each subject, resulting in one “con image” per subject representing visual task vs. baseline and in addition one image for auditory task vs. baseline. Random effects analyses were used to perform all group analyses. To restrict comparisons to the unisensory cortices, masks were created for auditory and visual responsive cortices. The masks were generated by including all subjects (n = 61) in random‐effects analyses for each sensory modality and a binary mask was created from activated regions (P = 0.001 uncorrected) within auditory and visual Brodmann areas as defined by the Wake Forest University Pickatlas tool [Lancaster et al.,2000; Maldjian et al.,2003]. Two different random‐effects analyses were performed within the auditory and visual ROI to evaluate age‐related changes in unisensory activation and deactivation—(1) regression of age against BOLD activity and (2) two‐sample T‐tests of BOLD activity between age groups (i.e., young/middle, young/older, middle/older). For both paradigms task vs. baseline comparison, all subjects from each age group were combined into a group analysis to characterize the main effect of sensory stimulation in each age group. These analyses were used to clarify the direction (activation or deactivation) of group differences. Further post‐hoc analyses included individual response times for both the auditory and visual tasks as regression variables to investigate whether individual response time differences contributed to age‐related fMRI activity differences within sensory cortices. The Wake Forest University Pickatlas tool was used to label the Brodmann area for coordinates of peak voxels in clusters identified in any analysis [Lancaster et al.,2000; Maldjian et al.,2003]. To control all analyses for multiple comparisons, an extent correction was applied within each sensory ROI [Cao and Worsley,2001; Hayasaka and Nichols,2003].
BPM [Casanova et al.,2007] was used to perform regression and ANCOVA analyses where voxel‐specific gray matter volume was included in the regression as a nuisance variable using the general linear model (GLM). In other words, BPM provides a mechanism to combine information from multiple imaging modalities for each participant and perform a unique regression in every brain voxel that adjusts the BOLD signal for gray matter volume. This process assumes that data from the different imaging modalities are within a common space and are independent. The above fMRI analyses were repeated using BPM to evaluate age‐related changes in unisensory activation and deactivation while including each subject's modulated gray matter VBM and total ICV applied as variables‐of‐no‐interest. Each subject's original modulated gray matter probability map was resampled to a 4 × 4 × 5 mm3 voxel size and smoothed with an 8 × 8 × 10 mm3 FWHM Gaussian kernel to match the parameters of the fMRI image. The dependent variable in BPM multiple regressions were the imaging variable of primary interest, in this case the auditory or visual fMRI paradigm. Independent variables included one other imaging modality (modulated gray matter maps from the VBM analysis) and two nonimaging covariates (age and ICV). Maps examining the relationship between fMRI activity and age included VBM and ICV as covariates in the regression and weighted age. Similarly, the BPM ANCOVA analysis employed the GLM to compare age‐groups with the dependent variable of auditory or visual fMRI, imaging covariate of modulated VBM maps, and nonimaging covariate of ICV. Significant BPM findings would represent areas with age‐related BOLD differences that cannot be accounted for by local age‐related gray matter volume changes.
An additional BPM analysis was also performed to investigate the impact of loss of cortical gray matter in auditory cortex to age‐related differences in deactivation of visual cortex during the auditory task. An average amount of gray matter volume was calculated for each individual within the auditory responsive ROI to be used as a nonimaging covariate along with age, modulated VBM (local voxel‐by‐voxel gray matter measures), and ICV in a BPM regression of visual cortical response to the auditory task (i.e., deactivation differences) and the BPM ANCOVA between young and older adults.
RESULTS
Behavior
Regarding age differences in response percentage and/or speed for the auditory and visual tasks, the only significant age‐related difference occurred on the response times for the visual task (see Table I). Older adults were significantly slowed in their response relative to young and middle‐aged adults. Post‐hoc analyses assessing the impact of behavioral differences on gray matter loss and BOLD activity showed no significant effects or interaction with age on any of the comparisons summarized below.
Voxel‐Based Morphometry
Replicating previously published findings [Good et al.,2001; Hayasaka and Nichols,2003; Liu et al.,2003; Pruessner et al.,2001; Sowell et al.,2004; Tisserand et al.,2004], the VBM correlation analysis revealed that increased age was associated with decreased gray matter volume. These regression results were similar to ANCOVA findings that demonstrated the greatest differences occurred between young and older adults. In particular, young adults had greater gray matter volume than older adults predominately in bilateral temporal areas near the Sylvian fissure, insula, and medial areas including the caudate and thalamus (see Table II for significant maxima locations). Other areas in frontal, parietal, and occipital cortices showed diffuse areas with greater gray matter in young compared with older adults. When compared with the older group, the middle‐aged group also had areas of greater gray matter volume predominately near the left Sylvian fissure as well as along the midline near the thalamus and ventricles, though these areas were reduced in extent relative to the difference between young and older adults. Older adults did not show increased gray matter relative to young or middle‐aged adults. Between young and middle‐aged adults, one area within right middle frontal gyrus (BA9, local maxima at MNI coordinate 3, 48, 29) showed greater gray matter volume in younger adults.
Table II.
Summary of local maxima for the ANCOVA VBM results
Labels | Cluster local maxima of peak voxel(s) | Brodmann areas involved | |
---|---|---|---|
Young greater than older adults | Bilateral Sylvian fissure: Insula, superior and transverse temporal gyri | 46, 2, −7/−44, −2, −2 | 13, 22, 41, 42 |
Thalamus/putamen | 1, −3, 11/−20, 6, 1 | N/A | |
Bilateral pre‐ and postcentral gyri | 50, −33, 58/−47, −14, 46 | 2, 3, 4 | |
Bilateral frontal areas: Superior, middle, and inferior frontal gyri | 31, 29, 52/−33, 57, 19 | 6, 8, 9, 10, 45 | |
Bilateral interparietal lobule | 44, −49, 55/−41, −57, 50 | 40 | |
Occipital areas: Left superior occipital and fusiform gyri, right precuneus | 18, −54, 31/−34, −93, 23, and −56, −56, −29 | 19, 37 | |
Young greater than middle‐aged adults | Right middle frontal gyrus | 3, 48, 29 | 9 |
Middle‐aged greater than older adults | Bilateral Sylvian fissure: Insula, superior and transverse temporal gyri | 39, −18, 19/−40, −11, −16 | 13, 22, 41 |
Thalamus | −1, −16, −8 | N/A | |
Bilateral caudate\putamen | 6, 15, 1/−7, 10, 6\22, 4, −5/−21, 3, −6 | N/A | |
Bilateral precentral gyri | 44, −13, 45/−47, −15, 45 | 4, 6 | |
Bilateral frontal areas: Superior, middle, and inferior frontal gyri | 37, 23, 5/−17, 43, 44 | 6, 8, 9, 10, 45 | |
Left interparietal lobule | −40, 40, 28 | ∼40 | |
Left inferior occipital gyrus | −42, −82, −9 | ∼19 |
FWE corrected for multiple comparisons; ICV entered as the covariate.
Since visual cortex was isolated for the fMRI analyses, ROI analysis of the VBM results for visual cortex was performed and indicated that younger adults had significantly more gray matter in several areas of visual cortex (see Fig. 1). These included bilateral areas of BA17 and superior portions of BA19. In addition, middle‐aged adults had significantly more gray matter in right BA 17 and left superior BA19 than older adults. There were no significant differences between young and middle‐aged adults in gray matter in the visual cortex.
Figure 1.
Voxel‐based morphometry (VBM) results within visual cortex: Middle‐aged adults had greater gray matter volume than older adults in right BA17 and left superior BA18. When young and older adults were compared, these areas remained significant and expanded to show bilateral differences in both BA17 and 19. These were identified within a ROI of visual responsive cortex and used intracranial volume as a covariate and a FWE correction (P < 0.05) for multiple comparisons.
Functional Magnetic Resonance Imaging
Within each age group, activation was noted in the sensory cortex matching the stimulation paradigm (e.g., auditory paradigm: activation in superior aspects of temporal lobe) and deactivation in the sensory cortex not matching the paradigm of interest (e.g., auditory paradigm: deactivation in occipital lobe). Table III lists the T‐value (positive = activation; negative = deactivation), extent and location of the peak voxel per cluster for each age group in both the auditory and visual paradigms. More robust age‐related statistical findings were seen within visual rather than auditory cortex, and thus, age‐related functional differences within visual cortex are discussed below. Results in visual cortex of the regression analyses are listed in Table IV, and the ANOVA and BPM ANCOVA findings from visual cortex are listed in Table V. Auditory areas, while responsive in all age groups across all conditions, did not show significant age‐related differences within the regression or age‐group comparisons.
Table III.
Significant BOLD activity in sensory ROIs for the visual and auditory tasks
Visual task | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|
Visual cortex | Auditory cortex | |||||||||
T‐value | Cluster Size | Peak voxel | Label | BA | T‐value | Cluster size | Peak voxel | Label | BA | |
Young | 17.33 | 885 | −12, −92, −10 | Lingual gyrus | ∼18 | −6.48 | 39 | 55, −16, 5 | STG | ∼21/44 |
−4.11 | 6 | −64, −40, 15 | STG | ∼22 | ||||||
−3.94 | 8 | −64, −20, −5 | STG | 21 | ||||||
Middle | 13.59 | 809 | 28, −56, −15 | Fusiform gyrus | ∼19 | −7.12 | 24 | 56, −12, 0 | STG | ∼22 |
−5.58 | 10 | −60, −48, 5 | Middle TG | 21 | ||||||
Older | 15.48 | 794 | 32, −88, 10 | Middle occipital gyrus | ∼18 | −5.42 | 11 | −48, −16, 5 | STG | 22 |
−4.76 | 10 | −32, −40, 15 | Subgyral | ∼41 | ||||||
−4.57 | 7 | 44, −8, −10 | Subgyral | 21 | ||||||
−4.35 | 10 | 40, −24, 10 | Transverse TG | 41 | ||||||
Auditory task | ||||||||||
Visual cortex | Auditory cortex | |||||||||
T‐value | Cluster Size | Peak voxel | Label | BA | T‐value | Cluster Size | Peak voxel | Label | BA | |
Young | −6.14 | 47 | 28, −60, −5 | Subgyral | ∼19 | 16.67 | 233 | 60, −28, 15 | STG | 42 |
−6.08 | 41 | 32, −84, 30 | Cuneus | 19 | 5.98 | 215 | −55, −40, 20 | “Insula” | ∼22/42 | |
−5.29 | 65 | −28, −84, 30 | Cuneus | 19 | ||||||
Middle | −6.95 | 137 | 44, −80, 10 | Middle occipital gyrus | 19 | 12.50 | 234 | −52, −24, 5 | STG | ∼41 |
−6.25 | 48 | −28, −88, 25 | Cuneus | 19 | 12.26 | 248 | 56, −24, 5 | STG | ∼41 | |
−5.51 | 42 | −44, −68, −10 | Middle occipital gyrus | ∼19 | ||||||
Older | −10.34 | 181 | 36, −84, 25 | Superior occipital gyrus | ∼19 | 11.92 | 219 | −56, −28, 10 | STG | 41 |
−8.31 | 138 | −44, −76, 0 | Middle occipital gyrus | ∼19 | 11.40 | 230 | 56, −24, 10 | Transverse TG | 41 |
Table IV.
Simple regression and BPM regression results for age on visual and auditory task activity in visual cortex
Peak voxel (x,y,z) | Regression with age | BPM regression with age and VBM | ||||||
---|---|---|---|---|---|---|---|---|
Brodmann area | Cluster size | Significance | Brodmann area | Cluster size | Significance | |||
Visual task | Negative Correlation (BOLD signal decreases with increasing age) | 28, −84, 10 | 17/18/19 | 45 | P < 0.001 | 18/19 | 16 | P < 0.01 |
24, −100, 5a | — | — | — | 17/18 | 9 | P < 0.03 | ||
−16, −96, −5 | 17/18 | 10 | P < 0.04 | 17/18 | 9 | P < 0.03 | ||
Positive Correlation (BOLD signal increases with increasing age) | n.s. | — | — | — | — | — | — | |
Auditory task | Negative Correlation (BOLD signal decreases with increasing age) | 28, −88, 5 | 17/18/19 | 25 | P < 0.01 | 17/18/19 | 26 | P < 0.01 |
−36, −92, −5 | 18/19 | 22 | P < 0.01 | 18/19 | 19 | P < 0.01 | ||
Positive Correlation (BOLD signal increases with increasing age) | 40, −76, −15 | 18/19 | 19 | P < 0.02 | 18/19 | 13 | P < 0.02 | |
0, −88, −10 | 17/18 | 18 | P < 0.02 | 17/18 | 15 | P < 0.02 | ||
−48, −68 −15 | 19 | 7 | P < 0.053 | 19 | 7 | P < 0.05 | ||
28, −64, −15 | 19 | 2 | n.s. | 19 | 8 | P < 0.04 |
Significance level is corrected for multiple comparisons using extent correction within each sensory ROI.
Voxel is included within the 45 voxel cluster of the Regression with Age but is separated from the rest of the cluster when gray matter is accounted for in the BPM Regression.
Table V.
ANOVA and BPM ANCOVA results for all age group comparisons in visual cortex for the visual and auditory task
Peak voxel (x,y,z) | ANOVA | BPM ANCOVA | ||||||
---|---|---|---|---|---|---|---|---|
Brodmann area | Cluster size | Significance | Brodmann area | Cluster size | Significance | |||
Visual task comparisons | Young vs. Older Adults | 32, −92, 10 | 17/18/19 | 56 | P < 0.001 | 18/19 | 25 | P < 0.001 |
Young greater than older adults | −12, −100, −5 | 17/18 | 7 | P < 0.05 | 17 | 6 | P < 0.05 | |
Older adults greater than young | n.s. | – | – | – | – | – | – | |
Middle Aged vs. Older Adults | n.s. | – | – | – | – | – | – | |
Young vs. Middle Aged | n.s. | – | – | – | – | – | – | |
Auditory task comparisons | Young vs. Older Adults | 28, −88, 0 | 18/19 | 10 | P < 0.04 | 18/19 | 7 | P < 0.04 |
Young greater than older adults | −32, −92, −5 | 18/19 | 14 | P < 0.03 | 18/19 | 5 | P < 0.07 | |
Older adults greater than young | 40, −76, −15 | 18/19 | 17 | P < 0.02 | 18/19 | 6 | P < 0.05 | |
Middle Aged vs. Older Adults | 32, −76, 0 | 18/19 | 18 | P < 0.01 | 18/19 | 17 | P < 0.01 | |
Middle greater than older adults | −32, −76, −5 | 18/19 | 6 | P < 0.057 | 18/19 | 7 | P < 0.04 | |
Older adults greater than middle | 40, −76, −20 | 18/19 | 7 | P < 0.05 | 18/19 | 10 | P < 0.02 | |
Young vs. Middle Aged | n.s. | – | – | – | – | – | – |
Significance level is corrected for multiple comparisons using extent correction within each sensory ROI.
Visual cortex response to the visual paradigm
Not surprisingly, activation within the visual cortex was induced by the visual paradigm and showed a decreased activation level as age increased. In other words, younger participants had greater activation than older participants. This was supported by group ANOVA comparisons where significant clusters within both primary and secondary visual areas (left hemisphere BA17; left and right hemisphere BA18, right hemisphere BA19) were seen where young adults had greater activity than older adults (see Fig. 2). No areas within visual cortex were identified as having greater activity in older than younger subjects in either the regression or group ANOVA analysis. Further group comparisons showed that the middle‐aged group BOLD response did not significantly differ from either the young or older adult BOLD response.
Figure 2.
Visual task response differences. Axial slices depict clusters where young adults have significantly more activity than older adults in right BA18. To graphically depict the regression by age results, activity within the 28,−84,10 MNI coordinate voxel is shown. Interestingly, this voxel falls within the cluster identified in the ANOVA group results.
When age‐related gray matter differences were accounted for in the BPM analyses, activation differences were reduced in all analyses but remained significant (see Tables IV and V). For example, in the young vs. older adult group comparison the large cluster in the right hemisphere middle occipital gyrus was reduced by 50%, yet remained significant (see Table V and Fig. 3). These BPM results suggest that the functional disparities between older and younger adults are not fully explained by local gray matter atrophy. Interestingly, this cortical area did not have significant gray matter volume differences identified in the VBM between group analysis; however, in the uncorrected VBM map (T score = 5.55, P < 0.1), this area did show subthreshold gray matter decreases in older adults relative to younger adults (see Fig. 3).
Figure 3.
Biological parametric mapping (BPM) ANCOVA results for visual task activity difference between young and older adults: BPM results indicate that the increased fMRI activation in younger compared with older adults for the visual stimuli remains when differences in gray matter volume from voxel‐based morphometry (VBM) are accounted for in the analysis. It is interesting to note that the area involved in functional differences does not show a significant gray matter volume difference in the VBM analysis, but greater gray matter volume in young adults can be seen at subthreshold levels in this area in the uncorrected VBM map.
In addition, BPM regression results support the notion that gray matter variation across age incompletely accounts for age‐related differences in visual sensory activation. As can be seen in the R 2 values, indicating the variable's proportion of explained variance in the respective models, the influence of age is typically reduced but remains statistically robust after adjusting for gray matter volume (see Table VI). Voxels that were no longer significant in the BPM regression showed a decreased influence of age and reduced age R 2 values (i.e., positive difference score in column E). Such cases indicate that age‐related gray matter volume loss could fully account for changes in the BOLD signal. A few selected voxels showed an increase in the variance explained by age when gray matter was added to the model. For example, in auditory task, the signal in the voxel at 28,−64,−15 was not significant in the regression with age but showed a significant relationship with age when gray matter volume was included in the model. Therefore, it exhibited a larger R 2 value for age in the BPM regression (i.e., negative difference score in column E), indicating that adjusting the BOLD signal for gray matter volume reduced variability in the BOLD response to the auditory task unrelated to age.
Table VI.
Regression coefficients, their significance values, and partial R 2 values for the age and BPM age regressions
Peak voxel (x,y,z) | Regression coefficients and their significance values | ||||||
---|---|---|---|---|---|---|---|
Regression with age | BPM regression with age and VBM | ||||||
Age β value | Significance | Age β value | Gray matter β value | Icv β value | |||
Visual task | Negative correlation | 28, −84, 10 | 1.59 | P < 0.009 | −0.0128 | 2.10 | −0.000668 |
24, −100, 5a | – | – | −0.0411 | −3.75 | 0.00112 | ||
−16, −96, −5 | 4.74 | P < 0.012 | −0.0379 | −1.39 | −0.000630 | ||
Auditory task | Negative correlation | 28, −88, 5 | 0.348 | P < 0.021 | −0.00870 | −0.0957 | 0.0000785 |
−36, −92, −5 | 0.585 | P < 0.002 | −0.0115 | −0.506 | 0.000296 | ||
Positive correlation | 40, −76, −15 | −1.24 | P < 0.002 | 0.0263 | 1.04 | 0.000272 | |
0, −88, −10 | −0.912 | P < 0.032 | 0.0223 | −0.757 | 0.000329 | ||
−48, −68 −15 | −0.981 | P < 0.061 | 0.0168 | −0.213 | 0.000129 | ||
28, −64, −15 | −0.696 | P < 0.089 | 0.0156 | 1.89 | −0.000618 | ||
Peak voxel (x,y,z) | Partial R 2 values | ||||||
Regression with age | BPM regression with age and VBM | ||||||
Age | Gray matter and age together | Gray matter not shared with age | Age not shared with gray matter | ||||
Visual task | Negative correlation | 28, −84, 10 | 0.259 | 0.373 | 0.153 | 0.197 | |
24, −100, 5a | – | 0.199 | 0.00316 | 0.198 | |||
−16, −96, −5 | 0.250 | 0.279 | 0.0394 | 0.240 | |||
Auditory task | Negative correlation | 28, −88, 5 | 0.232 | 0.234 | 0.00339 | 0.222 | |
−36, −92, −5 | 0.293 | 0.316 | 0.0323 | 0.309 | |||
Positive correlation | 40, −76, −15 | 0.294 | 0.314 | 0.0274 | 0.262 | ||
0, −88, −10 | 0.219 | 0.229 | 0.0124 | 0.217 | |||
−48, −68 −15 | 0.199 | 0.200 | 0.000993 | 0.134 | |||
28, −64, −15 | 0.183 | 0.2273 | 0.05422 | 0.2268 |
Significance reflects the FDR corrected value of the individual local maximum reported in the SPM analysis.
Voxel is included within the 45 voxel cluster of the Regression with Age but is separated from the rest of the cluster when gray matter is accounted for in the BPM Regression.
Visual cortex response to the auditory paradigm
For both the negative and positive age regressions, significant voxels in spatially distinct cortical regions were seen, and this finding implied that unlike the activation response to visual stimuli, older adults do not simply show a reduced amount of deactivation to auditory stimuli in visual cortex relative to younger adults. The negative regression identifies significant areas where BOLD activity decreases as age increases and voxels in more medial aspects of occipital cortex (middle occipital gyrus) were seen superior to areas found in the positive regression (Table IV and Fig. 4). These more lateral inferior areas of occipital cortex (fusiform gyrus) were identified in the positive regression and depicted increasing activity with increasing age. When gray matter differences were controlled for in the BPM regression analysis, some clusters did reduce slightly in size but not to the same degree as in the activation differences on the visual task. Further, in some clusters significant voxels were added in the BPM analysis, and in fact, one cluster quadrupled in size (see Table IV).
Figure 4.
Auditory task response in occipital cortex: Clusters depicted on axial brain sections were identified as significant in the regression of age and BOLD signal, and the signal from the peak voxel is graphically depicted within the scatter plot. Blue spectrum colors depict the negative regression where deactivation activity is associated with older age (peak voxel −36,−92,−5 MNI coordinate), whereas red spectrum colors show areas identified in the positive regression that have deactivation activity associated with younger age (peak voxel 40,−76,−15 MNI coordinate). Between age group analyses identified similar cortical areas as in the regression analyses. Within the superior BA18 area, older adults tended to deactivate while younger adults activated. This was reversed within the inferior BA18 cluster where younger adults tended to deactivate and older adults activated. These same areas showed significant differences between older and middle‐aged adults as well.
When young and older adult age groups were compared, increased activity in young adults was seen in clusters within the superior area of the middle occipital gyrus (BA 18/19) (see Table V). In reviewing the within age‐group maps to interpret this activity difference, it was noted that these areas were shown to have deactivation (negative T‐scores) for older adults but slight activation in young adults. This age‐dependent alteration in activity can also be seen in the graph of the peak negative regression results for this area of middle occipital gyrus (upper left of Fig. 4). The inferior portion of occipital cortex identified in the positive regression was similarly identified in the ANOVA as an area with significantly more activity in older adults than young adults. Inspecting the within age‐group maps as well as data from the positive regression results, it was noted that this area showed deactivation in young subjects while slight activation in older adults (lower right of Fig. 4). When middle‐aged adults were compared to older adults similar results were seen (see Table V), although no significant differences were observed between young and middle‐aged participants.
Lastly, these group differences were assessed while controlling for group variation in gray matter volume using BPM ANCOVA. Significant clusters for differences between young and older adults were again reduced in size by ∼50% (see Table V). Of the three clusters identified in the group ANOVA, two remained significant. Activity differences between middle‐aged and older adults were impacted to a lesser extent when gray matter differences were controlled. In fact, two of the three clusters expanded spatially within the BPM ANCOVA.
DISCUSSION
We have shown that when voxel‐by‐voxel gray matter volume changes are controlled, age‐related BOLD activity differences in visual cortex remain significant although reduced by ∼50% in extent when young and older adults are compared directly. Comparing BOLD activity between middle‐aged and older adults after controlling for gray matter atrophy showed an even smaller effect and appeared to enhance differences between the age groups instead of reducing them. Further, age‐related deactivation differences in occipital cortex induced by auditory stimuli do not show the same age‐related reductions. Rather, older adults deactivated spatially distinct areas of occipital cortex not deactivated in young or middle‐aged participants while other areas were deactivated in young and middle‐aged adults but not in older adults. As such, cross‐modal deactivation of visual cortex does not reflect a scalar difference in neural activity between the age groups but may indicate strategy differences that are implemented to adjust for cross‐modal interference. These observations also contrast with previous identified differences in deactivation of the default‐mode network where older adults have an overall decrease in the amount of deactivation compared to younger subjects [Lustig et al.,2003].
Interestingly, the visual areas showing age‐related differences in deactivation roughly correspond to the ventral and dorsal stream of visual processing pathways, and thus, the data could be interpreted as age‐differences in visual deactivation based on strategies of reducing visual interference during an auditory task by suppressing categorical “where” and “what” visual perceptions. The deactivation that varied in older adults was in the dorsal pathway and may reflect a strategy of not paying attention to “where” visual stimuli occur (ahead of them in the goggles), whereas young and middle‐aged adults have deactivation in the visual ventral pathway and would not be paying attention to processing “what” a visual stimulus could be. In other words, the deactivation reflects a reduction from baseline visual processing during an auditory task in different areas of visual cortex [Pasley et al.,2007]. These processing streams have been identified as contributing differentially to the integration of audiovisual signals within cross‐modal studies of recognition and localization [Sestieri et al.,2006]. Alternatively, the spatially distinct pattern of age‐related differences in deactivation may allude to an indirect effect of age‐related sensory reweighting between the auditory and visual systems. Sensory reweighting contributes to increased instability in older adults due to changes between the influence of visual, vestibular, and proprioceptive senses on balance [Calvert et al.,2004]. This strategy difference or change in intersensory weighting may contribute to why older adults are more disturbed by auditory noise when performing a nonspatial visual selective attention task than younger adults, yet no age‐related difference was seen in the same population on a similar spatial visual task with auditory interference [e.g., Talsma et al.,2006]. Future experiments will explore whether this deactivation is due to strategy differences or to some other as yet elucidated reason.
Age‐related structural changes, such as loss of local gray matter volume, do not appear to account for the full extent of functional activity differences seen in aging. In light of PET findings which directly measure a reduction in cellular activity in older adults [e.g., Haxby et al.,1994], it is not surprising that BPM analyses reduced but did not completely eliminate differences in BOLD activity between the age groups. Aging alters sensory systems both at the sense organ and in the morphology of the sensory cortex. Since the cortex is receiving less input from the sensory organ and has less gray matter in sensory cortex, differences in visual task activation have greater overlap with age‐related gray matter differences. Further, the BPM method does not account for gray matter changes in other areas of the brain which could contribute to age‐related sensory cortical activity differences.
In contrast, clusters where young and older adults showed differences in the degree of deactivation tended to increase in extent after adjusting for gray matter volume differences within visual cortex, even though R 2 values tended to show small reductions in the variability that age could explain when adjusting for gray matter (Table VI). Deactivation of visual cortex may reflect top‐down redirection of attentional resources. As such, deactivations of visual cortex may be less affected by local changes in gray matter volume, since they are driven by areas outside of the sensory cortex itself. Alternatively, loss of grey matter in auditory cortex may influence the differences in visual cortex deactivation associated with the auditory task; however, a post‐hoc regression showed no significant association between the average grey matter volume of the auditory cortex ROI and the deactivations within visual cortex for this data.
For the most part, BOLD signal differences were altered by subthreshold differences in gray matter volume that did not survive correction for multiple comparisons within the VBM analysis. These findings suggest that gray matter volume loss depicted in VBM analyses underestimates the extent of functionally meaningful local structural changes in the aging brain. As such, relying on VBM results to provide overlapping areas where functional differences may be accounted for by structural differences is insufficient. In addition, as a research tool adjusting the BOLD signal in a location specific manner for changes in structure preceding second‐order analyses (e.g., post‐hoc analysis of gray matter loss from other brain areas contributing to local functional differences) may provide insight into debates concerning research topics in aging of compensatory neural activity [Cabeza,2002; Logan et al.,2002]. In addition, utilizing BPM analyses in future aging studies will provide an efficient means of adjusting age‐related BOLD signal differences for other local physiological effects know to be associated with increasing age. For example, age is associated with an increase in arteriosclerosis and differences in resting perfusion rates which may also be affecting the analysis of age‐related differences in the fMRI BOLD response. Further, by utilizing BPM with perfusion imaging and VBM, cautions discussed by D'Esposito et al. [1999] concerning the age‐related changes in hemodynamic coupling between neural activities can be directly controlled in future fMRI studies with older subjects. The findings of these future studies may address the current discrepancy in the literature as to whether or not the hemodynamic response actually varies with age [Brodtmann et al.,2003; Buckner et al.,2000; D'Esposito et al.,1999; Huettel et al.,2001; Ross et al.,1997].
In conclusion, age‐related differences in deactivation of visual cortex in response to an auditory task do not exhibit a similar scalar‐difference pattern as age‐related differences in activation of visual cortex in response to a visual task. When age‐related gray matter volume differences are accounted for in a voxel‐by‐voxel manner, activation differences are reduced but remain significant while deactivation differences tended to expand to encompass more areas of visual cortex. The spatially unique areas that show deactivation differences in young and older adults imply that these groups are utilizing differing strategies for coping with cross‐modal sensory information that is not pertinent to the task at hand.
Acknowledgements
The authors thank the Center for Biomolecular Imaging Staff, Ms. Debra Hege, Ms. Jennifer Mozolic, Ms. Kathy Pearson, and Mr. Aaron Baer for their assistance.
REFERENCES
- Aguirre GK,Zarahn E,D'Esposito M ( 1998): The inferential impact of global signal covariates in functional neuroimaging analyses. Neuroimage 8: 302–306. [DOI] [PubMed] [Google Scholar]
- Alavi A,Newberg AB,Souder E,Berlin JA ( 1993): Quantitative analysis of PET and MRI data in normal aging and Alzheimer's disease: Atrophy weighted total brain metabolism and absolute whole brain metabolism as reliable discriminators. J Nucl Med 34: 1681–1687. [PubMed] [Google Scholar]
- Alsius A,Navarra J,Campbell R,Soto‐Faraco S ( 2005): Audiovisual integration of speech falters under high attention demands. Curr Biol 15: 839–843. [DOI] [PubMed] [Google Scholar]
- Brodtmann A,Puce A,Syngeniotis A,Darby D,Donnan G ( 2003): The functional magnetic resonance imaging hemodynamic response to faces remains stable until the ninth decade. Neuroimage 20: 520–528. [DOI] [PubMed] [Google Scholar]
- Buckner RL,Snyder AZ,Sanders AL,Raichle ME,Morris JC ( 2000): Functional brain imaging of young, nondemented, and demented older adults. J Cogn Neurosci 12 ( Suppl 2): 24–34. [DOI] [PubMed] [Google Scholar]
- Cabeza R ( 2002): Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychol Aging 17: 85–100. [DOI] [PubMed] [Google Scholar]
- Calvert GA,Spence C,Stein BE ( 2004): The Handbook of Multisensory Processes. Cambridge, MA: MIT Press. [Google Scholar]
- Cao J,Worsley KJ ( 2001): Applications of random fields in human brain mapping In: Moore M, editor. Spatial Statistics: Methodological Aspects and Applications, Vol. 159 New York: Springer; pp 169–182. [Google Scholar]
- Casanova R,Srikanth R,Baer A,Laurienti PJ,Burdette JH,Hayasaka S,Flowers L,Wood F,Maldjian JA ( 2007): Biological parametric mapping: A statistical toolbox for multimodality brain image analysis. Neuroimage 34: 137–143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Colcombe SJ,Erickson KI,Raz N,Webb AG,Cohen NJ,McAuley E,Kramer AF ( 2003): Aerobic fitness reduces brain tissue loss in aging humans. J Gerontol A Biol Sci Med Sci 58: 176–180. [DOI] [PubMed] [Google Scholar]
- Colcombe SJ,Kramer AF,Erickson KI,Scalf P,McAuley E,Cohen NJ,Webb A,Jerome GJ,Marquez DX,Elavsky S ( 2004): Cardiovascular fitness, cortical plasticity, and aging. Proc Natl Acad Sci USA 101: 3316–3321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Colcombe SJ,Kramer AF,Erickson KI,Scalf P ( 2005): The implications of cortical recruitment and brain morphology for individual differences in inhibitory function in aging humans. Psychol Aging 20: 363–375. [DOI] [PubMed] [Google Scholar]
- Crum RM,Anthony JC,Bassett SS,Folstein MF ( 1993): Population‐based norms for the Mini‐Mental State Examination by age and educational level. JAMA 269: 2386–2391. [PubMed] [Google Scholar]
- D'Esposito M,Zarahn E,Aguirre GK,Rypma B ( 1999): The effect of normal aging on the coupling of neural activity to the bold hemodynamic response. Neuroimage 10: 6–14. [DOI] [PubMed] [Google Scholar]
- Fjell AM,Walhovd KB,Reinvang I,Lundervold A,Salat D,Quinn BT, et al. ( 2006): Selective increase of cortical thickness in high‐performing elderly‐structural indices of optimal cognitive aging. Neuroimage 29: 984–994. [DOI] [PubMed] [Google Scholar]
- Good CD,Johnsrude IS,Ashburner J,Henson RN,Friston KJ,Frackowiak RS ( 2001): A voxel‐based morphometric study of ageing in 465 normal adult human brains. Neuroimage 14: 21–36. [DOI] [PubMed] [Google Scholar]
- Haier RJ,Jung RE,Yeo RA,Head K,Alkire MT ( 2005): Structural brain variation, age, and response time. Cogn Affect Behav Neurosci 5: 246–251. [DOI] [PubMed] [Google Scholar]
- Haxby JV,Horwitz B,Ungerleider LG,Maisog JM,Pietrini P,Grady CL ( 1994): The functional organization of human extrastriate cortex: A PET‐rCBF study of selective attention to faces and locations. J Neurosci 14: 6336–6353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hayasaka S,Nichols TE ( 2003): Validating cluster size inference: Random field and permutation methods. Neuroimage 20: 2343–2356. [DOI] [PubMed] [Google Scholar]
- Huettel SA,Singerman JD,McCarthy G ( 2001): The effects of aging upon the hemodynamic response measured by functional MRI. Neuroimage 13: 161–175. [DOI] [PubMed] [Google Scholar]
- Jernigan TL,Gamst AC ( 2005): Changes in volume with age—Consistency and interpretation of observed effects. Neurobiol Aging 26: 1271–1274; discussion 1275–1278. [DOI] [PubMed] [Google Scholar]
- Jernigan TL,Archibald SL,Fennema‐Notestine C,Gamst AC,Stout JC,Bonner J,Hesselink JR ( 2001): Effects of age on tissues and regions of the cerebrum and cerebellum. Neurobiol Aging 22: 581–594. [DOI] [PubMed] [Google Scholar]
- Johnson SC,Saykin AJ,Baxter LC,Flashman LA,Santulli RB,McAllister TW, et al. ( 2000): The relationship between fMRI activation and cerebral atrophy: Comparison of normal aging and Alzheimer disease. Neuroimage 11: 179–187. [DOI] [PubMed] [Google Scholar]
- Justino L,Kergoat H,Kergoat MJ ( 2001): Changes in the retinocortical evoked potentials in subjects 75 years of age and older. Clin Neurophysiol 112: 1343–1348. [DOI] [PubMed] [Google Scholar]
- Kawashima R,O'Sullivan BT,Roland PE ( 1995): Positron‐emission tomography studies of cross‐modality inhibition in selective attentional tasks: Closing the “mind's eye”. Proc Natl Acad Sci USA 92: 5969–5972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lancaster JL,Woldorff MG,Parsons LM,Liotti M,Freitas CS,Rainey L,Kochunov PV,Nickerson D,Mikiten SA,Fox PT ( 2000): Automated Talairach atlas labels for functional brain mapping. Hum Brain Mapp 10: 120–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Laurienti PJ ( 2004): Deactivations, global signal, and the default mode of brain function. J Cogn Neurosci 16: 1481–1483. [DOI] [PubMed] [Google Scholar]
- Laurienti PJ,Burdette JH,Wallace MT,Yen YF,Field AS,Stein BE ( 2002): Deactivation of sensory‐specific cortex by cross‐modal stimuli. J Cogn Neurosci 14: 420–429. [DOI] [PubMed] [Google Scholar]
- Lemaitre H,Crivello F,Grassiot B,Alperovitch A,Tzourio C,Mazoyer B ( 2005): Age‐ and sex‐related effects on the neuroanatomy of healthy elderly. Neuroimage 26: 900–911. [DOI] [PubMed] [Google Scholar]
- Levine BK,Beason‐Held LL,Purpura KP,Aronchick DM,Optican LM,Alexander GE,Horwitz B,Rapoport SI,Schapiro MB ( 2000): Age‐related differences in visual perception: A PET study. Neurobiol Aging 21: 577–584. [DOI] [PubMed] [Google Scholar]
- Li RW,Edwards MH,Brown B ( 2000): Variation in Vernier acuity with age. Vision Res 40: 3775–3781. [DOI] [PubMed] [Google Scholar]
- Liu RS,Lemieux L,Bell GS,Sisodiya SM,Shorvon SD,Sander JW,Duncan JS ( 2003): A longitudinal study of brain morphometrics using quantitative magnetic resonance imaging and difference image analysis. Neuroimage 20: 22–33. [DOI] [PubMed] [Google Scholar]
- Logan JM,Sanders AL,Snyder AZ,Morris JC,Buckner RL ( 2002): Under‐recruitment and nonselective recruitment: Dissociable neural mechanisms associated with aging. Neuron 33: 827–840. [DOI] [PubMed] [Google Scholar]
- Lustig C,Snyder AZ,Bhakta M,O'Brien KC,McAvoy M,Raichle ME,Morris JC,Buckner RL ( 2003): Functional deactivations: Change with age and dementia of the Alzheimer type. Proc Natl Acad Sci USA 100: 14504–14509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Macaluso E,Frith CD,Driver J ( 2000): Modulation of human visual cortex by crossmodal spatial attention. Science 289: 1206–1208. [DOI] [PubMed] [Google Scholar]
- Maldjian JA,Schulder M,Liu WC,Mun IK,Hirschorn D,Murthy R,Carmel P,Kalnin A ( 1997): Intraoperative functional MRI using a real‐time neurosurgical navigation system. J Comput Assist Tomogr 21: 910–912. [DOI] [PubMed] [Google Scholar]
- Maldjian JA,Laurienti PJ,Kraft RA,Burdette JH ( 2003): An automated method for neuroanatomic and cytoarchitectonic atlas‐based interrogation of fMRI data sets. Neuroimage 19: 1233–1239. [DOI] [PubMed] [Google Scholar]
- Mansfield P ( 1977): Multi‐planar image formation using NMR spin echoes. J Phys C: Solid State Phys 10: 55–58. [Google Scholar]
- Mechelli A,Price CJ,Friston KJ,Ashburner J ( 2005): Voxel‐based morphometry of the human brain: Methods and applications. Curr Med Imaging Rev 1: 1–9. [Google Scholar]
- Meltzer CC,Zubieta JK,Links JM,Brakeman P,Stumpf MJ,Frost JJ ( 1996): MR‐based correction of brain PET measurements for heterogeneous gray matter radioactivity distribution. J Cereb Blood Flow Metab 16: 650–658. [DOI] [PubMed] [Google Scholar]
- Muller‐Gartner HW,Links JM,Prince JL,Bryan RN,McVeigh E,Leal JP,Davatzikos C,Frost JJ ( 1992): Measurement of radiotracer concentration in brain gray matter using positron emission tomography: MRI‐based correction for partial volume effects. J Cereb Blood Flow Metab 12: 571–583. [DOI] [PubMed] [Google Scholar]
- Ogawa S,Lee TM ( 1990): Magnetic resonance imaging of blood vessels at high fields: In vivo and in vitro measurements and image simulation. Magn Reson Med 16: 9–18. [DOI] [PubMed] [Google Scholar]
- Pasley BN,Inglis BA,Freeman RD ( 2007): Analysis of oxygen metabolism implies a neural origin for the negative BOLD response in human visual cortex. Neuroimage 36: 269–276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pruessner JC,Collins DL,Pruessner M,Evans AC ( 2001): Age and gender predict volume decline in the anterior and posterior hippocampus in early adulthood. J Neurosci 21: 194–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Prvulovic D,Van de Ven V,Sack AT,Maurer K,Linden DE ( 2005): Functional activation imaging in aging and dementia. Psychiatry Res 140: 97–113. [DOI] [PubMed] [Google Scholar]
- Ross MH,Yurgelun‐Todd DA,Renshaw PF,Maas LC,Mendelson JH,Mello NK,Cohen BM,Levin JM ( 1997): Age‐related reduction in functional MRI response to photic stimulation. Neurology 48: 173–176. [DOI] [PubMed] [Google Scholar]
- Sestieri C,Di Matteo R,Ferretti A,Del Gratta C,Caulo M,Tartaro A,Olivetti Belardinelli M,Romani GL ( 2006): “What” versus “where” in the audiovisual domain: An fMRI study. Neuroimage 33: 672–680. [DOI] [PubMed] [Google Scholar]
- Shulman GL,Corbetta M,Buckner RL,Raichle ME,Fiez JA,Miezin FM,Petersen SE ( 1997): Top‐down modulation of early sensory cortex. Cereb Cortex 7: 193–206. [DOI] [PubMed] [Google Scholar]
- Sowell ER,Peterson BS,Thompson PM,Welcome SE,Henkenius AL,Toga AW ( 2003): Mapping cortical change across the human life span. Nat Neurosci 6: 309–315. [DOI] [PubMed] [Google Scholar]
- Sowell ER,Thompson PM,Toga AW ( 2004): Mapping changes in the human cortex throughout the span of life. Neuroscientist 10: 372–392. [DOI] [PubMed] [Google Scholar]
- Spence C,Driver J ( 1997): On measuring selective attention to an expected sensory modality. Percept Psychophys 59: 389–403. [DOI] [PubMed] [Google Scholar]
- Spence C,Nicholls ME,Driver J ( 2001): The cost of expecting events in the wrong sensory modality. Percept Psychophys 63: 330–336. [DOI] [PubMed] [Google Scholar]
- Talsma D,Kok A,Ridderinkhof KR ( 2006): Selective attention to spatial and non‐spatial visual stimuli is affected differentially by age: Effects on event‐related brain potentials and performance data. Int J Psychophysiol 62: 249–261. [DOI] [PubMed] [Google Scholar]
- Tanna NK,Kohn MI,Horwich DN,Jolles PR,Zimmerman RA,Alves WM,Alavi A ( 1991): Analysis of brain and cerebrospinal fluid volumes with MR imaging: Impact on PET data correction for atrophy. Part 2: Aging and Alzheimer dementia. Radiology 178: 123–130. [DOI] [PubMed] [Google Scholar]
- Tisserand DJ,van Boxtel MP,Pruessner JC,Hofman P,Evans AC,Jolles J ( 2004): A voxel‐based morphometric study to determine individual differences in gray matter density associated with age and cognitive change over time. Cereb Cortex 14: 966–973. [DOI] [PubMed] [Google Scholar]
- Worsley KJ,Marrett S,Neelin P,Vandal AC,Friston KJ,Evans AC ( 1996): A unified statistical approach for determining significant signals in images of cerebral activation. Hum Brain Mapp 4: 58–73. [DOI] [PubMed] [Google Scholar]
- Yanase D,Matsunari I,Yajima K,Chen W,Fujikawa A,Nishimura S,Matsuda H,Yamada M ( 2005): Brain FDG PET study of normal aging in Japanese: Effect of atrophy correction. Eur J Nucl Med Mol Imaging 32: 794–805. [DOI] [PubMed] [Google Scholar]