Cerebral Cortex Advance Access originally published online on June 16, 2006
Cerebral Cortex 2007 17(5):1092-1099; doi:10.1093/cercor/bhl019
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Normal Developmental Changes in Inferior Frontal Gray Matter Are Associated with Improvement in Phonological Processing: A Longitudinal MRI Analysis
1 Laboratory of Neuro Imaging, Department of Neurology, University of California, Los Angeles, Los Angeles, CA 90095-1769, USA, 2 Department of Neuroscience, University of Florida, Gainesville, FL 32611, USA
Address correspondence to Dr Lisa H. Lu, Laboratory of Neuro Imaging, David Geffen School of Medicine at UCLA, 635 Charles Young Drive South, Room 225, Los Angeles, CA 90095, USA. Email: lisa.lu{at}loni.ucla.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study asked whether previously identified developmental changes in the gray matter of the left inferior frontal gyrus are associated with maturation of a linguistic skill. To test this hypothesis, we examined whether thickening of this region was correlated with developmental improvements in phonological processing but not hand motor skills in a unique longitudinal data set of 45 normally developing children (between ages 5 and 11 years) studied over a 2-year interval. We analyzed structural magnetic resonance imaging data using cortical pattern matching methods and correlated within-individual changes in cortical thickness with 2 neurocognitive scores. As predicted, gray matter thickening in the left inferior frontal cortex was associated with improving phonological processing scores but not with improving hand motor skills. By contrast, motor skill improvement was associated with thinning in the hand region of the left motor cortex, and cortical change in this region was not associated with phonological processing. This study illustrates a specific correspondence between regional gray matter thickness change and language skill change in normally developing children.
Key Words: cortical thickness • inferior frontal gyrus • morphometry • motor • phonological processing
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A century and a half of aphasia research has identified the left inferior frontal gyrus as a key area in language production. In addition to supporting the role of this gyrus in language production, functional imaging studies have also identified tightly clustered and interconnected modules within the left inferior frontal gyrus that work together to achieve complex language processing (Bookheimer 2002
). Much of this research has been aimed at understanding underlying neural processes that may be disrupted in dyslexia. In the present study, we correlated changes in cortical gray matter thickness with behavioral results from tests of language and motor functioning in normally developing children studied longitudinally to determine whether specific structural changes are associated with language maturation.
Cross-sectional magnetic resonance imaging (MRI) studies of normally developing children and adults have identified differences between the structure and asymmetry of the peri-Sylvian language region in children and adults. In one study, duplication of Heschl's gyrus was found to occur with approximately equal frequency in the 2 hemispheres of children, but in adults, duplication was more frequent in the right hemisphere (Leonard and others 1998). In addition, the left and right Sylvian fissures become more asymmetrical with age. The left Sylvian fissure extends more posteriorly than the right, and the slope of the right Sylvian fissure becomes steeper with maturation (Sowell and others 2002). The location of the greatest Sylvian fissure maturational displacement coincides with localized regions of gray matter density increase, and the correspondence between these 2 measures of anatomical change is more marked in the left hemisphere. In one of the few longitudinal studies published, we found a general pattern of gray matter thinning punctuated by 2 left hemisphere areas where gray matter increased in thickness. Interestingly, these 2 areas in the inferior frontal gyrus (Brodmann Areas [BA] 6, 44, and 45) and in the posterior peri-Sylvian region (BA 40, 41, 42) (Sowell, Thompson, Leonard, and others 2004) coincided with areas implicated in the development of reading and language skills (Shaywitz and others 1998; Pugh and others 2000; Simos and others 2000; Temple and others 2001; Georgiewa and others 2002). Because these findings of altered asymmetry, duplication of Heschl's gyrus, and cortical thickness increase all occur in classical language regions during early to mid childhood, it becomes of interest to determine which of these many changes is related to the improvements in specific cognitive skills that occur over this time period. Only longitudinal studies with repeated assessments can address how within-subject maturational changes in brain structure correlate to improving cognitive skills.
Our understanding of the cortical regions involved in language function is largely based on studies of adult brain disorders such as stroke. Given that relatively little is known about normative brain maturation and its cognitive correlates, it is possible that studies of developmental brain disorders such as dyslexia might shed light on the maturation and development of cortical regions important for language functioning. Neuropsychological studies have shown phonological processing deficits in dyslexic individuals who consistently perform worse than nonimpaired readers on tasks such as reading pronounceable nonwords, manipulation of phonemes, and elision (Snowling 1981; Liberman and Shankweiler 1985; Pratt and Brady 1988; Pennington and others 1990). Using a voxel-based analysis, Brown and others (2001) showed decreased gray matter among dyslexic men in the left posterior superior temporal gyrus and in bilateral inferior frontal gyri. Eckert and others (2003) measured pars triangularis volumes in dyslexic and normal children and found bilateral volume reductions in this structure in the dyslexic subjects. Positron emission tomography (PET) studies implicated the left inferior frontal lobe and posterior language areas in reading. Independent research groups have shown increased activation for dyslexic individuals in the left premotor region during reading (Rumsey and others 1992; Brunswick and others 1999). Other PET studies have shown relatively normal activation in the frontal areas, but abnormal activation in posterior language areas (Rumsey and others 1994, 1997). Functional MRI studies have shown that dyslexic individuals rely more on left inferior frontal and right superior temporal regions in response to increasing phonological decoding demands compared with normal readers (Shaywitz and others 1998, 2002). The left inferior frontal gyrus may serve to transform phonological representations of words to articulatory programs for speech production, whereas the left posterior temporal and angular gyrus may map phonological representations to meaning (Fiez and Petersen 1998; McCrory 2004).
Although the structural findings lead to speculation that anatomical changes (e.g., increasing gray matter, changing asymmetry) in the classical language regions may be related to establishment of complex neural networks that underlie and support improving language skills, the bi-hemispheric nature of most of the above findings raises the suspicion that these anatomical changes may also be required for more general cognitive development not specific to language. For example, in the right hemisphere, a general pattern of gray matter thinning was also punctuated by gray matter thickening in the posterior peri-Sylvian region. Gray matter thickening was unilateral only in the left inferior frontal gyrus (Sowell, Thompson, Leonard, and others 2004). Because functional imaging studies documenting activation abnormalities during phonological processing tasks relied on the dyslexic populations, we do not know which structural changes in implicated brain regions may be related to abnormal development and which changes reflect healthy development. Most of the above studies also used cross-sectional design, which is insufficient for addressing maturational changes within subjects. We approached the problem of understanding morphological changes associated with normal development by using a longitudinal design with normally developing children. Because of the substantial literature in dyslexia establishing the prominent role of the left inferior frontal gyrus in phonological processing, we chose phonological processing as an operationalized representation of language processing in normally developing children. We asked whether gray matter thickening in the left inferior frontal gyrus, the only unilateral region of gray matter thickening in the left hemisphere, is associated with developmental changes in this language skill. In order to establish the specificity of the relation between gray matter thickening in the left inferior frontal gyrus and phonological processing, we predicted that thickness change in this region would correlate with improving phonological skills but not with nonlanguage measures such as motor dexterity and strength. Accordingly, we predicted that increased motor facility with age would correlate with gray matter thickness change in the hand region of the primary motor cortex but not in the left inferior frontal gyrus (Penfield and Boldrey 1937; Toma and Nakai 2002). We aimed to establish this double dissociation, which would support the notion that gray matter thickening in the left inferior frontal gyrus is associated with the maturation of specific language skills. Further, an understanding of brain–behavior relationships in normal language development can be used to generate hypotheses about abnormal development in children with dyslexia.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects
Forty-five children were selected from an original sample of 77 children who received neuropsychological assessments and underwent MRI on 2 occasions, an average of 2.2 years apart. The selected children had movement-free scans suitable for processing. The 32 children who were excluded due to movement artifacts at one or the other time point did not differ from the retained sample of children in terms of age at time 1, measured intelligence, handedness, scan interval, or sex ratio (see Table 1). The original sample of 77 children was recruited by word of mouth and from presentation to parent groups at elementary schools in Alachua County, FL. Children were excluded if their parent reported a neurological or psychiatric diagnosis, placement in remedial classes, or testing for gifted programs. Informed consent-assent was obtained from the children and their parents according to procedures approved by the Institutional Review Board.
|
Behavioral Measures
Within a large psychoeducational test battery, 2 measures of phonological processing were administered 1) the Lindamood Auditory Conceptualization Test (LAC) (Lindamood C and Lindamood P 1979) and 2) a prepublished version of the Elision subtest from the Comprehensive Test of Phonological Processing (CTOPP) (Wagner and others 1999). The LAC is a criterion-referenced assessment that measures the ability to 1) discriminate one speech sound or phoneme from another and 2) segment a spoken word into its constituent phonemic units. It uses colored blocks to visually represent phonemes and requires children to manipulate these blocks to form meaningful units of sound or phonemes (Lindamood C and Lindamood P 1979). In this test, the child hears a pseudoword together with a block sequence that corresponds to the order of the phonemes in the presented pseudoword. Then the aurally presented pseudoword changes, and the child indicates addition, subtraction, or transposition of phonemes in the changed pseudoword by manipulating the position of blocks. The Elision subtest of the CTOPP requires the child to first repeat a real word, then to say the phonemic sequence that remains after a part has been deleted (e.g., "Say Toothbrush. Now say toothbrush without saying tooth" or "Say cup. Now say cup without saying /k/"). Each subject's raw score from time 1 was subtracted from the raw score from time 2 to obtain a change score. The change scores for the LAC and Elision were averaged to obtain a phonological processing composite change score (PHONO) for analysis with the brain measures.
Subjects' motor ability was assessed with the Purdue Pegboard Test (Tiffin and Asher 1948; Tiffin 1968) and the Hand Dynamometer (Reitan and Davison 1974). The Purdue Pegboard is a test of fine motor dexterity requiring the child to place pegs in a column of small round holes within 30 s, using one hand at a time. Raw scores were obtained for each hand. A hand "dynamometer" was used to test grip strength in kilograms. It requires the child to hold the apparatus with his/her arm extended down, away from the body, and to squeeze the apparatus as hard as possible. One practice and 2 recorded trials were administered with each hand, with at least 10 s between each trial. The 2 recorded trials were averaged to obtain one score for each hand. For both the peg and grip strength tasks, each subject's raw score from time 1 was subtracted from its corresponding time 2 raw score to obtain a change score for each hand. Because we were interested in general motor skill development and not in laterality effects, the change scores for the right and left hands were averaged to obtain a change score for each task (peg and grip strength). Finally, the change scores for the 2 tasks were then averaged to obtain a motor composite change score (MOTOR) for analysis.
Image Acquisition
Brain image data were obtained with a Siemens 1.5-T magnet with a 3-dimensional (3D) T1-weighted protocol (Siemens AG, Erlangen, Germany). Imaging parameters were as follows: time repetition, 10 msec; echo time, 4 msec; flip angle, 10°; matrix, 256 x 256 x 128; image voxel size, 0.98 x 0.98 x 1.25 mm; acquisition time, 6 min.
Image Processing
Image analysis procedures and their validation are reported in detail elsewhere (Sowell and others 2002; Sowell, Thompson, Leonard, and others 2004). Briefly, the magnetic resonance images were preprocessed with a series of manual and automated procedures executed by analysts blind to subject age, sex, and scan session: 1) transforming brain volumes into a standardized 3D coordinate space (Mazziotta and others 1995) using a 12-parameter, linear, automated image registration algorithm (Woods and others 1993); 2) classifying images into gray matter, white matter, and cerebral spinal fluid 3) removing nonbrain tissue (i.e., scalp, orbits) and the cerebellum and excluding the left hemisphere from the right; 4) automatically extracting the cortical surface (MacDonald and others 1994); 5) tracing 35 sulcal and gyral landmarks on the lateral and medial surfaces of each hemisphere (detailed in Sowell and others 2002; Sowell, Thompson, Leonard, and others 2004); 6) transforming the image volumes back into their own native image acquisition space by mathematically inverting the transformation which took them into standard space; 7) spatially registering all segmented images and brain surfaces for each individual by defining 80 standardized, manually defined, anatomical landmarks (40 in each hemisphere; the first and last points on each of 20 of the 35 sulcal lines drawn in each hemisphere) (Sowell and others 2002, 2003); 8) spatially "time locking" each individual's brain data acquired at time 2 to the data acquired at time 1 by manually selecting anatomical landmarks in both scans; and 9) measuring cortical thickness in millimeters averaged within a 15-mm sphere centered on each anatomically matched cortical surface point.
Once preprocessing was completed, points on the cortical surfaces surrounding and between the sulcal contours drawn on each individual's brain surface were calculated using the averaged sulcal contours as anchors to drive 3D cortical surface mesh models from each subject into correspondence (Thompson and others 2000). This cortical pattern matching technique allows the creation of average surface models and the creation of statistical maps of correlations between thickness change values and change scores on behavioral measures. Note all analyses of the thickness maps were conducted in each subject's native space.
Statistical Analysis
Time 1 gray matter thickness at each cortical surface point was subtracted from its corresponding time 2 thickness measure to generate a thickness change value. Each subject's behavioral change scores (i.e., PHONO and MOTOR) were correlated with the change in thickness at each of the 65 536 matched cortical surface points in each hemisphere. Results for the correlation between thickness change and behavioral change were then mapped to the average cortical surface using color coding to represent correlation coefficients and associated P values.
To control for multiple comparisons, we conducted permutation analyses by randomly permuting the behavioral change scores and the thickness change values in 10 000 new analyses. To test our a priori hypothesis that phonological processing would be associated with thickness increase in the left inferior frontal gyrus, permutation analyses were conducted within 9 regions of interest (ROIs) in each hemisphere. Coarse ROIs (of dorsal and ventral frontal lobe, parietal lobe, temporal lobe, occipital lobe) for the lateral and medial surfaces were created for each individual from a probabilistic atlas (Evans and others 1994). Although our a priori hypothesis was specific to the left inferior frontal gyrus, we chose to define an ROI that included this region based on data rather than anatomical landmarks that may not reflect the actual observed effects in this sample. Thus, we created a peri-Sylvian ROI based on the region where gray matter thickness increased in the left hemisphere of this population of subjects (Sowell, Thompson, Leonard, and others 2004). The ROIs for each individual were averaged to create regional masks (Fig. 1).
|
Data were analyzed with and without the 4 left-handed subjects. Patterns of results were similar with less statistical power to detect significant correlations when the left-handed subjects were excluded. Because the majority of left-handed subjects are also left hemisphere dominant for language (Branch and others 1964
; Strauss and Wada 1983; Strauss and others 1984), they were retained in the sample. |
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 2 summarizes performance on behavioral measures composing the PHONO and MOTOR change scores. Correlation of these change scores with gray matter thickness change supports the first prediction that gray matter thickening in the left inferior frontal cortex would correlate with improvements in phonological skill. Figure 2a shows the correlation between changes in thickness and PHONO for every cortical surface point. White areas centered in the left inferior precentral sulcus including BA 6 and 44 have statistically significant positive Pearson's r coefficients at P
0.05. Figure 3 shows an enlargement of the area with significant positive Pearson's r coefficients and identifies the cortical surface points with significant positive Pearson's r coefficients at P 0.01, 0.005, and 0.001. As can be seen from Figure 3, the surface points with the most robust positive correlations are located in or just adjacent to the left inferior precentral sulcus. The major cluster of significant positive correlation falls mainly within the peri-Sylvian ROI, although some peripheral points fall within the lateral dorsofrontal ROI. Permutation test results indicate that the probability of the number of positive correlations thresholded at P 0.05 in the left peri-Sylvian ROI is 10 out of 100 (white areas in Fig. 2a; refer to Table 3 for permutation results). The probability of the number of positive correlations thresholded at P 0.001 is 4 out of 100 (black dots in Fig. 3). In other words, there are cortical surface points within the a priori predicted inferior frontal peri-Sylvian region at which gray matter thickness increase is robustly correlated (i.e., P 0.001) with improvement in phonological processing skills, and the permutation analyses show that this regional pattern of correlations does not occur by chance (peri-Sylvian ROI permutation analysis, P = 0.04). These surface points are surrounded by cortex where thickness increase is also correlated with improving phonological skills, though at decreased levels of significance in the permutation analyses (P = 0.05–0.10, see Table 3).
|
|
|
|
To establish double dissociation, we predicted that the correlation between changes in gray matter thickness and a nonlanguage measure would not reach significance in the left inferior frontal region but would reach significance in the hand area of the motor strip. Indeed, the correlations between changes in thickness and the MOTOR change score did not approach significance in the left peri-Sylvian region (permutation P > 0.20) but was significant in the hand region of the left motor strip (Fig. 2b). This region of significant correlation extended to the lateral dorsofrontal cortex (parts of BA 6, 8, and 9). Significant negative Pearson's r coefficients at P
0.05 have been color coded red in Figure 2b. These negative correlations indicate thinning of gray matter with improvement in motor hand skills. Permutation test results indicate that the probability of the number of significant negative correlations in the left lateral dorsofrontal region is 4 out of 100 (or P = 0.04, see Table 3). Some unexpected areas with correlation trends between cortical thickness change and behavioral change were found. The medial ventrofrontal region was at trend level significance for both the PHONO (positive correlation) and MOTOR (negative correlation) change scores at a surface point threshold of 0.05 in the permutation analyses (PHONO, P = 0.06; MOTOR, P = 0.06). Given that the correlation between thickness change and the behavioral scores in the medial ventrofrontal region was not predicted and these findings were only at trend level of significance, interpretation should be done cautiously.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This unique data set has made it possible to identify morphological changes that underlie maturation in specific skills. Due to variability in individual anatomy and maturational rate, answering these kinds of questions is best accomplished with longitudinal data. We found that the previously identified island of gray matter thickening in the left inferior frontal gyrus (Sowell, Thompson, Leonard, and others 2004
) was indeed associated with maturational improvements in phonological skill. Crucially, this relationship does not appear to be an effect of general maturation because thickness change in this region does not correlate with another behavioral measure that also improves with age, namely, hand motor skills. Thus, the data are consistent with the view that the isolated thickening of the left inferior frontal gyrus is associated with improvement of language skills during childhood, rather than a nonspecific developmental change.
The area of significant correlation between thickness increase and improving phonological skills centering on the left inferior precentral sulcus (Fig. 2) corresponds to the area in which stimulation produces tongue, mouth, and jaw movements in the classic cortical stimulation study of Penfield and Boldrey (1937). The more dorsal part of this region also partially overlaps with areas that elicit or inhibit vocalization upon stimulation (Penfield and Boldrey 1937). The cortical surface points with the most robust correlation between increasing gray matter thickness and improving phonological processing skills are limited to the left inferior precentral sulcus and nearby cortices (Fig. 3). The region of significant correlation did not extend anteriorly to BAs 45 or 47. Our finding that phonological improvement was associated with changes in BAs 6/44 matches well with adult functional imaging studies. Functional imaging studies designed to parcel submodular roles of the left inferior frontal gyrus in language processing have shown BAs 6/44 to be selectively activated during phonological processing (Poldrack and others 1999; Bookheimer 2002), BA 45 to be selectively activated during syntactic processing (Dapretto and Bookheimer 1999; Bookheimer 2002), and BA 47/45 to be involved in semantic processing (Dapretto and Bookheimer 1999; Poldrack and others 1999; Bookheimer 2002). Such findings suggest finer parcellation of functioning in the left inferior frontal gyrus than previously appreciated from lesion studies. Posterior peri-Sylvian areas, on the other hand, appear to support the matching of phonological and semantic information in order to comprehend meaningful discourse (Beeson and Rapcsak 1998; Walsh and Darby 1999). Individuals with lesions in posterior peri-Sylvian regions produce relatively well-preserved phonological discourse but impaired semantic content (Beeson and Rapcsak 1998; Walsh and Darby 1999). Therefore, it may not be surprising that significant correlations between gray matter change and improvements in a purely phonological task were not found.
The meta-analysis of the imaging literature on word production by Indefrey and Levelt (2004) attempted to identify brain regions subserving syllabification, defined as operation on abstract segmental representations that follow phonological code retrieval and precede phonetic coding and articulation of motor representations. They identified the left posterior inferior frontal gyrus as the region subserving syllabification. Whether this region's role is in linguistic phonological processing or motor speech output, or both, is a fascinating question that is beyond the scope of this report. Interested readers are referred to readings discussing overlap between language and motor systems (Liberman and Mattingly 1985; Corballis 1999; Liberman and Whalen 2000; Fadiga and others 2002).
In contrast to the positive relationship between phonological development and gray matter thickening, we found that improvement in motor skills correlated "negatively" with changes in gray matter thickness in the hand area of the left motor cortex. Although the significant negative correlation extended beyond the motor cortex to the dorsofrontal region, correlation between thickness and motor skill nevertheless was spatially dissociated from correlation between thickness and phonological skill. Whereas simple association between changes in anatomical region and behavior may be related to confounding tertiary variables (e.g., general maturation), the double dissociation observed here rules out effects of tertiary variables and underscores the specificity of morphological changes in the region centering on the left inferior precentral sulcus as a correlate of corresponding changes in phonological skill. The consistency of our findings with our a priori hypotheses and with functional imaging studies gives confidence to the interpretation that maturational changes in the left BAs 6 and 44 underlie developmental improvements in phonological skills.
The lateralization of the significant correlation between cortical thinning and improving motor skills to the left hemisphere was consistent with our predominantly right-handed subjects. Hand motor skills improved for both hands, but hand skills changed systematically with only left frontal thinning. This lateralization may reflect a morphological basis underpinning preference for using the right hand for complex learned motor movements. The extension of the area of significant correlation between cortical thinning and motor skill improvement into the lateral dorsofrontal cortex, while unpredicted from lesion studies, may reflect a natural developmental change during childhood. Predictions based on lesions studies, which usually rely on adult populations with pathology, may not be appropriate for understanding emerging cognitive skills in normally developing children.
Of interest was the finding that the correlation between improving phonological skills and gray matter thickness is positive in the same regions where we observed cortical thickness increases in this population, whereas the correlation between improving hand motor skills and gray matter thickness is negative, in the same regions where we observed cortical thinning with age in this population (Sowell, Thompson, Leonard, and others 2004). In other words, improving phonological skills are associated with gray matter thickening, whereas improving hand motor skills are associated with gray matter thinning. Increases in the thickness of brain tissue that segments as gray matter may reflect proliferation of dendritic processes and synaptic connections or increases in soma size (Wellman and Sengelaub 1991; O'Kusky and others 1996; Giedd 2004). Thinning gray matter may reflect synaptic pruning and loss of dendritic processes (Huttenlocher and Dabholkar 1997). An alternative explanation to cortical thinning is that it results from an increase in myelin as it extends into the neuropil bordering gray and white matter. This could cause an apparent "thinning" in the gray matter that was actually related to an increase in white matter (Sowell and others 2003; Sowell, Thompson, and Toga 2004). Both synaptic pruning (Huttenlocher and Dabholkar 1997) and myelination (Benes and others 1994) are known from postmortem studies to continue during the first 2 decades of life, spanning the age range studied here. Given the complexity of human language in both oral and written forms and the critical role of phonological skills in reading, it makes sense that maturation of these language skills requires the establishment of complex and perhaps "new" neural networks, which may be manifested at the neuronal level by increased synaptic connections or increased somal size and reflected as increasing gray matter thickness on MRI. Fine motor skills, on the other hand, may be in a different part of their developmental trajectory, in a consolidating rather than an accelerating phase where improvement does not require the establishment of new neural networks. Possibly, this improved fine motor processing (i.e., increased motor speed) more likely results from increased myelination as existing networks and connections are fine-tuned. Although these explanations are only speculative, as MRI does not allow the spatial resolution to assess the etiology of cortical changes at the cellular level, converging evidence from an independent longitudinal study supports the notion that maturation is associated with both cortical thickening and thinning. Shaw and others (2006) found that whereas maturational trajectory is manifested by gray matter thickening in certain brain regions (e.g., prefrontal gyri) during young childhood, maturation is characterized by gray matter thinning in those same regions during adolescence. Although this effect was moderated by intelligence, the findings of Shaw and others (2006) underscore that maturational trajectories are complex and depend on both localized brain regions and the age examined.
Turkeltaub and others (2003) examined performance on the Lindamood Conceptualization Test (the same phonological processing task used in the present report) and correlated out-of-scanner performance on this test with activation during an implicit reading functional imaging task. They found that activation during implicit reading was correlated with phonological processing in the left inferior frontal gyrus (BA 47), left middle frontal gyrus (BA 9), and left inferior temporal gyrus (BA 20). These anatomical regions do not correspond with our finding of a significant correlation between thickness and phonological processing change in BA 6/44. Although these findings may seem inconsistent at first glance, direct comparison of the 2 studies is not appropriate due to important differences between these studies. The findings of Turkeltaub and others (2003) reflect the relationship between phonological processing and activation during "implicit reading" at one point in time. Activation patterns depend on the task and take place across a cortical surface that has invariant morphological features within a given functional run. Our study examined the relationship between cortical thickness and phonological processing change over a span of 2 years. We did not measure implicit reading. Although both studies measured phonological processing, activation during their functional imaging task reflected implicit reading and not phonological processing. Thus, it is not appropriate to compare activation reflective of implicit reading with morphological changes associated with phonological processing.
We have discussed functional roles and physiological changes of anatomical regions where correlations were found. We acknowledge that a strict and more conservative interpretation of our results can only localize correlations to the ROI within which permutation tests were conducted and not to the more specific regions appearing on the maps (Friston and others 1996). We have done so partly because a priori hypotheses justified eschewing the most conservative interpretation (Friston and others 1996) and partly because double dissociation was a more powerful method to localize function with structure than simple associations usually reported in brain mapping studies (Jernigan and others 2003).
In the present study, we used phonological processing as an operational definition of language processing in order to demonstrate a method of correlating functional maturation to structural changes to elucidate physiological changes underlying development within living subjects. We acknowledge that language skills, including phonological processing skills, most likely evolved using cortical networks that have other functional roles (Bookheimer 2002; Gelfand and Bookheimer 2003). Our proposition that structural changes in the inferior frontal gyrus are specific to language processing is offered with the intention of demonstrating that structural changes in this region cannot be attributed solely to general maturation. This study provides evidence that structural changes in this region may underlie developmental improvements in a specific linguistic skill in individual children. However, whether gray matter thickening allows for increased facility with language or the ability to master language leads to cortical thickening cannot be inferred from correlation studies.
|
Acknowledgments |
|---|
We are very grateful to the children and parents who participated in this study as well as to Sarah Smith and Jennifer Mockler for their tireless efforts in recruitment and testing. This work was supported by National Institute of Drug Abuse Grant R21 DA15878, RO1 DA017831, and March of Dimes Grant 5FY03-12 awarded to ERS; National Institute on Deafness and Other Communication Disorders R01 DC02292 awarded to CML; National Institutes of Health (NIH)/National Center for Research Resources grant P41 RR013642, and National Institute of Neurological Disorders and Stroke grant NS3753 awarded to AWT. PMT was also supported by AG016570, LM05639, EB01651, and RR019771. Additional support was provided by the NIH through the NIH Roadmap for Medical Research, Grant U54 RR021813 entitled Center for Computational Biology. Information on the National Centers for Biomedical Computing can be obtained from http://nihroadmap.nih.gov/bioinformatics. Conflict of Interest: None declared.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Beeson PM and Rapcsak SZ. (1998) The aphasias. In Snyder PJ and Nussbaum PD (Eds.). Clinical neuropsychology: a pocket handbook for assessment(American Psychological Association, Washington, DC) pp. 403–425.
Benes FM, Turtle M, Khan Y, Farol P. (1994) Myelination of a key relay zone in the hippocampal formation occurs in the human brain during childhood, adolescence, and adulthood. Arch Gen Psychiatry 51:477–484.[Abstract]
Bookheimer S. (2002) Functional MRI of language: new approaches to understanding the cortical organization of semantic processing. Annu Rev Neurosci 25:151–188.[CrossRef][ISI][Medline]
Branch C, Milner B, Rasmussen T. (1964) Intracarotid sodium amytal for the lateralization of cerebral speech dominance; observations in 123 patients. J Neurosurg 21:399–405.[ISI][Medline]
Brown WE, Eliez S, Menon V, Rumsey JM, White CD, Reiss AL. (2001) Preliminary evidence of widespread morphological variations of the brain in dyslexia. Neurology 56:781–783.
Brunswick N, McCrory E, Price CJ, Frith CD, Frith U. (1999) Explicit and implicit processing of words and pseudowords by adult developmental dyslexics: a search for Wernicke's Wortschatz? Brain 122:Pt 101901–1917.
Corballis MC. (1999) The gestural origins of language. Am Sci 87:138–145.[CrossRef][ISI]
Dapretto M and Bookheimer SY. (1999) Form and content: dissociating syntax and semantics in sentence comprehension. Neuron 24:427–432.[CrossRef][ISI][Medline]
Eckert MA, Leonard CM, Richards TL, Aylward EH, Thomson J, Berninger VW. (2003) Anatomical correlates of dyslexia: frontal and cerebellar findings. Brain 126:482–494.
Evans AC, Kamber M, Collins DL, MacDonald D. (1994) An MRI-based probabilistic atlas of neuroanatomy. In Shorvon SD (Ed.). Magnetic resonance scanning and epilepsy(Plenum, New York) pp. 263–274.
Fadiga L, Craighero L, Buccino G, Rizzolatti G. (2002) Speech listening specifically modulates the excitability of tongue muscles: a TMS study. Eur J Neurosci 15:399–402.[CrossRef][ISI][Medline]
Fiez JA and Petersen SE. (1998) Neuroimaging studies of word reading. Proc Natl Acad Sci USA 95:914–921.
Friston KJ, Holmes A, Poline JB, Price CJ, Frith CD. (1996) Detecting activations in PET and fMRI: levels of inference and power. Neuroimage 4:223–235.[CrossRef][ISI][Medline]
Gelfand JR and Bookheimer SY. (2003) Dissociating neural mechanisms of temporal sequencing and processing phonemes. Neuron 38:831–842.[CrossRef][ISI][Medline]
Georgiewa P, Rzanny R, Gaser C, Gerhard UJ, Vieweg U, Freesmeyer D, Mentzel HJ, Kaiser WA, Blanz B. (2002) Phonological processing in dyslexic children: a study combining functional imaging and event related potentials. Neurosci Lett 318:5–8.[CrossRef][ISI][Medline]
Giedd JN. (2004) Structural magnetic resonance imaging of the adolescent brain. Ann N Y Acad Sci 1021:77–85.
Huttenlocher PR and Dabholkar AS. (1997) Regional differences in synaptogenesis in human cerebral cortex. J Comp Neurol 387:167–178.[CrossRef][ISI][Medline]
Indefrey P and Levelt WJ. (2004) The spatial and temporal signatures of word production components. Cognition 92:101–144.[CrossRef][ISI][Medline]
Jernigan TL, Gamst AC, Fennema-Notestine C, Ostergaard AL. (2003) More "mapping" in brain mapping: statistical comparison of effects. Hum Brain Mapp 19:90–95.[CrossRef][ISI][Medline]
Leonard CM, Puranik C, Kuldau JM, Lombardino LJ. (1998) Normal variation in the frequency and location of human auditory cortex landmarks. Heschl's gyrus: where is it? Cereb Cortex 8:397–406.
Liberman AM and Mattingly IG. (1985) The motor theory of speech perception revised. Cognition 21:1–36.[CrossRef][ISI][Medline]
Liberman AM and Whalen DH. (2000) On the relation of speech to language. Trends Cogn Sci 4:187–196.[CrossRef][ISI][Medline]
Liberman IY and Shankweiler D. (1985) Phonology and the problems of learning to read and write. Rem Spec Educ 6:8–17.
Lindamood C and Lindamood P. (1979) Lindamood auditory conceptualization test(Austin, TX: PRO-ED).
MacDonald D, Avis D, Evans A. (1994) Multiple surface identification and matching in magnetic resonance images. Proc Vis Biomed Comput 2359:160–169.
Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J. (1995) A probabilistic atlas of the human brain: theory and rationale for its development. The International Consortium for Brain Mapping (ICBM). Neuroimage 2:89–101.[CrossRef][ISI][Medline]
McCrory E. (2004) The neurocognitive basis of developmental dyslexia. In Frackowiak RSJ, Friston KJ, Frith CD, Dolan RJ, Price CJ, Zeki S, Ashburner J, Penny W (Eds.). Human brain function 2nd ed (Elsevier Academic Press, New York) pp. 563–581.
O'Kusky JR, Akers MA, Vinters HV. (1996) Synaptogenesis in hemimegalencephaly: the numerical density of asymmetric and symmetric synapses in the cerebral cortex. Acta Neuropathol (Berl) 92:156–163.[CrossRef][Medline]
Penfield W and Boldrey E. (1937) Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 60:389–443.
Pennington BF, Van Orden GC, Smith SD, Green PA, Haith MM. (1990) Phonological processing skills and deficits in adult dyslexics. Child Dev 61:1753–1778.[CrossRef][ISI][Medline]
Poldrack RA, Wagner AD, Prull MW, Desmond JE, Glover GH, Gabrieli JD. (1999) Functional specialization for semantic and phonological processing in the left inferior prefrontal cortex. Neuroimage 10:15–35.[CrossRef][ISI][Medline]
Pratt AC and Brady S. (1988) Relation of phonological awareness to reading disability in children and adults. J Educ Psychol 80:319–323.[CrossRef][ISI]
Pugh KR, Mencl WE, Jenner AR, Katz L, Frost SJ, Lee JR, Shaywitz SE, Shaywitz BA. (2000) Functional neuroimaging studies of reading and reading disability (developmental dyslexia). Ment Retard Dev Disabil Res Rev 6:207–213.[CrossRef][ISI][Medline]
Reitan RM and Davison LA. (1974) Clinical neuropsychology: current status and applications(V.H. Winston, Washington, DC).
Rumsey JM, Andreason P, Zametkin AJ, Aquino T, King AC, Hamburger SD, Pikus A, Rapoport JL, Cohen RM. (1992) Failure to activate the left temporoparietal cortex in dyslexia. An oxygen 15 positron emission tomographic study. Arch Neurol 49:527–534.[Abstract]
Rumsey JM, Nace K, Donohue B, Wise D, Maisog JM, Andreason P. (1997) A positron emission tomographic study of impaired word recognition and phonological processing in dyslexic men. Arch Neurol 54:562–573.[Abstract]
Rumsey JM, Zametkin AJ, Andreason P, Hanahan AP, Hamburger SD, Aquino T, King AC, Pikus A, Cohen RM. (1994) Normal activation of frontotemporal language cortex in dyslexia, as measured with oxygen 15 positron emission tomography. Arch Neurol 51:27–38.[Abstract]
Shaw P, Greenstein D, Lerch J, Clasen L, Lenroot R, Gogtay N, Evans A, Rapoport J, Giedd J. (2006) Intellectual ability and cortical development in children and adolescents. Nature 440:676–679.[CrossRef][Medline]
Shaywitz BA, Shaywitz SE, Pugh KR, Mencl WE, Fulbright RK, Skudlarski P, Constable RT, Marchione KE, Fletcher JM, Lyon GR, Gore JC. (2002) Disruption of posterior brain systems for reading in children with developmental dyslexia. Biol Psychiatry 52:101–110.[CrossRef][ISI][Medline]
Shaywitz SE, Shaywitz BA, Pugh KR, Fulbright RK, Constable RT, Mencl WE, Shankweiler DP, Liberman AM, 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 USA 95:2636–2641.
Simos PG, Breier JI, Fletcher JM, Foorman BR, Bergman E, Fishbeck K, Papanicolaou AC. (2000) Brain activation profiles in dyslexic children during non-word reading: a magnetic source imaging study. Neurosci Lett 290:61–65.[CrossRef][ISI][Medline]
Snowling MJ. (1981) Phonemic deficits in developmental dyslexia. Psychol Res 43:219–234.[CrossRef][ISI][Medline]
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.[CrossRef][ISI][Medline]
Sowell ER, Thompson PM, Leonard CM, Welcome SE, Kan E, Toga AW. (2004) Longitudinal mapping of cortical thickness and brain growth in normal children. J Neurosci 24:8223–8231.
Sowell ER, Thompson PM, Rex D, Kornsand D, Tessner KD, Jernigan TL, Toga AW. (2002) Mapping sulcal pattern asymmetry and local cortical surface gray matter distribution in vivo: maturation in perisylvian cortices. Cereb Cortex 12:17–26.
Sowell ER, Thompson PM, Toga AW. (2004) Mapping changes in the human cortex throughout the span of life. Neuroscientist 10:372–392.[Abstract]
Strauss E and Wada J. (1983) Lateral preferences and cerebral speech dominance. Cortex 19:165–177.[ISI][Medline]
Strauss E, Wada J, Kosaka B. (1984) Writing hand posture and cerebral dominance for speech. Cortex 20:143–147.[ISI][Medline]
Temple E, Poldrack RA, Salidis J, Deutsch GK, Tallal P, Merzenich MM, Gabrieli JD. (2001) Disrupted neural responses to phonological and orthographic processing in dyslexic children: an fMRI study. Neuroreport 12:299–307.[CrossRef][ISI][Medline]
Thompson PM, Woods RP, Mega MS, Toga AW. (2000) Mathematical/computational challenges in creating deformable and probabilistic atlases of the human brain. Hum Brain Mapp 9:81–92.[CrossRef][ISI][Medline]
Tiffin J. (1968) Purdue pegboard: examiner manual(Science Research Associates, Chicago).
Tiffin J and Asher EJ. (1948) The purdue pegboard: norms and studies of reliability and validity. J Appl Psychol 32:234–247.[CrossRef]
Toma K and Nakai T. (2002) Functional MRI in human motor control studies and clinical applications. Magn Reson Med Sci 1:109–120.[CrossRef][Medline]
Turkeltaub PE, Gareau L, Flowers DL, Zeffiro TA, Eden GF. (2003) Development of neural mechanisms for reading. Nat Neurosci 6:767–773.[CrossRef][ISI][Medline]
Wagner RK, Torgesen JK, Rashotte C. (1999) Comprehensive test of phonological processing(Austin, TX: PRO-ED).
Walsh K and Darby D. (1999) The temporal lobes. Neuropsychology: a clinical approach 4th ed. (Churchill Livingstone, New York) pp. 179–223.
Wellman CL and Sengelaub DR. (1991) Cortical neuroanatomical correlates of behavioral deficits produced by lesion of the basal forebrain in rats. Behav Neural Biol 56:1–24.[CrossRef][ISI][Medline]
Woods RP, Mazziotta JC, Cherry SR. (1993) MRI-PET registration with automated algorithm. J Comput Assist Tomogr 17:536–546.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Shaw, N. J. Kabani, J. P. Lerch, K. Eckstrand, R. Lenroot, N. Gogtay, D. Greenstein, L. Clasen, A. Evans, J. L. Rapoport, et al. Neurodevelopmental Trajectories of the Human Cerebral Cortex J. Neurosci., April 2, 2008; 28(14): 3586 - 3594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. R. Sowell, S. N. Mattson, E. Kan, P. M. Thompson, E. P. Riley, and A. W. Toga Abnormal Cortical Thickness and Brain Behavior Correlation Patterns in Individuals with Heavy Prenatal Alcohol Exposure Cereb Cortex, January 1, 2008; 18(1): 136 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P M THOMPSON and L G APOSTOLOVA Computational anatomical methods as applied to ageing and dementia Br. J. Radiol., December 1, 2007; 80(Special_Issue_2): S78 - S91. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||