Abstract

Previous investigations of cerebral anatomy in persistent developmental stutterers have reported bilateral anomalies in the perisylvian region and atypical patterns of cerebral asymmetry. In this study, perisylvian sulcal patterns were analyzed to compare subjects with persistent developmental stuttering (PDS) and an age-, hand-, and gender-matched control group. This analysis was accomplished using software designed for 3-dimensional sulcal identification and extraction. Patterns of cerebral asymmetry were also investigated with standard planimetric measurements. PDS subjects showed a small but significant increase in both the number of sulci connecting with the second segment of the right Sylvian fissure and in the number of suprasylvian gyral banks (of sulci) along this segment. No differences were seen in the left perisylvian region for either sulcal number or gyral bank number. Measurements of asymmetry revealed typical patterns of cerebral asymmetry in both groups with no significant differences in frontal and occipital width asymmetry, frontal and occipital pole asymmetry, or planum temporale and Sylvian fissure asymmetries. The subtle difference in cortical folding of the right perisylvian region observed in PDS subjects may correlate with functional imaging studies that have reported increased right-hemisphere activity during stuttered speech.

Introduction

Persistent developmental stuttering (PDS) is a speech disorder that typically appears between 2 and 4 years of age and is characterized by sound and syllable repetitions and audible and silent prolongations (Bloodstein 1995). The prevalence of stuttering in adults is 1% with a lifetime incidence of 4%, and the disorder affects males more than females with sex ratios estimated at 2–3:1 in childhood and 10:1 in adulthood (Bloodstein 1995). There appear to be significant genetic factors involved in the development of the disorder (Howie 1981), though the influence and the location of genes that may be involved is still under investigation (Ambrose et al. 1997; Riaz et al. 2005).

Functional imaging studies have consistently shown abnormal activation patterns during stuttered speech. In one of the early positron emission tomography (PET) studies looking at the neural systems of stuttered speech, overactivations of the motor system and right lateralization of primary and extraprimary motor cortices were detected, along with an absence of left-lateralized activations of the auditory system (Fox et al. 1996). Subsequent PET investigations of developmental stutterers using stuttering performance correlations reported a similar lack of left hemisphere speech-motor lateralization during stuttering, along with bilateral deactivations in the auditory association area (Braun et al. 1997; Fox et al. 2000; Ingham et al. 2004). One recent meta-analysis (Brown et al. 2005) concluded that stutterers utilize the same core areas of the speech-motor system seen in fluent controls, with a rightward shift of activity during stuttered speech that appears to recede during induced fluency (Fox et al. 1996, 2000; Ingham et al. 2004).

Structural imaging studies of PDS subjects have reported several significant morphological differences. One study described anomalous anatomy along the superior bank of the Sylvian fissure in PDS subjects. These anomalies included an increase in the gyrification of the bilateral suprasylvian opercula, slight differences in the configuration of the sulcal boundaries of the pars triangularis portion of the inferior frontal gyrus, and the presence of a doubled diagonal sulcus (DgS) in the pars opercularis (Foundas et al. 2001). However, these perisylvian anomalies were not present in all subjects studied and “no one anatomic feature accounted for the group differences” observed (Foundas et al. 2001, p. 212). Additionally, several studies have described alterations in patterns of cerebral asymmetry in developmental stutterers. These findings include atypical patterns of lobar width asymmetry and petalia in both the occipital and frontal lobes (Strub et al. 1987), an increase in planum temporale (PT) size with an overall reduction in the degree of leftward asymmetry (Foundas et al. 2001), and atypical prefrontal and occipital lobe volume asymmetries that deviate from expected distributions in healthy individuals (Foundas et al. 2003). In addition to atypical perisylvian anatomy and cerebral asymmetry patterns, a final class of structural imaging findings in PDS involves aberrant features of white matter tract connectivity and right-hemispheric white matter volume. A decrease in fractional anisotropy in the left rolandic operculum of PDS subjects has been detected with diffusion tensor imaging (DTI) (Sommer et al. 2002). This was interpreted as a white matter disconnection within the left hemisphere that might interfere with “the sensorimotor integration necessary for fluent speech production” (Sommer et al. 2002, p. 382). One voxel-based morphometry (VBM) study reported an increase in right-hemisphere white matter volume for the superior temporal gyrus, the precentral gyrus, the pars opercularis portion of the inferior frontal gyrus, and the middle frontal gyrus of PDS subjects (Jancke et al. 2004). The same VBM study did not detect any gray matter density or volume differences in the entire cerebrum or white matter anomalies in the left hemisphere of PDS subjects (Jancke et al. 2004). The information provided by these studies is intriguing and suggests that structural differences in PDS subjects may relate to the abnormal functional activity of stuttered speech. However, there are still unanswered questions as some studies have identified bilateral anomalies (Foundas et al. 2001, 2003), left-lateralized anomalies (Sommer et al. 2002), or right-lateralized anomalies (Jancke et al. 2004). Moreover, not all studies have observed these abnormalities to be present in all PDS subjects (Foundas et al. 2001, 2003).

The major aim of the present study was to evaluate perisylvian anatomy and patterns of cerebral asymmetry in developmental stuttering. As some previous descriptions of perisylvian anomalies in PDS have been qualitative, this study was designed to provide quantifiable measures of the region. Therefore, an automated program for 3-dimensional (3-D) sulcal identification, recognition, and extraction was utilized. This was done to remove the “viewpoint dependency” (Lohmann et al. 1999) that can arise when only successive 2-D sections are used to evaluate complex morphological features such as cerebral sulci. In addition, standard planimetric measurements of cerebral asymmetry were applied (Geschwind and Levitsky 1968; Yeni-Komshian and Benson 1976; Chui and Damasio 1980; Galaburda et al. 1987) with some modification. As this is the first morphological study of PDS to include only right-handed male subjects, the removal of all group differences except stuttering status allowed the potential correlation between cerebral asymmetry and the disorder to better be evaluated.

Materials and Methods

Subjects

Participants were 19 male PDS subjects and 16 male controls. PDS subjects ranged from 19 to 58 years (mean age: 35.1 ± 12.4 years) and control subjects ranged from 21 to 63 years (mean age: 34.8 ± 12.2 years). The developmental stutterers were healthy, adult volunteers who self-reported a previous developmental stuttering diagnosis and displayed stuttering as confirmed by 2 speech-language pathologists (J.C.I. and R.J.I.) using standard clinical assessments. Control subjects were recruited as part of the International Consortium for Brain Mapping project (Mazziotta et al. 1995, 2001). Imaging of all subjects was performed with their informed written consent and Institutional Review Board approval.

The 2 groups were not matched one-to-one but were instead constituted to provide 2 groups of male subjects with similar mean age and degree of right handedness. Similar approaches in matching PDS subjects and controls to generate representative groups can be found in other studies of the disorder (Braun et al. 1997; Stager and Ludlow 1998; De Nil et al. 2001). All subjects had a review of their medical history to rule out endocrinological, neurological, and/or psychiatric illnesses, and all subjects had a minimum of a high school education. All controls were specifically screened for any history of stuttering or other speech disorder, and no controls had a history of either.

Subjects were initially asked to self-report handedness, and all subjects reported that they were right handed. Following that screening, the degree of right handedness of all subjects was classified by a modified Edinburgh handedness inventory (Oldfield 1971) (see Appendix 1). For PDS subjects, the mean Edinburgh score was 97.1 ± 7.4 and for controls the mean score was 94.6 ± 8.4. One control reported right handedness during the initial screening process, but the modified Edinburgh score was not available.

Speech Measurements

PDS subjects had all reported a history of stuttering since childhood and had participated in a variety of therapy programs. All reported that treatment had not been beneficial. Measurements of stuttered speech in the PDS group were based on recorded samples of 3-min oral reading and 3-min monologue. Group percentage syllables stuttered (%SS) for oral reading samples ranged from 1.2 to 27.6 %SS with a group mean %SS of 8.7. Group percentage syllables stuttered for monologue samples ranged from 3.2 to 33.1 %SS with a group mean %SS of 8.5. Averaging the scores for oral reading and monologue in each subject, the group ranged from 2.3 to 23.7 %SS with a group mean of %SS on all tasks of 8.6. No speech sample was available for control subjects, though all control subjects answered negatively to a specific screening question regarding a history of stuttering or other speech disorder. To establish measurement reliability, an independent trained speech-language pathologist evaluated 15 of the 38 samples produced in the above recordings. Measurement agreement was excellent with a Pearson's r correlation of 0.99 in the rating of percentage of syllables stuttered.

Magnetic Resonance Imaging Protocol and Preprocessing

All subjects were imaged at the Research Imaging Center on a Siemens 3-T Trio magnetic resonance imaging scanner with a high-resolution 8-channel head coil. T1-weighted imaging was performed using a high-resolution (800 μm resolution in x-, y-, and z- directions, axial orientation) retrospective motion-corrected protocol (Kochunov et al. 2006). Six full-resolution volumes were acquired with a 3-D TurboFlash sequence with an adiabatic inversion contrast pulse and averaged following motion correction. The scan parameters used were as follows: time echo/time repitition/time to inversion = 3.04/2100/785 ms, flip angle = 13 degrees.

Images were preprocessed according to the following steps: 1) nonbrain tissue removal, 2) image registration to the Talairach coordinate system, 3) radiofrequency inhomogeneity correction, 4) image segmentation into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF), and 5) sulcal identification and extraction with the program BrainVISA (BV) (Mangin et al. 2004). The steps used in image preprocessing at our center are described in detail elsewhere (Kochunov et al. 2005, 2006) and are briefly summarized here. The automated skull-stripping procedure Brain Extraction Tool (Smith 2002) was used to remove nonbrain tissue, and the extracted brain images were subsequently cropped at the level of the brainstem. The images were then globally, spatially normalized to the Talairach coordinate system in order to correct for gross differences in subject brain size and orientation. This normalization was achieved with FSL FLIRT spatial normalization (SN) software (Jenkinson et al. 2002) using a 3-D, 15-voxel–wide sinc interpolation kernel.

Extraction and Identification of Cerebral Sulci with BV

The processed images were imported into the BV sulcal extraction and identification “pipeline.” This image-processing pipeline uses a neural network–based recognition system to generate a subject-specific representation of cerebral sulci that are automatically labeled. BV has been used extensively as a tool to perform quantitative analysis on sulcal features (e.g., depth, sulcal length, center of gravity, surface area) in other studies of cerebral morphology (Molko et al. 2003, 2004; Mangin et al. 2004; Juch et al. 2005; Kochunov et al. 2005; Eckert et al. 2006).

Briefly, the program first identifies the cerebral hemispheres and cerebellum using a technique based on 3-D erosion and template-based 3-D seed growth (Mangin et al. 1996, 2004). Segmented GM and WM images are then utilized in order to create hemispheric meshes. To create meshes of the cerebral sulci, a homotopic erosion technique and a “crevasse detector” are used to obtain the medial surface of the cortical sulci (Mangin et al. 1995). The sulcal structures are reconstructed as the medial surfaces of the 2 opposing gyral banks of the sulcus, and the reconstructed meshes span from the most internal point of the sulcus at the gray/CSF border to the convex hull of the cortical surface (Mangin et al. 1995, 2004). The sulcal recognition pipeline incorporates 500 artificial neural network–based pattern classifiers for the automated recognition of cerebral sulci. The neural network–based classifiers in the BV pipeline were established on a data set of 26 expertly labeled images and are used to identify 58 sulci per hemisphere (Mangin et al. 2004). The recognition rate of the automated sulcal labeling tool in BV, prior to correction, is 76% for all sulci (Riviere et al. 2002) and ≥94% in the perisylvian region for the central sulcus (CS) and lateral fissure of Sylvius (Mangin et al. 2004).

For sulci of interest in the study, the automated recognition procedure was checked for errors and corrected as necessary. Manual adjustments to the sulcal labeling scheme were done according to standard anatomical convention (Ono et al. 1990; Tamraz and Comair 2006) by a neuroanatomist (M.C.).

BV was then utilized to render a 3-D triangular mesh of all sulci (Riviere et al. 2002; Mangin et al. 2004). Figure 1A shows the combination of hemisphere and sulcal meshes as viewed in BV (after review of the automated sulcal extraction and recognition pipeline) and Figure 1B shows the same subject as seen in a 2-D sagittal section.

Figure 1.

Perisylvian sulcal morphology as rendered in BV. (A) Mesh of hemisphere with perisylvian sulcal meshes. The hemispheric mesh is made slightly transparent so that the contour and depth of sulcal structures is revealed. The second segment of the Sylvian fissure is outlined as the portion between the CmS, giving rise to the AHR and AAR, and the beginning of PAR of the Sylvian fissure. (B) A 2-D sagittal section from the same subject seen in (A) showing the second segment of the Sylvian fissure and the connections of several sulci. Notably, a VR is seen ascending into the frontal operculum from the Sylvian fissure (caudal to the CmS) after the ASC and PSC sulci are identified. Note that this VR can scarcely be detected as a slight green mesh through the transparent hemisphere in (A) just rostral to the ASC but is otherwise difficult to detect without utilizing the combination of the 2-D section. (C) The same subject as seen in (A) and (B) with an area of interest detailed in the inset on the right. The temporal lobe has been removed in the inset image so that the CrSI is seen as the medial border for the region of interest. The portion of mesh in the inset below the label “CrSI” is the superior portion of the insular gyri. The sulci and gyri directly above the CrSI in this view represent the frontal and parietal opercula of the suprasylvian bank as seen from the Sylvian fissure. The sulci detected in (A) and (B) are used as natural partitions between opposing gyral banks (numbered 1–6) in this subject. Other sulci depicted in (AC): IPeC, inferior precentral sulcus; SPeC, superior precentral sulcus; IPoC, inferior postcentral sulcus.

Figure 1.

Perisylvian sulcal morphology as rendered in BV. (A) Mesh of hemisphere with perisylvian sulcal meshes. The hemispheric mesh is made slightly transparent so that the contour and depth of sulcal structures is revealed. The second segment of the Sylvian fissure is outlined as the portion between the CmS, giving rise to the AHR and AAR, and the beginning of PAR of the Sylvian fissure. (B) A 2-D sagittal section from the same subject seen in (A) showing the second segment of the Sylvian fissure and the connections of several sulci. Notably, a VR is seen ascending into the frontal operculum from the Sylvian fissure (caudal to the CmS) after the ASC and PSC sulci are identified. Note that this VR can scarcely be detected as a slight green mesh through the transparent hemisphere in (A) just rostral to the ASC but is otherwise difficult to detect without utilizing the combination of the 2-D section. (C) The same subject as seen in (A) and (B) with an area of interest detailed in the inset on the right. The temporal lobe has been removed in the inset image so that the CrSI is seen as the medial border for the region of interest. The portion of mesh in the inset below the label “CrSI” is the superior portion of the insular gyri. The sulci and gyri directly above the CrSI in this view represent the frontal and parietal opercula of the suprasylvian bank as seen from the Sylvian fissure. The sulci detected in (A) and (B) are used as natural partitions between opposing gyral banks (numbered 1–6) in this subject. Other sulci depicted in (AC): IPeC, inferior precentral sulcus; SPeC, superior precentral sulcus; IPoC, inferior postcentral sulcus.

Quantifying Perisylvian Sulci and Gyral Bank Numbers

A combination of primary, secondary, and tertiary sulci were selected in areas found to have anomalous perisylvian anatomy or functional activation patterns in previous studies of PDS (Fox et al. 1996, 2000; Foundas et al. 2001; Sommer et al. 2002; Ingham et al. 2004; Jancke et al. 2004; Brown et al. 2005). Primary sulci are defined as those sulci present prior to the 30th week of development (Tamraz and Comair 2006). These sulci have alternately been described as developing between the 16th and 28th weeks (Armstrong et al. 1995). Secondary sulci are those forming at the 32nd week, and tertiary sulci are those appearing from the 36th week through the first year of life (Armstrong et al. 1995). Table 1 lists all sulci labeled in this study and their approximate range of development, and Figure 1(A,C) illustrate their location along the suprasylvian bank (for those sulci present on the representative brain and in the slice shown of the same subject in Fig. 1). In addition to sulci listed in Table 1, the following sulci were labeled: the common stem (CmS) for the anterior ascending ramus (AAR) and anterior horizontal ramus (AHR) of the Sylvian fissure, the posterior ascending ramus (PAR) and posterior descending ramus (PDR) of the Sylvian fissure. Based on established neuroanatomical descriptions (Ono et al. 1990; Tamraz and Comair 2006), all regularly segmented sulci (e.g., precentral sulci) were labeled by position as necessary (i.e., superior, intermediate, inferior).

Table 1

Perisylvian sulci labeled with BV

Sulcus Classification Estimate of maturationa 
Sylvian fissure Primary 10-15 weeks, OW 16 
CrSI Primary 16-23 weeks, OW 25-28 
CS Primary 20-26 weeks, OW 21 
PeC Primary 24-27 weeks, OW 25-28 
PoC Primary 24-30 weeks, OW 29 
TSS Secondary 32-35 weeks, OW 32 
DgS Tertiary >36 weeks 
ASC Tertiary >36 weeks 
PSC Tertiary >36 weeks 
VR of Sylvian fissureb Tertiaryb >36 weeksb 
AAR and AHR of the Sylvian fissurec Unknownc Unknownc 
Sulcus Classification Estimate of maturationa 
Sylvian fissure Primary 10-15 weeks, OW 16 
CrSI Primary 16-23 weeks, OW 25-28 
CS Primary 20-26 weeks, OW 21 
PeC Primary 24-27 weeks, OW 25-28 
PoC Primary 24-30 weeks, OW 29 
TSS Secondary 32-35 weeks, OW 32 
DgS Tertiary >36 weeks 
ASC Tertiary >36 weeks 
PSC Tertiary >36 weeks 
VR of Sylvian fissureb Tertiaryb >36 weeksb 
AAR and AHR of the Sylvian fissurec Unknownc Unknownc 

Note: OW, ontogenetic weeks.

a

Range compiled from several studies (Chi et al. 1977; Ono et al. 1990; Tamraz and Comair 2006). Range in “OW” (Armstrong et al. 1995).

b

VR were small tertiary sulci detected by BV to which no standard name could be applied.

c

According to Chi et al. (1977), the pars triangularis and pars opercularis components of the inferior frontal gyrus can be distinguished at 28 weeks of development. This suggests that these sulci are present at <30 weeks.

Boundaries were then established for quantifying sulci with true connections to the Sylvian fissure and the number of related gyral banks (see Fig. 1). The Sylvian fissure is often described as consisting of 3 consecutive components including a stem (first or hidden segment), a posterior ramus (second or posterior horizontal segment), and a terminal ascending segment (third or posterior ascending segment) (Ono et al. 1990; Tamraz and Comair 2006). The boundaries established in the present study limited the region of interest to the second (or posterior horizontal) segment of the Sylvian fissure (Ono et al. 1990), and this segment is marked on Figure 1(A,B). Rostrally, the boundary was the CmS or AAR of the Sylvian fissure depending on the configuration present in the subject. The caudal boundary was the point of origin of the PAR of the Sylvian fissure. Only in cases where there was no clear PAR, the posterior transverse supratemporal sulcus (TSS) caudal to the last Heschl's gyrus was used (Tamraz and Comair 2006). Medially, the boundary was the circular sulcus of the insula (CrSI), and laterally, it was the point where the perisylvian opercula meet the lateral surface of the hemisphere. Superiorly directed rami originating entirely from the CrSI, from the PAR of the Sylvian fissure, or from the stem of the Sylvian fissure rostral to the AAR or CmS were excluded. In addition, sulci with only a “pseudoconnection” to the Sylvian fissure (Ono et al. 1990) were excluded if there was an intervening gyral passage between the sulcal mesh and the Sylvian fissure mesh (e.g., the subcentral gyrus underling the CS). Connecting sulci were determined to be within these boundaries by simultaneously viewing each subject's 3-D hemispheric mesh and successive 2-D sagittal sections in BV (see Fig. 1). The simultaneous use of the 3-D rendering prevented the incorrect detection of 2 separate sulci for complex sulcal structures that moved obliquely relative to successively deeper sagittal planes.

With the boundaries defined, the number of sulci connecting with the second segment of the Sylvian fissure was quantified. This included sulcal meshes descending from a suprasylvian position (e.g., CS) and tertiary rami originating from the Sylvian fissure (e.g., anterior subcentral [ASC] sulcus). Sulci within these boundaries and with a connection to the Sylvian fissure were then used as natural division points between the opposing suprasylvian gyral banks of the sulcus (see Fig. 1C). For instance, if the parietal operculum was crossed by the inferior postcentral sulcus (IPoC) connecting to the second part of the Sylvian fissure, 2 gyral banks were recorded (one on either side of the sulcus). For this portion of the analysis, interruptions made by the DgS into the frontal operculum were included (as separating 2 banks), even in instances when it did not have a connection to the Sylvian fissure but was connected to the AAR or CmS and divided 2 suprasylvian gyral banks.

Finally, for each subject, a total left and total right perisylvian sulcal mesh area was extracted using the sulcal mesh data in BV. This sulcus mesh area is equal to the sum of the areas of triangles composing each sulcus-specific smooth mesh (Mangin et al. 2004) and therefore provides an estimate of the gyral bank area directly opposing both sides of the 3-D sulcal mesh. This total surface area value was determined for all labeled sulcal structures along the suprasylvian bank (relating to the second segment of the Sylvian fissure) and was done independent of a Sylvian fissure connection. This cumulative area included the following sulci: AAR, DgS, precentral sulcus (PeC) (all segments), CS, ASC, posterior subcentral sulcus (PSC), PoC (all segments), and variant ramus (VR) (Table 1).

Measurements of Asymmetry

Manipulation of imaging planes and calculation of Euclidean distances for each of the measures below were performed using the software Spatial Normalization (SN) as a visualization and an x-, y-coordinate–marking tool (Lancaster et al. 1995). Except as noted below, all right–left measures were used to compute a coefficient of asymmetry according to the formula: (R − L)/(0.5 × (R + L)) (Galaburda et al. 1987).

Global, spatial normalization is a linear normalization procedure involving an affine transformation consisting of translation, rotation, and scaling of the brain (Lancaster et al. 1995). The major advantage of the process is that intersubject differences in brain position, orientation, and overall size are accounted for. Unlike nonlinear transformations, this process does not involve regional deformations of the subject brains (Ashburner and Friston 1999). Linear normalization does not affect intrasubject cerebral asymmetry patterns nor individual subject sulcal and gyral anatomy. There are a number of studies reporting expected intrasubject asymmetry patterns for a variety of measures after transformations normalizing brain structure (Thompson et al. 1998; Narr et al. 2001; Watkins et al. 2001; Sowell et al. 2002). Comparable anatomical studies of developmental stutterers (Sommer et al. 2002; Jancke et al. 2004) have also utilized spatial normalization prior to group comparisons.

Planum Temporale

An imaging plane was created through the chiasmatico-commissural plane (CH-PC) that correlates with the plane of the Sylvian fissure used by Geschwind and Levitsky in their study of PT asymmetry (Geschwind and Levitsky 1968). This allows clear identification of the entirety of supratemporal structures (Tamraz and Comair 2006). Visualizing each hemisphere simultaneously in axial, coronal, and sagittal views, the maximal rostrocaudal y-coordinate distance of the PT was determined using the CH-PC plane. In the few cases where the CH-PC plane did not correlate with the Sylvian plane, the image was adjusted to correlate with the Sylvian plane prior to measurement. The measurement of PT length followed the procedure used by numerous authors (Geschwind and Levitsky 1968; Galaburda et al. 1987; Steinmetz 1996). The labeled hemisphere of each subject, as exemplified in Figure 1(A,B) was used as a reference for subject-specific landmark determination in the SN program. The anterior border was determined as the posterior TSS. In cases of anatomical variability where more than one transverse temporal gyrus (Heschl's gyrus) was present, the second gyrus was included in the rostrocaudal length. Therefore, the measurement followed modifications outlined by Galaburda et al. (1987). This is considered an accurate estimate of the PT size because cytoarchitectonic evidence suggests that additional Heschl's gyri caudal to the first gyrus contain auditory association cortex and not primary auditory cortex (Shapleske et al. 1999). The posterior border of the region was defined as the origin of the PDR of the Sylvian fissure (Geschwind and Levitsky 1968; Galaburda et al. 1987; Steinmetz 1996). In hemispheres where there was no PDR in the configuration of the third segment of the Sylvian fissure, the entire supratemporal extent of the planum was included (Tamraz and Comair 2006).

Sylvian Fissure

The rostrocaudal length of the Sylvian fissure was measured as the maximum distance between y-coordinates. The posterior y-coordinate of the Sylvian fissure was determined with the normalized brain in the anterior commissure-posterior commissure (AC-PC) reference plane by tracing successive coronal slices to the end of PAR of the Sylvian fissure or to the end of the terminal horizontal Sylvian fissure segment caudal to the TSS in cases where there was no clear PAR. The anterior coordinate used was a line drawn perpendicular to the AC-PC plane from the origin of either the CmS or the AAR, depending on the sulcal configuration present. The labeled reference image in BV was again used for correct determination of landmarks during this measurement. Because the presence of a CmS or AAR are among the most stable landmarks in imaging the lateral frontal lobe (Juch et al. 2005), these proved to be reliable reference points. A second anterior point was chosen at the approximate point where the stem and second portions of the Sylvian fissure meet (Yeni-Komshian and Benson 1976) to rule out the possibility that individual variations in pars triangularis (and therefore CmS and AAR) configuration might bias the result.

Occipital and Frontal Lobe Asymmetry

The most inferior axial plane that disclosed both the frontal horns of the lateral ventricles and the collateral trigone of each ventricle (Chui and Damasio 1980) was established for each subject. The anteroposterior (AP) diameter of the brain was calculated in this plane using the most anterior and posterior y-coordinates. At distances of 90% and 16% of the total AP diameter, the frontal and occipital lobe widths were determined, respectively, as the distance between x-coordinates. Asymmetry coefficients were calculated for width measures according to (R − L)/(R + L) (Chui and Damasio 1980). The y-coordinate distance between the left and right frontal and left and right occipital poles was determined as a measure of petalia. Only differences >1 voxel (of 0.8 mm) were classified as leftward or rightward petalia with the remainder classified as symmetrical. This was a modification of the “accuracy of the millimeter rule” used in previous studies of petalia (Chui and Damasio 1980).

Statistical Analysis and Reliability Measures

To test measurement reliability, the original observer (M.C.) repeated the complete set of measures as outlined in the methods on 6 subjects (3 from each group), blind to subject status and age. Intraobserver reliability was assessed (utilizing the Pearson correlation coefficient) for quantification of sulcal connections and the related gyral bank number (r = 0.88), for measurements of petalia (r = 0.92), for measurements of width asymmetry (r = 0.99), for measurements of PT length asymmetry (r = 0.97), and Sylvian fissure length asymmetry (r = 0.99). In addition, paired t-tests were performed on the values recorded during reliability testing, and the original values reported for the same measures on each of the 6 subjects. There were no significant differences between the repeated and original measures for the mean number of sulcal connections (4.92 vs. 5.0; P = 0.59) or the mean number of gyral banks (6.33 vs. 6.50; P = 0.34) for the 12 hemispheres of the 6 subjects utilized in reliability testing. Paired t-tests of repeated and original asymmetry measures on the 6 subject brains utilized were also nonsignificant (all P ≥ 0.2) for occipital petalia (3.67 vs. 3.5 voxels; P = 0.70), frontal petalia (4.17 vs. 3.5 voxels; P = 0.24), frontal width asymmetry coefficients (−0.006 vs. −0.008; P = 0.84), occipital width asymmetry coefficients (−0.019 vs. −0.013; P = 0.23), PT asymmetry coefficients (−0.14 vs. −0.21; P = 0.37), and Sylvian fissure asymmetry coefficients (−0.15 vs. −0.14; P = 0.46). Therefore, intraobserver differences were nonsignificant and were not considered a threat to the validity of the findings.

The quantitative measures of the number of sulcal connections and gyral bank number, perisylvian sulcal mesh area, and all asymmetry coefficients for both groups were compared with a 2-tailed Student's t-test (Zar 1996) assuming unequal variance. The absolute voxel number difference between right and left frontal and occipital poles (i.e., petalia) was similarly compared between PDS subjects and controls. Because the group sizes were relatively small, a correction for multiple comparisons was not performed, and therefore, the results must be considered exploratory.

Results

Sulcal Connections and Gyral Bank

A greater number of sulcal connections to the second segment of the right Sylvian fissure were seen (t = 2.05, P ≤ 0.04, 2-tailed Student's t-test, unequal variance) in PDS subjects compared with controls. As shown in Table 2, developmental stutterers averaged nearly one additional connecting sulcus along this segment of the right Sylvian fissure when compared with controls. A similar comparison of left-to-left connecting sulci between PDS subjects and controls was nonsignificant. A greater number of right perisylvian gyral banks were observed in the PDS subjects (t = 2.05, P ≤ 0.02), also shown in Table 2. The same comparison was not significant in comparing left-to-left perisylvian gyral bank numbers between controls and PDS subjects. Comparing left-to-right connecting sulci in the control group, a significant difference was found (t = 2.07, P ≤ 0.02), with nearly one additional connecting sulcus along the left Sylvian fissure. No similar left > right difference occurred for the PDS group for the number of connecting sulci.

Table 2

Mean (standard deviation) of measures in PDS subjects and controls

 Left perisylvian region Right perisylvian region 
 PDS subjects Controls PDS subjects Controls 
Perisylvian measurements     
    Number of sulcal connections 5.32 (0.48) 5.13 (0.62)a 5.11 (0.94)b 4.31 (1.20)b 
    Number of gyral banks 6.74 (0.65) 6.69 (0.80) 6.42 (0.96)b 5.56 (1.15)b 
    VR present 0.95 (0.71) 0.88 (0.72) 0.79 (0.63) 0.63 (0.72) 
    Surface area of all sulci (mm215185.0 (1498.0) 14633.4 (2499.7) 14765.0 (1302.3) 13851.5 (1818.1) 
Asymmetry coefficients PDS subjects  Controls  
    PT −0.243 (0.464)  −0.248 (0.357)  
    Sylvian fissure from AAR −0.136 (0.097) −0.149 (0.109) 
    Sylvian fissure from temporal pole −0.127 (0.094) −0.144 (0.075) 
    Occipital width −0.020 (0.023) −0.022 (0.022) 
    Frontal width 0.027 (0.034) 0.006 (0.031) 
 Left perisylvian region Right perisylvian region 
 PDS subjects Controls PDS subjects Controls 
Perisylvian measurements     
    Number of sulcal connections 5.32 (0.48) 5.13 (0.62)a 5.11 (0.94)b 4.31 (1.20)b 
    Number of gyral banks 6.74 (0.65) 6.69 (0.80) 6.42 (0.96)b 5.56 (1.15)b 
    VR present 0.95 (0.71) 0.88 (0.72) 0.79 (0.63) 0.63 (0.72) 
    Surface area of all sulci (mm215185.0 (1498.0) 14633.4 (2499.7) 14765.0 (1302.3) 13851.5 (1818.1) 
Asymmetry coefficients PDS subjects  Controls  
    PT −0.243 (0.464)  −0.248 (0.357)  
    Sylvian fissure from AAR −0.136 (0.097) −0.149 (0.109) 
    Sylvian fissure from temporal pole −0.127 (0.094) −0.144 (0.075) 
    Occipital width −0.020 (0.023) −0.022 (0.022) 
    Frontal width 0.027 (0.034) 0.006 (0.031) 

Note: Perisylvian measurements related to the second segment of the Sylvian fissure only. Negative asymmetry coefficient values indicate leftward asymmetry and positive values indicate rightward asymmetry.

a

This was significantly higher (P < 0.05) than the number of right-sided connections within the control group.

b

PDS subjects showed a slight increase (P < 0.05) in the number of connecting sulci to the right Sylvian fissure and an increase in the number of gyral banks in the frontal and parietal opercula.

Table 3 shows the percentage of PDS and control subjects who displayed particular perisylvian sulcal patterns as compared with available references (Ono et al. 1990; Juch et al. 2005; Tamraz and Comair 2006). There were no definitive abnormalities for individual sulcal configurations detected, though a greater number of PDS subjects showed tertiary VR in the bilateral parietal opercula, along with a decrease in the percentage of PDS subjects with VR in the left rolandic operculum. Among primary sulci, the only apparent difference between the 2 groups was seen in the proportion of PDS subjects displaying a connection of the right inferior precentral sulcus to the Sylvian fissure (PDS 53%, controls 25%), though both groups fell within or near the wide reference range available from the literature.

Table 3

Perisylvian sulcal patterns in PDS subjects and controls (%)

 Left perisylvian region (%) Right perisylvian region (%) 
 PDS subjects Controls Reference rangea PDS subjects Controls Reference rangea 
Configuration of sulci related to the pars triangularis and opercularis       
    Separate AAR and AHRb 74 69 60-78 63 75 64-69 
    CmS for AAR and AHRb 26 25c 24-33 37 25 22-33 
    DgS present 58 69 70-72 58 63 64-70 
Sulci with a connection to the Sylvian fissure (second segment)       
    DgS 26 19 8-43 11 13 0-26 
    IPeC 47 44 16-43 53 25 28-57 
    CS 4-16 11 13-16 
Inferior postcentral (IPoC)d 47 56 30-48 53 38 40-68 
Variant/tertiary ramus patterns       
    Secondary tertiary ramus crossing pars opercularise 11 13 NR NR 
    Variant tertiary rami deep in frontal operculum 21 19 NR 16 19 NR 
    Variant tertiary rami deep in rolandic operculum 26 50 NR 21 31 NR 
    Variant tertiary rami deep in parietal operculum 47 19 NR 42 13 NR 
 Left perisylvian region (%) Right perisylvian region (%) 
 PDS subjects Controls Reference rangea PDS subjects Controls Reference rangea 
Configuration of sulci related to the pars triangularis and opercularis       
    Separate AAR and AHRb 74 69 60-78 63 75 64-69 
    CmS for AAR and AHRb 26 25c 24-33 37 25 22-33 
    DgS present 58 69 70-72 58 63 64-70 
Sulci with a connection to the Sylvian fissure (second segment)       
    DgS 26 19 8-43 11 13 0-26 
    IPeC 47 44 16-43 53 25 28-57 
    CS 4-16 11 13-16 
Inferior postcentral (IPoC)d 47 56 30-48 53 38 40-68 
Variant/tertiary ramus patterns       
    Secondary tertiary ramus crossing pars opercularise 11 13 NR NR 
    Variant tertiary rami deep in frontal operculum 21 19 NR 16 19 NR 
    Variant tertiary rami deep in rolandic operculum 26 50 NR 21 31 NR 
    Variant tertiary rami deep in parietal operculum 47 19 NR 42 13 NR 

Note: IPeC, inferior precentral sulcus; NR, no reference range available regarding these measures.

a

Reference ranges for previously investigated patterns from several studies (Ono et al. 1990; Grefkes et al. 2001; Juch et al. 2005; Tamraz and Comair 2006).

b

Range includes estimates interpreted from the classification of Juch et al. (2005). “V,” “U,” “J” configurations (Juch et al. 2005), grouped here as separate AHR and AAR and “Y” configuration as having a CmS.

c

One control subject without AHR present (6%).

d

One reference range (Ono et al. 1990) includes cases where this sulcus contacts the third segment of the Sylvian fissure, whereas these frequencies are only for the second segment.

e

See Discussion for comment.

Surface Area of Sulcal Meshes

A cumulative surface area value was determined from the sum of all perisylvian sulci relating to the second segment of the Sylvian fissure. This was done for all labeled sulcal structures along the suprasylvian bank independent of a Sylvian fissure connection and included the AAR, DgS, PeC, CS, ASC, PSC, PoC, and VR (Table 1). As Table 2 shows, neither the right-to-right comparison (t = 2.05, P = 0.10) nor the left-to-left comparison (t = 2.06, P = 0.45) between controls and PDS subjects reached significance.

Asymmetry Measurements

None of the comparisons of coefficients of asymmetry revealed significant differences between PDS subjects and controls (see Table 2). Though a greater rightward asymmetry of frontal width in PDS subjects was observed, as reflected in a comparison of the mean value for PDS subjects (mean = 0.027) to controls (mean = 0.006), this comparison did not quite reach significance (t = 2.03, P = 0.06).

Table 4 displays the frequency of leftward and rightward asymmetry, as well as symmetry, for all measures in this study. The table caption describes how the continuous variable (asymmetry coefficient) was classified as leftward, rightward, or symmetric for each measure and displays a reference range derived from several similar studies. The Kolmogorov–Smirnov (KS) goodness of fit test was applied to the frequencies of asymmetry observed in PDS subjects. Control frequencies observed in the study were used to determine the frequency expected for the PDS group (n = 19, k = 3). The KS test is the most appropriate test to compare observed and expected frequency distribution in the PDS group because the data were ordered (by asymmetry coefficient from negative to positive values) and the sample size is relatively small (Zar 1996). The KS test showed a statistically significant frequency distribution for the width of the occipital lobe (n = 19; k = 3; dimax = 6.27; P = 0.02) (Table 4). However, a 2-tailed t-test comparing the mean asymmetry coefficients of PDS subjects and controls (see Table 2) was not significant (t = 2.04; P = 0.74). Other frequency distribution differences between the groups were found to be nonsignificant after evaluation with the KS goodness of fit test.

Table 4

Cerebral asymmetry patterns in PDS subjects and controls (%)

 PDS subjects Controls Reference ranges (%)ab 
 Sc Sc Sc 
PT 63 26 11 75 12.5 12.5 62-83.3 5-23 8.3-25 
Sylvian fissure 74 21 75 25 75-84 3-12 4-17 
Frontal width 11 63 26 25 50 25 9-22 36-70 21-42 
Occipital width 78 11 11 56 44 36-67 5-20 20-44 
Frontal petalia 26 53 21 25 50 25 8-28 36-53 23-56 
Occipital petalia 74 21 62 19 19 53-68 8-27 20-25 
 PDS subjects Controls Reference ranges (%)ab 
 Sc Sc Sc 
PT 63 26 11 75 12.5 12.5 62-83.3 5-23 8.3-25 
Sylvian fissure 74 21 75 25 75-84 3-12 4-17 
Frontal width 11 63 26 25 50 25 9-22 36-70 21-42 
Occipital width 78 11 11 56 44 36-67 5-20 20-44 
Frontal petalia 26 53 21 25 50 25 8-28 36-53 23-56 
Occipital petalia 74 21 62 19 19 53-68 8-27 20-25 

Note: L, leftward (negative asymmetry coefficient [AC]); R, rightward (positive AC); S, symmetrical.

a

Healthy population reference ranges from magnetic resonance studies reporting data for right-handed healthy males for PT asymmetry (Kulynch et al. 1994; Rossi et al. 1994; Heiervang et al. 2000).

b

Healthy population reference ranges for additional measures include several studies of mixed handedness and gender where right-handed male data could not be extracted. Ranges are for Sylvian fissure asymmetry (estimated from Yeni-Komshian and Benson [1976, p. 388], Fig. 1; Falkai et al. [1992, p. 29] Fig. 2; Rubens et al. 1976), frontal/occipital width, and petalia (Gundara and Zivanovic 1968; LeMay [1976], Table 2, p. 354; Lemay [1977], Table 1, p. 246; Chui and Damasio 1980; Galaburda et al. 1978; Kennedy et al. 1999).

c

Symmetry for the PT and Sylvian fissure determined as −0.1 < AC < +0.1 (i.e., >10% interhemispheric difference is asymmetric according to Galaburda et al. [1987]). Symmetry for width and petalia measures approximates the “rule of a millimeter” (Chui and Damasio 1980) where symmetry equals either a 0 or 1 voxel (0.8 μm) difference between hemispheres.

Finally, to compare petalias of the frontal and occipital lobe, the mean absolute voxel difference for each group was compared for both frontal and occipital poles. For frontal pole asymmetry (PDS mean = 1.68; control mean = 1.19), this comparison was not significant (t = 2.04, P = 0.69). In both groups, the mean voxel difference between frontal poles represented a more anterior right frontal pole, as Table 4 reflects. A quantitative comparison of absolute voxel difference for occipital pole asymmetry (PDS mean = 1.74; control mean = 2.06) was also nonsignificant (t = 2.04, P = 0.80) and leftward occipital petalia was common to both groups.

Correlation of Stuttering Rate and Anatomical Findings

Post hoc analyses were performed using Pearson correlations to evaluate the relationship between the rates of stuttered speech and select structural features described above. The maximum percentage of syllables stuttered, during either the subject's oral reading or monologue sample, was used in the correlation analysis (see Appendix 2).

A one-tailed test was used to determine critical values of the correlation coefficient, r, for structural features that appear to be increased in association with PDS (i.e., the presence of the feature is associated with the condition). This included the number of sulcal connections to the right Sylvian fissure, rightward frontal width asymmetry, right perisylvian sulcal mesh area (independent of connection to the Sylvian fissure), and the frequency of right inferior precentral sulcus connection to the Sylvian fissure.

The correlation was weak (−0.12 < r < 0.12) for rightward frontal width asymmetry and right perisylvian total sulcal mesh area. There was a moderate, positive correlation for the number of sulcal connections to the second segment of the right Sylvian fissure (r = 0.28; P = 0.12). A stronger correlation existed for the frequency of the inferior precentral sulcus connection to the Sylvian fissure and the maximum percentage of syllables stuttered (r = 0.37, P = 0.055). To further evaluate these positive correlations in PDS subjects, the number of primary sulci and nonprimary sulci (Table 1) observed to connect to the right Sylvan fissure in each subject were correlated separately with stuttering severity. For primary sulci, the relationship was a significant positive correlation (r = 0.40, P = 0.04). For nonprimary sulci (e.g., tertiary variant rami and subcentral sulci), there was no such direct, positive correlation (r = −0.22).

A 2-tailed hypothesis was applied for correlating the structural features of the left perisylvian region and PT with stuttering severity. Though our study found these features not to be significantly different in PDS, abnormalities in these features have been reported (Foundas et al. 2001, 2004; Jancke et al. 2004). The number of sulcal connections to the second segment of the left Sylvian fissure did not significantly correlate with the maximum %SS (r = 0.05; P = 0.83), nor did the degree of PT asymmetry (r = −0.19; P = 0.42; severity correlated weakly with leftward asymmetry).

Discussion

A slightly greater total number of sulci making connections with the right Sylvian fissure (P ≤ 0.04) was seen in PDS subjects. However, the same comparison of left perisylvian sulcal anatomy was not significant. Using sulci as natural divisions between the opposing gyral banks of those sulci, a slight increment in the number of gyral banks along the right suprasylvian segment was seen in PDS subjects (P < 0.02). The right-to-right comparison of mesh area for all sulcal structures (independent of connection to the Sylvian fissure) revealed a higher average surface area in the right perisylvian region of PDS subjects but this did not reach significance (P = 0.10).

In regard to cerebral asymmetries, quantitative comparisons between controls and PDS subjects were nonsignificant for frontal and occipital asymmetry, PT asymmetry, Sylvian fissure asymmetry, and frontal/occipital petalias. A KS goodness of fit test revealed no significant difference between the frequencies of leftward and rightward asymmetry and symmetry in these structures (Table 4) with the exception of occipital width. However, leftward occipital width was seen in both groups, and the parametric comparison of the asymmetry coefficient was not significant. The qualitative difference in leftward asymmetry of occipital width may be an artifact of the small group sizes utilized in this study, but this measure should be revisited with a larger sample size.

Perisylvian Morphology in PDS and Controls

Previous analyses of structural differences in developmental stuttering have reported sulcal configuration abnormalities and “extra gyri” (Foundas et al. 2001, p. 212) in the bilateral perisylvian opercula of the frontal and parietal lobes of some developmental stutterers. One DTI study of PDS and control groups reported a decrease in the value of fractional anisotropy for white matter in the left rolandic operculum for the PDS group (Sommer et al. 2002). In addition to structural imaging studies, functional studies have reported abnormal right-sided activations in perisylvian cortex (particularly frontal and rolandic opercula) during stuttered speech (Fox et al. 1996; Brown et al. 2005). Therefore, it was important to devise a quantitative method of assessing perisylvian sulcal configuration and gyrification. The reasoning behind this approach was that if there were increased perisylvian cortical folding as previously reported (Foundas et al. 2001), then there should be an increase in the numbers of Sylvian-connecting sulci between opposing gyral banks.

Placing our results in the context of previous reports on cerebral morphology in PDS, they may actually be most consistent with findings by Jancke et al. (2004). In spite of different methods utilized in that study (VBM), a right-lateralized increase was detected in the volume of frontal lobe white matter (Jancke et al. 2004). This included white matter increases in the right pars opercularis of the inferior frontal gyrus and the inferior right precentral gyrus near the right rolandic operculum. Those findings may be related to the increase in cortical folding of the right perisylvian region seen here. However, the results presented here do not completely overlap with several features previously reported in other studies of neuroanatomic features in PDS. We do not find anomalies bilaterally in the perisylvian region as reported by Foundas et al. (2001), which included increased gyrification bilaterally, a doubled DgS, and atypical configurations of the sulci bounding the pars triangularis portion of the inferior frontal gyrus. As Tables 2 and 3 show, the present study did not find these features to be atypical in the PDS group. In the present study, the increased number of sulcal connections seen was limited to the right perisylvian region only, and the between-group differences reported here are quite subtle (see Fig. 2A). In addition, our control and PDS subjects showed some overlap in the quantitative measures of the right perisylvian region for individual subjects and only with group comparisons did the differences observed emerge.

Figure 2.

(A) Ten right-hemispheric meshes are shown (5 separate PDS subjects and 5 control subjects) with the right perisylvian region magnified. The whole brain shown at the top illustrates the right perisylvian region as magnified in the 10-subject insets shown below. For orientation, the AAR is marked by an asterisk, whereas the PAR, or terminal segment of the Sylvian fissure, is marked by a solid white arrow on each separate subject inset. Note that not all sulcal connections are apparent when simply viewing the convex surface of the mesh (e.g., detecting the variant ramus in Fig. 1B vs. A) because some arise from deeper in the Sylvian fissure (see also 2B). The PDS (6 ± 1 average connections, range of 5–7) and control subjects (5.4 ± 0.54, range of 5–6) selected for Figure 2A were at the high end of the number of true sulcal connections with the second segment of the right Sylvian fissure for both groups (the 5 controls shown had the 5 highest values of all controls in the study). Using labeled Figure 1(A,B) as a reference, note that both a considerable overlap of gyrification patterns between the 2 groups is apparent, and subtle differences found in PDS subjects may be detected (i.e., the frequency of PeC connection to the Sylvian fissure and the general contribution of larger, primary sulci to the total number of sulcal connections in the PDS group). (B) Composite images of both groups (16 age-matched subjects selected from the PDS group were utilized to match the complete set of 16 controls in creating the composite images of 2B). The legend given below the images applies to both groups. These composite images represent the extraction of the meshes of BV-labeled perisylvian sulci for meshes that connected to the Sylvian fissure (see Fig. 1A for a reference of sulcal meshes merged with the hemispheric mesh). Except for the second segment of the Sylvian fissure (whose mesh is simplified from only several representative subjects per group so that small rami are not obscured), the composite images are overlays of all connecting sulci of each group of 16. For example, a single CS is present for both composite images because one subject in each group displayed this feature (i.e., true connection of the CS to the Sylvian fissure).

Figure 2.

(A) Ten right-hemispheric meshes are shown (5 separate PDS subjects and 5 control subjects) with the right perisylvian region magnified. The whole brain shown at the top illustrates the right perisylvian region as magnified in the 10-subject insets shown below. For orientation, the AAR is marked by an asterisk, whereas the PAR, or terminal segment of the Sylvian fissure, is marked by a solid white arrow on each separate subject inset. Note that not all sulcal connections are apparent when simply viewing the convex surface of the mesh (e.g., detecting the variant ramus in Fig. 1B vs. A) because some arise from deeper in the Sylvian fissure (see also 2B). The PDS (6 ± 1 average connections, range of 5–7) and control subjects (5.4 ± 0.54, range of 5–6) selected for Figure 2A were at the high end of the number of true sulcal connections with the second segment of the right Sylvian fissure for both groups (the 5 controls shown had the 5 highest values of all controls in the study). Using labeled Figure 1(A,B) as a reference, note that both a considerable overlap of gyrification patterns between the 2 groups is apparent, and subtle differences found in PDS subjects may be detected (i.e., the frequency of PeC connection to the Sylvian fissure and the general contribution of larger, primary sulci to the total number of sulcal connections in the PDS group). (B) Composite images of both groups (16 age-matched subjects selected from the PDS group were utilized to match the complete set of 16 controls in creating the composite images of 2B). The legend given below the images applies to both groups. These composite images represent the extraction of the meshes of BV-labeled perisylvian sulci for meshes that connected to the Sylvian fissure (see Fig. 1A for a reference of sulcal meshes merged with the hemispheric mesh). Except for the second segment of the Sylvian fissure (whose mesh is simplified from only several representative subjects per group so that small rami are not obscured), the composite images are overlays of all connecting sulci of each group of 16. For example, a single CS is present for both composite images because one subject in each group displayed this feature (i.e., true connection of the CS to the Sylvian fissure).

The term “extra gyri” (Foundas et al. 2001) was previously used to describe perisylvian morphology in PDS subjects. We refrain from using that term to describe our findings because this phrase can imply “gross polymicrogyria” present in several developmental disorders (Schmitt et al. 2002; Piao et al. 2004), and this finding was not present either quantitatively or qualitatively in the subjects of this study. In addition, the tertiary sulci that were extracted by our automated method do not always make clear, deep boundaries between perisylvian gyri as the term extra gyri implies but rather indent a gyrus in the manner by which variant (and sometimes deep) medial frontal sulci can cross a uniform superior frontal gyrus (Ono et al. 1990).

There were no definitive sulcal configuration abnormalities (see Table 3 and Fig. 2) seen in either PDS subjects or controls, though some interesting group trends may be worth further investigation. Qualitative differences observed in the PDS group included a higher frequency of variant tertiary rami deep in the left and right parietal opercula and a lower frequency of variant rami present in the left rolandic operculum. The initial formation of tertiary rami begins around 36 weeks and continues into the postnatal period, likely under the result of the “same maturational processes” that deepen primary sulci and form secondary sulci (Armstrong et al. 1995, p. 61). Becaue tertiary sulci “appear random in their form and anatomical relationships” according to some authors (Richman et al. 1975, p. 18), and the forces causing their appearance are still undefined, the significance of the variability in tertiary rami seen between groups in the study is unclear. Interestingly, we did find a secondary tertiary ramus crossing the pars opercularis once the DgS had already been identified (see Table 3), and this may be the same structure previously reported as a doubled DgS in PDS subjects (Foundas et al. 2001). However, this secondary tertiary ramus originated from the Sylvian fissure, the AAR, or the CmS and was present in a small group of both control and PDS subjects. Though some studies have mentioned the presence of variant perisylvian rami or accessory gyri in the inferior parietal lobule of normal subjects (Naidich et al. 1995), the frequency of perisylvian sulcal variants in the normal population has not been well characterized. In post hoc correlation analyses, the number of secondary and tertiary rami in the right perisylvian region of PDS subjects did not appear to correlate with stuttering severity. However, these qualitative differences in later-developing sulcal structures may be related to the disorder, though not necessarily to stuttering severity.

The only apparent large separation in the patterns of primary sulci was seen in the right inferior precentral sulcus, with stutterers more commonly having a true connection (53%) between the inferior precentral sulcus and the Sylvian fissure (see Table 3 and Fig. 2B). This connection of the right inferior precentral sulcus was also positively correlated with stuttering severity (r = 0.37) in the PDS group, as was the total number of primary sulci connecting to the right Sylvian fissure (r = 0.40). These correlations are interesting in light of functional abnormalities described in the region. In male developmental stutterers, activity in the right inferior lateral premotor area (Brodmann area 44/6), among other regions, has been positively correlated with stutter rate. Local maxima in activations seen during stuttered speech have been identified where the right inferior precentral sulcus crosses the pars opercularis (x = 44, y = 22, z = 1 in Talairach space) (Ingham et al. 2004). The potential structural/functional correlation here is intriguing and deserves further investigation in studies assessing cortical thickness and fiber organization in the area of the right frontal operculum. However, both groups in this study fall within the wide range of previously published data for the inferior precentral sulcus (see Table 3), although our control population had a frequency of right inferior precentral sulcus connection (25%) similar to that of 28% observed in one atlas (Ono et al. 1990). Therefore, the significance of both this qualitative separation and the correlation analysis for the right PeC with stuttering severity are unclear until future studies can better characterize the relationship between the inferior precentral sulcus and Sylvian fissure in normal populations.

Finally, as Table 3 shows, the control subjects in this study are within reference standards for most patterns assessed, though in several cases they are slightly out of range. There are few standard ranges for whole-brain sulcal frequencies with high-resolution magnetic resonance, and those in the standard atlas (Ono et al. 1990) were determined from a study of gross brains. This may account for those instances of variability between perisylvian sulcal patterns observed in this study and the few reference studies available.

Implication of Perisylvian Findings in Both PDS and Control Groups

The findings of this study suggest a small, right-lateralized increase in perisylvian sulcal development and gyrification in PDS subjects, but the present data cannot provide a mechanism for how this came about. Postnatal structural plasticity has been reported that correlates with unique patterns of functional activity. These reports have included an increase in cortical thickness in the compensating sensorimotor cortex of chronic stroke patients (Schaechter et al. 2006) and in the cortical volume of the motor cortex (reflected by a CS depth increase) in musicians related to fine bimanual coordination, practice intensity, and experience (Amunts et al. 1997; Gaser and Schlaug 2003). Amunts et al. (1997) suggested that long-term microstructural changes in gyri (e.g., increase in the number of synaptic contacts and glial cells) can lead to detection by macrostructural methods. This interpretation of our current data may be relevant because the onset of stuttering behavior can occur prior to 3 years of age (Yairi and Ambrose 1992), and a lifetime of abnormal right-hemispheric involvement in speech tasks could plausibly lead to macrostructural changes.

Alternatively, in a disorder with a complex genetic basis (Riaz et al. 2005), it could be hypothesized that the results seen here have some relationship to events taking place during cortical development. For instance, the increase in right perisylvian sulcal development in PDS subjects may be influenced by larger, deeper, and earlier developing sulci, such as the right PeS and PoS (see Table 3 and Fig. 2B), which both had a slight qualitative increase in connectivity with the Sylvian fissure relative to the control group. On the other hand, the nonsignificant difference between the right perisylvian sulcal mesh areas of PDS subjects and controls (P = 0.10), independent of Sylvian connectivity, does not completely support that argument. Nonetheless, deep primary sulci have been shown to have a genetic component to depth and geometry (Lohmann et al. 1999; Molko et al. 2003), whereas younger, tertiary sulci appear to be more susceptible to nongenetic factors (Richman et al. 1975; Lohmann et al. 1999). Monozygotic twin studies have reported that for primary sulci there is a strong genetic influence on “the ontogenetic protomap of sulcal fundi” (Lohmann et al. 1999, p. 763); so if there are abnormal qualitative patterns in the development or placement of primary sulci, then there may be some developmental component worth further consideration. Though the current study is limited in its ability to answer the question of structural plasticity versus developmental abnormality, the question remains intriguing and warrants additional studies of cerebral morphology in PDS subjects, particular with methods capable of more focal comparisons in the right perisylvian region.

Finally, the control group in this study exhibited a significant left-to-right difference in the number of connecting sulci along the second segment of the Sylvian fissure (Table 2). This novel feature of cerebral asymmetry has not been previously reported in studies of Sylvian fissure and perisylvian morphology in healthy populations. However, several related studies predict the outcome obtained in the current study. Leftward asymmetry in absolute Sylvian fissure length is well established (Yeni-Komshian and Benson 1976) so that more cumulative connections would be expected along the left suprasylvian bank. In one postmortem study (Ide et al. 1999), an additional accessory PoC interposed between the PoC and the ascending ramus of the Sylvian fissure was described as a relatively consistent feature of left hemispheres. Our findings are consistent with this observed increased fissurization in the left perisylvian area, described as a potential consequence of increased surface area of the cortical sheet in the region (Ide et al. 1999). There is also indirect evidence that supports the finding of a left-to-right difference in our control group. Ono et al. (1990) described a longer left second Sylvian segment as well as a greater intersulcal distance along the right second Sylvian segment, both of which would contribute to the finding reported here. Witelson and Kigar (Witelson and Kigar 1992) reported a left “horizontal segment” of the Sylvian fissure, similar to the segment studied here, which was twice as long as the homologous segment on the right. Combining the anterior and horizontal lengths given in their study (because their “anterior segment” includes the sulci of the frontal operculum considered here) reveals a left value 1.35 times that on the right. These previous reports of Sylvian fissure morphology, while not addressing the novel measurement used here (additive true connections to the second segment), would predict the outcome seen in the left-to-right comparison of perisylvian sulci in our control population. Finally, this potential gyrification asymmetry in healthy controls may be related to the dynamic relationship between Sylvian fissure asymmetry and perisylvian gray matter asymmetry seen between childhood and adulthood by Sowell et al. (2002).

Comparisons of Cerebral Asymmetry

Several investigations of structural asymmetry in PDS have arrived at the conclusion that PDS is associated with atypical cerebral asymmetry in structures with well-established patterns of asymmetry in normal populations (see reference ranges in Table 4). These results have been interpreted as a potential structural correlate of theories that developmental stutterers have atypical laterality of cerebral function (Strub et al. 1987; Foundas et al. 2003). However, within the present study, control and PDS subjects were highly similar in the degree and the distribution of cerebral asymmetry. Therefore, we cannot corroborate any link between atypical functional laterality under the “cerebral dominance theory” (Travis 1978) and quantifiable neuroanatomical correlates in regards to the PT, the Sylvian fissure, or the frontal and occipital lobes. However, the findings presented here (i.e., typical structural asymmetry) do not necessarily provide any evidence in conflict with the cerebral dominance theory of Travis, which simply states, “stuttering is caused by aberrant interhemispheric relationships” (Travis 1978, p. 278).

The subject selection in the present study may explain the difference between the cerebral asymmetry findings obtained here and those previously reported in analyses of PDS. One feature of the present study was that both stuttering and control groups were comprised entirely of dextral males which eliminated confounding issues of gender and handedness on neuroanatomical measurements of asymmetry (Witelson and Kigar 1992; Beaton 1997; Shapleske et al. 1999). Previous anatomical studies of asymmetry in PDS subjects included studies utilizing one ambidextrous male and one left-handed female (Strub et al. 1987). Other studies utilized matched groups of both left- and right-handed males, as well as right-handed females (Foundas et al. 2001, 2003). The fact that this study controlled the number of independent variables by limiting the analysis of asymmetry to right-handed males may account for the incongruities between the current and previous reports.

The similarity between PDS subjects and controls in the gross patterns of asymmetry observed here (leftward asymmetry of the PT and Sylvian fissure, occipital width and petalia, and rightward occipital width and petalia) argues against atypical structural laterality as a major feature of persistent developmental stuttering. It is notable that right-handed adult males, the group studied here, display the strongest and most consistent patterns of neuroanatomical asymmetry (Watkins et al. 2001). Therefore, they constitute an ideal group in which to detect disorder-related deviations from established patterns of asymmetry. Further, most cerebral asymmetries are believed to be present during gestation (Galaburda, LeMay, et al. 1978; Weinberger 1982; Preis et al. 1999) and in nonhuman primates (Yeni-Komshian and Benson 1976; Gannon et al. 1998). Therefore, cerebral asymmetry is not likely to be susceptible to gross postnatal modification, except possibly for the development of Sylvian fissure asymmetry (Sowell et al. 2002).

The only quantitative measure of asymmetry close to significance (P = 0.06) was a slight increase in the degree of rightward asymmetry in the frontal width of PDS. It should be noted that rightward asymmetry in the frontal lobe is in and of itself a normal trend in right-handed males (see Table 4). Nonetheless, this feature may warrant additional study with a larger subject pool and more sensitive techniques, particularly because both our study and that of Jancke et al. (2004) have described right-lateralized structural differences.

The present study utilized 2 separate measures of Sylvian fissure asymmetry to see if there were correlations with PT asymmetry and to place perisylvian sulcal connectivity results in the context of overall Sylvian fissure morphology. Notably, a comparison of left-to-right Sylvian fissure length asymmetry coefficients between groups was not significant for either measure (see Table 2). The PDS subjects in our study, therefore, had a subtle increase in sulcal connectivity and gyral bank number without a longer or atypical right Sylvian fissure.

The modifications applied in this study for generating an asymmetry coefficient of PT length and Sylvian fissure length were also based on well-established measures. Because “Sylvian asymmetries seem to reflect planum asymmetries” (Galaburda, Sanides, et al. 1978, p. 817), these related measures of PT and Sylvian fissure length were important to establish the general patterns of cerebral asymmetry in PDS adults. Leftward asymmetry of PT length was seen in both groups with a mean asymmetry coefficient of −0.248 in controls and −0.243 in PDS subjects. This measurement was modified from techniques measuring PT area as applied by Galaburda et al. (1987); in that study of 100 brains, the mean asymmetry coefficient was −0.27. In fact, the results obtained here in both groups are comparable to the original work of Geschwind and Levitsky (1968). The PT length measures in that study, following slicing in the Sylvian plane as performed here, were 3.6 ± 1.0 cm on the left and 2.7 ± 1.2 cm on the right. The asymmetry coefficient generated from these values would be −0.285, remarkably close to our own value in controls of −0.248. The homology between measurements of the PT here and the classic findings of Geschwind and Levitsky (1968) and Galaburda et al. (1987) suggest that the methodology of measuring PT length in the Sylvian plane is appropriate. Other authors have replicated the results of these classic studies (Geschwind and Levitsky 1968) using similar length measurements (Witelson and Pallie 1973; Rubens et al. 1976; Larsen et al. 1989). Witelson and Pallie (1973) found that their separate surface area and length measurements of the PT were highly correlated.

However, for PT measurements, no standard approach yet exists (Shapleske et al. 1999), and other approaches include surface area measurements, volumetric measurements, and tissue segmentation approaches. In other studies of developmental stutterers, both volumetric measurements (Foundas et al. 2001) and tissue segmentation–based approaches (Jancke et al. 2004) have been used. These may explain the differences between the current study and those previous studies. Nonetheless, the asymmetry coefficients reported in several studies utilizing alternate methods in healthy controls are remarkably similar to the result here using a length measure of the PT (−0.248 in controls). Similar asymmetry coefficient values include a value of −0.24 for right-handed men (Knaus et al. 2006; a volumetric measure), a coefficient of −0.25 in right-handed twins (Steinmetz et al. 1995; a surface area measure), and a mean coefficient of −0.25 in right-handed men (Kulynch et al. 1994; a surface area measure). (All referenced studies normalized for the sign of the asymmetry coefficient by using (R − L)/(0.5 × (R + L)) [Galaburda et al. 1987]).

As with the PT, we detected leftward asymmetry of Sylvian fissure length in both groups using both the AAR and the temporal pole as the most rostral points for the measure. The latter of these measures (using the temporal pole) was based on well-established measures of Sylvian length (Yeni-Komshian and Benson 1976). The asymmetry coefficient generated from data in that article (83.6 mm for the length of the left Sylvian fissure and 73.4 mm for the right), according to the formula used in this study, would be −0.130. As Table 2 shows, both PDS subjects and controls have mean asymmetry coefficients very similar to that standard (−0.127 for PDS subjects, −0.144 for controls). These findings suggest that the planimetric methods utilized here are capable of replicating well-established trends of cerebral asymmetry.

Methodological differences may account for the divergence of results here with previous studies of the supratemporal region in developmental stutterers (Foundas et al. 2001, 2004; Jancke et al. 2004). However, the convergence of PT and Sylvian fissure asymmetry with control and reference values (see Tables 2 and 4), and the similarity of the asymmetry coefficients with several major studies independent of methodological differences, suggests that reduced or reversed patterns of structural asymmetry in the supratemporal region may not be a feature of right-handed male developmental stutterers.

Potential Structure/Function Relationships in Developmental Stuttering

The severity of stuttering was correlated with several anatomical measures of interest derived from the present study. The functional measure of severity used was the maximum percentage syllables stuttered on either the reading or monologue speech sample (see Appendix note 2).

Stuttering severity had a moderately positive correlation with the finding of increased right perisylvian sulcal connection to the Sylvian fissure (r = 0.28) in PDS subjects. Even more substantial positive correlations between anatomical features and stuttering severity were observed for the Sylvian connection of the right inferior precentral sulcus in PDS subjects (r = 0.37) and the cumulative number of primary sulci connecting to the right Sylvian fissure (r = 0.40). These anatomical features are developed prior to 30 weeks gestation (Chi et al. 1977; Armstrong et al. 1995), though the exact developmental week is only approximate in the available literature (see Table 1). Finally, the connection of nonprimary sulci to the right Sylvian fissure was not positively correlated (r = −0.22) with the severity of stuttering in the subjects of this study.

These positive correlations suggest that the severity of the disorder may have some relationship to abnormal developmental processes in the right perisylvian region, though additional regions are likely involved as well. However, with the small sample size used here, the anatomical findings of the present study need to be replicated in larger samples and correlated likewise with the severity of the disorder. In addition, there are limitations that must be considered in trying to correlate static anatomical features with the dynamic processes involved in speech and the severity of dysfunction during stuttered speech. Nonetheless, this post hoc correlation analysis suggests that the differences described here may be associated with the disorder. Features previously reported as significantly different in PDS subjects (i.e., left perisylvian sulcation and PT asymmetry), and found to be not different from the control group in the present study, did not correlate with the severity of the disorder in our sample.

Study Limitations

One major limitation of the present study is the relatively small sample size. With several negative findings contradicting reports in previous literature, the question of whether the parametric tests had enough power may be raised. However, the detection of any significant difference in the quantitative measures studied here (i.e., right perisylvian sulcation) verifies that enough power exist for all the between-group comparisons, in spite of small sample sizes. Additionally, in spite of unequal group sizes, matching was performed according to handedness, gender, and age to generate representative groups of PDS subjects and controls, an approach that has been successful in many studies of the disorder (see references in Materials and Methods).

The second major limitation of the present study is that significant anatomical variation is present in the perisylvian region (Ono et al. 1990), and anatomical anomalies (e.g., neuronal ectopias in the molecular layer) are known to exist, though rarely, in healthy subjects (Kaufmann and Galaburda 1989). Future studies with larger healthy control and stuttering groups are therefore warranted to replicate the findings presented here, preferably, with automated measures of gyrification (Luders et al. 2006).

A third and final limitation of the present study is that the region of interest sampled here was limited by previous reports of anomalies in developmental stutterers, so the existence of other abnormalities cannot be ruled out. Features such as cortical thickness need to be assessed within the perisylvian and other regions known to function abnormally during stuttered speech. Additionally, the morphological features of both subcortical structures and the cerebellum have not been assessed in any studies of developmental stutterers, though there is known to be atypical activity in both during stuttered speech (Fox et al. 1996; Brown et al. 2005).

Conclusion

This analysis revealed more sulcal connectivity with the right Sylvian fissure in the PDS subject group. The structural findings also appeared to have a modest positive correlation with the severity of the disorder. Earlier developing primary sulci appeared to be the most significant component of that correlation. Though previously described as atypical features in PDS, we did not find definitive left-sided gross abnormalities in the perisylvian region or quantitative evidence of atypical cerebral asymmetry patterns. Additional information provided by other modalities, such as diffusion tensor imaging, will increase the understanding of the relationship between structure and function in persistent developmental stuttering.

Funding

National Institute on Deafness and Other Communication Disorders (NIDCDC; RO1 grant DC007893); National Institute of Biomedical Imaging and Bioengineering (NIBIB; PO1 grant EB001955); National Center for Research Resources (NCRR; MO1 grant RR001346).

Conflict of Interest: None declared.

Appendix 1

The Edinburgh handedness inventory had 10 items including the hand of preference for: writing, drawing, throwing, using scissors, using a toothbrush, using a knife, using a spoon, using a broom (upper hand), opening a box lid, and striking a match. Subjects were capable of answering the questions of preference as “exclusively right,” “exclusively left,” “usually right,” “usually left,” or “no preference.” Items marked as “exclusive” use of a hand scored as 2 in the appropriate column, and a score of 1 was assigned to the appropriate column when a particular hand was preferred but not to the exclusion of the other hand. Items where both hands were used were scored with a 1 in both the left and right-hand columns. The score for all items was added for each column (left/right), and the difference between the two was taken (left–right). This difference between total left and right scores was divided by the cumulative total of both columns and multiplied by 100. The total possible score on the inventory ranged from −100 (strongly left handed) to +100 (strongly right handed). Subjects were classified as left handed if the score was less than −40, right handed if greater than +40, and ambidextrous between −40 and +40.

Appendix 2

Severity of stuttering is typically based on an overall rating or on frequency of stuttering score (Bloodstein 1995). We selected the latter and the maximum score on either reading or monologue because PDS subjects may display maximum disability on either task—as was apparent in the data from PDS participants in this study.

References

Ambrose
NG
Cox
NJ
Yairi
E
The genetic basis of persistence and recovery in stuttering
J Speech Lang Hear Res
 , 
1997
, vol. 
40
 (pg. 
567
-
580
)
Amunts
K
Schlaug
G
Jancke
L
Steinmetz
H
Schleicher
A
Dabringhaus
A
Zilles
K
Motor cortex and hand motor skills: structural compliance in the human brain
Hum Brain Mapp
 , 
1997
, vol. 
5
 (pg. 
206
-
215
)
Armstrong
E
Schleicher
A
Omran
H
Curtis
M
Zilles
K
The ontogeny of human gyrification
Cereb Cortex
 , 
1995
, vol. 
5
 (pg. 
56
-
63
)
Ashburner
J
Friston
KJ
Nonlinear spatial normalization using basis functions
Hum Brain Mapp
 , 
1999
, vol. 
7
 (pg. 
254
-
266
)
Beaton
AA
The relation of planum temporale asymmetry and morphology of the corpus callosum to handedness, gender, and dyslexia: a review of the evidence
Brain Lang
 , 
1997
, vol. 
60
 (pg. 
255
-
322
)
Bloodstein
O
A handbook on stuttering
1995
5th ed
San Diego (CA)
Singular
Braun
AR
Varga
M
Stager
S
Schulz
G
Selbie
S
Maisog
JM
Carson
RE
Ludlow
CL
Altered patterns of cerebral activity during speech and language production in developmental stuttering. An H2(15)O positron emission tomography study
Brain
 , 
1997
, vol. 
120
 (pg. 
761
-
784
)
Brown
S
Ingham
RJ
Ingham
JC
Laird
AR
Fox
PT
Stuttered and fluent speech production: an ALE meta-analysis of functional neuroimaging studies
Hum Brain Mapp
 , 
2005
, vol. 
25
 (pg. 
105
-
117
)
Chi
JG
Dooling
EC
Gilles
FH
Gyral development of the human brain
Ann Neurol
 , 
1977
, vol. 
1
 (pg. 
86
-
93
)
Chui
HC
Damasio
AR
Human cerebral asymmetries evaluated by computed tomography
J Neurol Neurosurg Psychiatry
 , 
1980
, vol. 
43
 (pg. 
873
-
878
)
De Nil
LF
Kroll
RM
Houle
S
Functional neuroimaging of cerebellar activation during single word reading and verb generation in stuttering and nonstuttering adults
Neurosci Lett
 , 
2001
, vol. 
302
 (pg. 
77
-
80
)
Eckert
MA
Galaburda
AM
Karchemskiy
A
Liang
A
Thompson
P
Dutton
RA
Lee
AD
Bellugi
U
Korenberg
JR
Mills
D
, et al.  . 
Anomalous sylvian fissure morphology in Williams syndrome
Neuroimage
 , 
2006
, vol. 
33
 (pg. 
39
-
45
)
Falkai
P
Bogerts
B
Greve
B
Pfeiffer
U
Machus
B
Folsch-Reetz
B
Majtenyi
C
Ovary
I
Loss of sylvian fissure asymmetry in schizophrenia. A quantitative post mortem study
Schizophrenia Res
 , 
1992
, vol. 
7
 (pg. 
23
-
32
)
Foundas
AL
Bollich
AM
Corey
DM
Hurley
M
Heilman
KM
Anomalous anatomy of speech-language areas in adults with persistent developmental stuttering
Neurology
 , 
2001
, vol. 
57
 (pg. 
207
-
215
)
Foundas
AL
Bollich
AM
Feldman
J
Corey
DM
Hurley
M
Lemen
LC
Heilman
KM
Aberrant auditory processing and atypical planum temporale in developmental stuttering
Neurology
 , 
2004
, vol. 
63
 (pg. 
1640
-
1646
)
Foundas
AL
Corey
DM
Angeles
V
Bollich
AM
Crabtree-Hartman
E
Heilman
KM
Atypical cerebral laterality in adults with persistent developmental stuttering
Neurology
 , 
2003
, vol. 
61
 (pg. 
1378
-
1385
)
Fox
PT
Ingham
RJ
Ingham
JC
Hirsch
TB
Downs
JH
Martin
C
Jerabek
P
Glass
T
Lancaster
JL
A PET study of the neural systems of stuttering
Nature
 , 
1996
, vol. 
382
 (pg. 
158
-
161
)
Fox
PT
Ingham
RJ
Ingham
JC
Zamarripa
F
Xiong
JH
Lancaster
JL
Brain correlates of stuttering and syllable production: a PET performance-correlation analysis
Brain
 , 
2000
, vol. 
123
 (pg. 
1985
-
2004
)
Galaburda
AM
Corsiglia
J
Rosen
GD
Sherman
GF
Planum temporale asymmetry, reappraisal since Geschwind and Levitsky
Neuropsychologia
 , 
1987
, vol. 
25
 (pg. 
853
-
868
)
Galaburda
AM
LeMay
M
Kemper
TL
Geschwind
N
Right-left asymmetries in the brain
Science
 , 
1978
, vol. 
199
 (pg. 
852
-
856
)
Galaburda
AM
Sanides
F
Geschwind
N
Human brain: cytoarchitectonic left-right asymmetries in the temporal speech region
Arch Neurol
 , 
1978
, vol. 
35
 (pg. 
812
-
817
)
Gannon
PJ
Holloway
RL
Broadfield
DC
Braun
AR
Asymmetry of chimpanzee planum temporale: humanlike pattern of Wernicke's brain language area homologue
Science
 , 
1998
, vol. 
279
 (pg. 
220
-
222
)
Gaser
C
Schlaug
G
Brain structures differ between musicians and non-musicians
J Neurosci
 , 
2003
, vol. 
23
 (pg. 
9240
-
9245
)
Geschwind
N
Levitsky
W
Human brain: left-right asymmetries in temporal speech region
Science
 , 
1968
, vol. 
161
 (pg. 
186
-
187
)
Grefkes
C
Geyer
S
Schormann
T
Roland
P
Zilles
K
Human somatosensory area 2: observer-independent cytoarchitectonic mapping, interindividual variability, and population map
Neuroimage
 , 
2001
, vol. 
14
 (pg. 
617
-
631
)
Gundara
N
Zivanovic
S
Asymmetry in East African skulls
Am J Phys Anthropol
 , 
1968
, vol. 
28
 (pg. 
331
-
7
)
Heiervang
E
Hugdahl
K
Steinmetz
H
Inge Smievoll
A
Stevenson
J
Lund
A
Ersland
L
Lundervold
A
Planum temporale, planum parietale and dichotic listening in dyslexia
Neuropsychologia
 , 
2000
, vol. 
38
 (pg. 
1704
-
1713
)
Howie
P
Concordance for stuttering in monozygotic and dizygotic twin pairs
J Speech Lang Hear Res
 , 
1981
, vol. 
24
 (pg. 
317
-
321
)
Ide
A
Dolezal
C
Fernandez
M
Labbe
E
Mandujano
R
Montes
S
Segura
P
Verschae
G
Yarmuch
P
Aboitiz
F
Hemispheric differences in variability of fissural patterns in parasylvian [sic] and cingulate regions of human brains
J Comp Neurol
 , 
1999
, vol. 
410
 (pg. 
235
-
242
)
Ingham
RJ
Fox
PT
Ingham
JC
Xiong
JH
Zamarripa
F
Hardies
LJ
Lancaster
JL
Brain correlates of stuttering and syllable production: gender comparison and replication
J Speech Lang Hear Res
 , 
2004
, vol. 
47
 (pg. 
321
-
341
)
Jancke
L
Hanggi
J
Steinmetz
H
Morphological brain differences between adult stutterers and non-stutterers
BMC Neurol
 , 
2004
, vol. 
4
 pg. 
23
 
Jenkinson
M
Bannister
P
Brady
M
Smith
S
Improved optimization for the robust and accurate linear registration and motion correction of brain images
Neuroimage
 , 
2002
, vol. 
17
 (pg. 
825
-
841
)
Juch
H
Zimine
I
Seghier
ML
Lazeyras
F
Fasel
JHD
Anatomical variability of the lateral frontal lobe surface: implication for intersubject variability in language neuroimaging
Neuroimage
 , 
2005
, vol. 
24
 (pg. 
504
-
514
)
Kaufmann
WE
Galaburda
AM
Cerebrocortical microdysgenesis in neurologically normal subjects: a histopathologic study
Neurology
 , 
1989
, vol. 
39
 (pg. 
238
-
244
)
Kennedy
DN
O'Craven
KM
Ticho
BS
Goldstein
AM
Makris
N
Henson
JW
Structural and functional brain asymmetries in human situs inversus totalis
Neurology
 , 
1999
, vol. 
53
 (pg. 
1260
-
5
)
Knaus
TA
Bollich
AM
Corey
DM
Lemen
LC
Foundas
AL
Variability in perisylvian brain anatomy in healthy adults
Brain Lang
 , 
2006
, vol. 
97
 (pg. 
219
-
232
)
Kochunov
P
Lancaster
JL
Glahn
DC
Purdy
D
Laird
AR
Gao
F
Fox
PT
Retrospective motion correction protocol for high-resolution anatomical MRI
Hum Brain Mapp
 , 
2006
, vol. 
27
 (pg. 
957
-
62
)
Kochunov
P
Mangin
JF
Coyle
T
Lancaster
JL
Thompson
P
Riviere
D
Cointepas
Y
Regis
J
Schlosser
A
Royall
DR
, et al.  . 
Age-related morphology trends of cortical sulci
Hum Brain Mapp
 , 
2005
, vol. 
26
 (pg. 
210
-
220
)
Kulynych
JJ
Vladar
K
Jones
DW
Weinberger
DR
Gender differences in the normal lateralization of the supratemporal cortex: MRI surface-rednering morphometry of Heschl's gyrus and the planum temporale
Cereb Cortex
 , 
1994
, vol. 
4
 (pg. 
107
-
118
)
Lancaster
JL
Glass
T
Lankipalli
BR
Downs
H
Mayberg
H
Fox
PT
A modality-independent approach to spatial normalization of tomographic images of the human brain
Hum Brain Mapp
 , 
1995
, vol. 
3
 (pg. 
209
-
223
)
Larsen
JP
Odegaard
H
Grude
TH
Hoien
T
Magnetic resonance imaging—a method of studying the size and asymmetry of the planum temporale
Acta Neurol Scand
 , 
1989
, vol. 
80
 (pg. 
438
-
445
)
LeMay
M
Morphological cerebral asymmetries of modern man, fossil man, and nonhuman primate
Ann N Y Acad Sci
 , 
1976
, vol. 
280
 (pg. 
349
-
66
)
LeMay
M
Asymmetries of the skull and handedness. Phrenology revisited
J Neurol Sci
 , 
1977
, vol. 
32
 (pg. 
243
-
53
)
Lohmann
G
von Cramon
DY
Steinmetz
H
Sulcal variability of twins
Cereb Cortex
 , 
1999
, vol. 
9
 (pg. 
754
-
763
)
Luders
E
Thompson
PM
Narr
KL
Toga
AW
Jancke
L
Gaser
C
A curvature-based approach to estimate local gyrification on the cortical surface
Neuroimage
 , 
2006
, vol. 
29
 (pg. 
1224
-
1230
)
Mangin
JF
Frouin
V
Bloch
I
Regis
J
Lopez-Krahe
J
From 3D magnetic resonance images to structural representations of the cortex topography using topology preserving deformations
J Math Imaging Vis
 , 
1995
, vol. 
5
 (pg. 
297
-
318
)
Mangin
JF
Frouin
V
Regis
J
Bloch
I
Belin
P
Samson
Y
Towards better management of cortical anatomy in multi-modal multi-individual brain studies
Phys Med
 , 
1996
, vol. 
12
 (pg. 
103
-
107
)
Mangin
JF
Riviere
D
Cachia
A
Duchesnay
E
Cointepas
Y
Papadopolous-Orfanos
D
Scifo
P
Ochiai
T
Brunelle
F
Regis
J
A framework to study the cortical folding patterns
Neuroimage
 , 
2004
, vol. 
23
 
Suppl 1
(pg. 
S129
-
S138
)
Mazziotta
JC
Toga
AW
Evans
A
Fox
PT
Lancaster
JL
A probabilistic atlas of the human brain: theory and rationale for its development. The International Consortium for Brain Mapping (ICBM)
Neuroimage
 , 
1995
, vol. 
2
 (pg. 
89
-
101
)
Mazziotta
JC
Toga
AW
Evans
A
Fox
PT
Lancaster
JL
Zilles
K
Woods
R
Paus
T
Simpson
G
Pike
B
, et al.  . 
A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM)
Philos Trans R Soc Lond B Biol Sci
 , 
2001
, vol. 
356
 (pg. 
1293
-
1322
)
Molko
N
Cachia
A
Riviere
D
Mangin
JF
Bruandet
M
LeBihan
D
Cohen
L
Dehaene
S
Functional and structural alterations of the intraparietal sulcus in a developmental dyscalculia of genetic origin
Neuron
 , 
2003
, vol. 
40
 (pg. 
847
-
858
)
Molko
N
Cachia
A
Riviere
D
Mangin
JF
Bruandet
M
LeBihan
D
Cohen
L
Dehaene
S
Brain anatomy in Turner syndrome: evidence for impaired social and spatial-numerical networks
Cereb Cortex
 , 
2004
, vol. 
14
 (pg. 
840
-
850
)
Naidich
TP
Valavanis
AG
Kubik
S
Anatomic relationships along the low-middle convexity: Part I—normal specimens and magnetic resonance imaging
Neurosurgery
 , 
1995
, vol. 
36
 (pg. 
517
-
532
)
Narr
K
Thompson
P
Sharma
T
Moussai
J
Zoumalan
C
Rayman
J
Toga
A
Three-dimensional mapping of gyral shape and cortical surface asymmetries in schizophrenia: gender effects
Am J Psychiatry
 , 
2001
, vol. 
158
 (pg. 
244
-
255
)
Oldfield
RC
The assessment and analysis of handedness: the Edinburgh inventory
Neuropsychologia
 , 
1971
, vol. 
9
 (pg. 
97
-
113
)
Ono
M
Kubik
S
Abernathy
CD
Atlas of the cerebral sulci
1990
New York
Thieme Medical Publishers
(pg. 
2
-
211
)
Piao
X
Hill
RS
Bodell
A
Chang
BS
Basel-Vanagaite
L
Straussberg
R
Dobyns
WB
Qasrawi
B
Winter
RM
Innes
AM
, et al.  . 
G protein-coupled receptor-dependent development of human frontal cortex
Science
 , 
2004
, vol. 
303
 (pg. 
2033
-
2036
)
Preis
S
Jancke
L
Schmitz-Hillebrecht
J
Steinmetz
H
Child age and planum temporale asymmetry
Brain Cogn
 , 
1999
, vol. 
40
 (pg. 
441
-
452
)
Riaz
N
Steinberg
S
Ahmad
J
Pluzhnikov
A
Riazuddin
S
Cox
NJ
Drayna
D
Genomewide significant linkage to stuttering on chromosome 12
Am J Hum Genet
 , 
2005
, vol. 
76
 (pg. 
647
-
651
)
Richman
DP
Stewart
MR
Hutchinson
JW
Caviness
VS
Mechanical model of brain convolutional development
Science
 , 
1975
, vol. 
189
 (pg. 
18
-
21
)
Riviere
D
Mangin
JF
Papadopoulos-Orfanos
D
Martinez
JM
Frouin
V
Regis
J
Automatic recognition of cortical sulci of the human brain using a congregation of neural networks
Med Image Anal
 , 
2002
, vol. 
6
 (pg. 
77
-
92
)
Rossi
A
Serio
A
Stratta
P
Petruzzi
C
Schiazza
G
Mattei
P
Mancini
F
Casacchia
M
Three-dimensional in vivo planum temporale reconstruction
Brain Lang
 , 
1994
, vol. 
47
 (pg. 
89
-
95
)
Rubens
AB
Mahowald
MW
Hutton
JT
Asymmetry of the lateral (Sylvian) fissures in man
Neurology
 , 
1976
, vol. 
26
 (pg. 
620
-
626
)
Schaechter
JD
Moore
CI
Connell
BD
Rosen
BR
Dijkhuizen
RM
Structural and functional plasticity in the somatosensory cortex of chronic stroke patients
Brain
 , 
2006
, vol. 
129
 (pg. 
2722
-
2733
)
Schmitt
JE
Watts
K
Eliez
S
Bellugi
U
Galaburda
AM
Reiss
AL
Increased gyrification in Williams syndrome: evidence using 3D MRI methods
Dev Med Child Neurol
 , 
2002
, vol. 
44
 (pg. 
292
-
295
)
Shapleske
J
Rossell
SL
Woodruff
PWR
David
AS
The planum temporale: a systematic, quantitative review of its structural, functional, and clinical significance
Brain Res Brain Res Rev
 , 
1999
, vol. 
29
 (pg. 
26
-
49
)
Smith
SM
Fast robust automated brain extraction
Hum Brain Mapp
 , 
2002
, vol. 
17
 (pg. 
143
-
155
)
Sommer
M
Koch
MA
Paulus
W
Weiller
C
Buchel
C
Disconnection of speech-relevant brain areas in persistent developmental stuttering
Lancet
 , 
2002
, vol. 
360
 (pg. 
380
-
383
)
Sowell
ER
Thompson
PM
Rex
D
Kornsand
D
Tessner
KD
Jernigan
TL
Toga
AW
Mapping sulcal pattern asymmetry and local cortical surface gray matter distribution in vivo: maturation in perisylvian cortices
Cereb Cortex
 , 
2002
, vol. 
12
 (pg. 
17
-
26
)
Stager
SV
Ludlow
CL
The effects of fluency-evoking conditions on voicing onset types in persons who do and do not stutter
J Commun Disord
 , 
1998
, vol. 
31
 (pg. 
33
-
51
)
Steinmetz
H
Structure, function and cerebral asymmetry: in vivo morphometry of the planum temporale
Neurosci Biobehav Rev
 , 
1996
, vol. 
20
 (pg. 
587
-
591
)
Steinmetz
H
Herzog
A
Schlaug
G
Huang
Y
Jancke
L
Brain asymmetry in monozygotic twins
Cereb Cortex
 , 
1995
, vol. 
5
 (pg. 
296
-
300
)
Strub
RL
Black
FW
Naeser
MA
Anomalous dominance in sibling stutterers: evidence from CT scan asymmetries, dichotic listening, neuropsychological testing, and handedness
Brain Lang
 , 
1987
, vol. 
30
 (pg. 
338
-
350
)
Tamraz
JC
Comair
YG
Atlas of regional anatomy of the brain using MRI: with functional correlations. New York: Springer-Verlag
p
 , 
2006
, vol. 
5
 (pg. 
1
-
158
)
Thompson
PM
Moussai
J
Zohoori
S
Goldkorn
A
Khan
AA
Mega
MS
Small
GW
Cummings
JL
Toga
AW
Cortical variability and asymmetry in normal aging and Alzheimer's disease
Cereb Cortex
 , 
1998
, vol. 
8
 (pg. 
492
-
509
)
Travis
LE
The cerebral dominance theory of stuttering: 1931–1978
J Speech Hear Disord
 , 
1978
, vol. 
43
 (pg. 
278
-
281
)
Watkins
KE
Paus
T
Lerch
JP
Zijdenbos
A
Collins
DL
Neelin
P
Taylor
J
Worsley
KJ
Evans
AC
Structural asymmetries in the human brain: a voxel-based statistical analysis of 142 MRI scans
Cereb Cortex
 , 
2001
, vol. 
11
 (pg. 
868
-
877
)
Weinberger
DR
Luchins
DJ
Morihisa
J
Wyatt
RJ
Asymmetrical volumes of the right and left frontal and occipital regions of the human brain
Ann Neurol
 , 
1982
, vol. 
11
 (pg. 
97
-
100
)
Witelson
SF
Kigar
DL
Sylvian fissure morphology and asymmetry in men and women: bilateral differences in relation to handedness in men
J Comp Neurol
 , 
1992
, vol. 
323
 (pg. 
326
-
340
)
Witelson
SF
Pallie
W
Left hemisphere specialization for language in the newborn. Neuroanatomical evidence of asymmetry
Brain
 , 
1973
, vol. 
96
 (pg. 
641
-
646
)
Yairi
E
Ambrose
N
Onset of stuttering in preschool children: Selected factors
J Speech Hear Res
 , 
1992
, vol. 
35
 (pg. 
782
-
788
)
Yeni-Komshain
GH
Benson
DA
Anatomical study of asymmetry in the temporal lobe of humans, chimpanzees, and rhesus monkeys
Science
 , 
1976
, vol. 
192
 (pg. 
387
-
389
)
Zar
JH
Biostatistical analysis
1996
3rd ed
Upper Saddle River (NJ)
Prentice-Hall
pg. 
94