Abstract

The sulci and gyri found within the anterior cingulate (AC), and across the cerebrum generally, have been found to vary in location and complexity from one individual to the next, making it difficult to analyze imaging data accurately and systematically. In this study, we examined the nature of morphometric variance in the AC of the left and right cerebral hemispheres using high-resolution structural magnetic resonance imaging (MRI) acquired from 176 healthy volunteers. Depending on the presence of a paracingulate sulcus (PCS) and its antero-posterior extent, three types of AC patterns were identified: ‘prominent’, ‘present’ and ‘absent’. Hemispheric comparisons across the whole sample showed the PCS to be more commonly ‘prominent’ in the left hemisphere and more commonly ‘absent’ in the right hemisphere. There was a significant gender difference, such that males showed an asymmetric pattern characterized by increased fissurization of the left AC, while females showed greater symmetry, with less fissurization of the left AC. Overall cerebral morphology, namely hemispheric volume and hemispheric fissurization, were also measured and used as independent variables as well as covariates in the analyses in order to ascertain the specificity of the results regarding AC morphology. Results showed that cerebral volume for males was larger on the right than on the left while fissurization showed the reverse asymmetry of greater leftward fissurization. In contrast, females were symmetric in both respects. The findings regarding AC morphology could not be explained by differences in these overall cerebral measures or by differences in age and handedness within the population. The results suggest that in the normal male brain, there exist morphological asymmetries at both the global and local levels that are less apparent in the female brain. The findings have implications for future studies examining the organization, development and functional anatomy of the AC.

Introduction

The anterior cingulate (AC) cortex is located bilaterally in the medial wall of the frontal lobes, forming a large region around the rostrum of the corpus callosum, bordered superiorly by the cingulate sulcus (CS) and inferiorly by the callosal sulcus. The advent of both structural and functional imaging techniques has recently allowed investigation of the functional and neurological roles of the AC. Recent research has shown the AC to be a pivotal component of brain networks that underlie executive functions such as motivation, emotion, attention, memory, learning, language, interpersonal and motor behavior [reviewed by Vogt and Devinsky (Vogt et al., 1992; Devinsky et al., 1995)]. Nevertheless, the specific function of the AC and its subregions remains unclear. This may in part be due to interindividual variation in the cortical sulci and gyri, particularly in the AC, which makes accurate and systematic analysis of imaging data difficult (Ono et al., 1990; Devinsky et al., 1995; Vogt et al., 1995; Paus et al., 1996a,b; Thompson et al., 1996). Specifically, the sulci and gyri within this region may be ‘doubled-up’ in some individuals in which case the more inferior elements are referred to as the cingulate sulcus/gyrus (also referred to as the inferior cingulate sulcus/gyrus), while the superior elements are referred to as the paracingulate sulcus (PCS)/gyrus (also referred to as the superior cingulate sulcus/gyrus). It is unclear whether the paracingulate region should be included as a part of the strict definition of the cingulate as it is cytoarchitectonically a cingulo-frontal transition area (Devinsky et al., 1995). Such variations in AC morphology have reduced the reliability of the volumetric delineation of AC boundaries that are required to ascertain AC structure–function relationships.

To date there have been five post-mortem studies that have examined the gross morphology of the AC (Eberstaller, 1884; Weinberg, 1905; Ono et al., 1990; Vogt et al., 1995; Ide et al., 1999), but these have been methodologically limited. Initial post-mortem studies were based on qualitative judgements of AC morphological patterns and found hemispheric differences in the frequency of PCS observed (Eberstaller, 1884; Weinberg, 1905). More recent post-mortem studies have either been unable to examine hemispheric effects on AC morphology as a consequence of only one hemisphere being examined from each subject (Vogt et al., 1995), or have used small sample sizes and found conflicting results. For example, Ono et al. (Ono et al., 1990) found no hemisphere difference in the manifestation of a PCS, while Ide et al. (Ide et al., 1999) found a hemispheric asymmetry such that the PCS was more often ‘prominent’ in the left hemisphere.

In a large MRI study of 247 healthy young volunteers, Paus et al. (Paus et al., 1996a) found large variability in the morphological features of the AC region and identified a hemispheric asymmetry related to the presence and extent of the PCS, which contrasted with the variable findings of previous post-mortem studies (Weinberg, 1905; Ono et al., 1990; Ide et al., 1999). They also found a previously unreported gender difference in the organization of the AC such that females were significantly more likely to be in the two extreme categories of the PCS, namely the ‘prominent’ and ‘absent’ PCS.

The lack of research into morphological variance may be due to an assumption that the gyrification/fissurization of the cortex, which occurs principally during the second and third trimester of gestation (Chi et al., 1977, Huang, 1991; Naidich et al., 1994), is a mechanical process that does not have any cytoarchitectural, connectional or functional relevance (Turner, 1948; Richman et al., 1975; Armstrong et al., 1991). This view has been super seded due to recent evidence from twin studies that suggests that sulcal/gyral formation is influenced by both genetic and epigenetic factors (Oppenheim et al., 1989; Steinmetz et al., 1994, 1995; Tramo et al., 1995; Bartley et al., 1997; Biondi et al., 1998; Lohmann et al., 1999) and, importantly, affected by underlying cytoarchitecture (Sanides, 1964; Watson et al., 1993; Rademacher et al., 1993; Roland and Zilles, 1994) and neural connectivity (Caviness et al., 1975, 1989; Goldman-Rakic, 1981; Rakic, 1988). In this regard, it has been found that variations in sulcal/gyral pattern are associated with differences in the size and distribution of cytoarchitectonically defined regions within the AC region (Vogt et al., 1995). Further, a systematic review of the evidence by Welker indicated that generally, gyrification has functional significance and that gyral crowns have qualitatively different organization and connectivity from sulcal walls and fundi (Welker, 1990). Van Essen has argued that the greater connectivity within gyral crowns may predate their development, and play an important role in their formation (Van Essen, 1997). There is also evidence from functional imaging studies that the morphological pattern within the AC is directly related to the location of functional activation peaks (Crosson et al., 1999).

These findings suggest that description of AC morphology, particularly sulcal/gyral patterns, and the nature of its variability in a large cohort of healthy volunteers is important to understanding the relationship between AC structure and function. Further, morphological studies of this region may provide an alternative to volumetric structural neuroimaging methods, which are limited because of the morphological variability. Knowledge of the morphological variability of this region is also critically important in evaluating functional neuroimaging studies. The aim of the current study was to develop a reliable method by which to examine the gross structural morphology of the AC in a large cohort of healthy volunteers. The present study extends previous work (Paus et al., 1996a) by taking into account the morphological relationships (asymmetry) between the left and right AC regions for each individual. We also examined the influence of hemisphere and gender on this relationship, while controlling for the effects of age, overall hemispheric volume and hemispheric fissurization on AC morphology. The use of hemispheric fissurization as a control measure is necessary to determine the specificity of any systematic differences in AC morphology.

Materials and Methods

Magnetic resonance imaging (MRI) scans were acquired from 176 healthy volunteers. The sample consisted of 103 males and 73 females (mean age 30.7 years, SD 12.5 years, range 15–69 years). In 135 of these subjects, handedness was assessed using the Edinburgh Inventory (Oldfield, 1971) and showed that there were 123 right-handed, 10 left-handed and 2 ambidextrous subjects in the sample. All subjects had a no history of neurological, psychiatric and alcohol or substance abuse disorders. Subjects were scanned using an identical GE Signa 1.5 T scanner at either the Royal Melbourne Hospital or Cabrini Hospital, Melbourne. Foam padding minimized head movement and velcro straps were placed across the subject's forehead and chin. A three-dimensional volumetric spoiled gradient recalled echo in the steady state sequence generated 124 contiguous, 1.5 mm coronal slices (subsequently resliced into 1 mm cubic voxels). Imaging parameters in the Cabrini/Royal Melbourne Hospitals were: TE 9/3.3 ms; TR 36/14.3 ms; flip angle 35/30; matrix size 256 × 192/256 × 256; field of view 20 × 15 cm/24 × 24 cm matrix; voxel dimensions 0.781 × 0.781× 1.5 mm/0.937 × 0.937 × 1.5 mm. Each scanner was calibrated fortnightly using the same proprietary phantom to ensure stability and accuracy of measurements. MRI data was transferred from DAT tape to an SGI-02 workstation. Image analysis was performed using the multi-modal image-processing package MEDx 3.0 (Sensor Systems). Code numbers were used to maintain patient confidentiality and to ensure blinded ratings of the MRI scan. Written informed consent was obtained from all subjects, and local research and ethics committees approved the study.

Classification of Anterior Cingulate Morphology

Classification Criteria

A protocol similar to that utilized by Paus et al. (Paus et al., 1996a) was generated, to classify the number of cingulate sulci (CS) and the explicitness of their folding. In addition, we described the explicitness of folding across two homologous areas of the same brain (asymmetry; described in more detail below). The type of AC surface morphology was classified according to the presence or absence of the PCS, as well as its antero-posterior extent. The PCS was defined as the sulcus located dorsal to the CS with a course clearly parallel to the CS. This yielded three categories of AC surface morphology — prominent-PCS, present-PCS and absent-PCS — depending upon the presence or absence of the PCS and its antero-posterior extent (see Fig. 1).

The PCS was considered ‘prominent’ (Fig. 1a) if there was as least one clearly developed horizontal element that made up a sulcus which was parallel to the CS and either (i) extended at least 40 mm or (ii) had no more than 20 mm of interruptions in total between its anterior origin and an imaginary vertical line passing through the anterior commissure (VAC). If the interruptions from the origin to the VAC were >20 mm, but the horizontal element parallel to the CS was at least 20 mm in length, the PCS was considered to be ‘present’ (Fig. 1b). Finally, if there were no clearly developed horizontal elements that make up a sulcus, parallel to the CS and extending at least 20 mm, the PCS was considered to be ‘absent’ (Fig. 1c).

Classification-related Issues

A number of steps were taken to improve the reliability of the classification method.

We reduced the ambiguity in the anterior regions (Paus et al., 1996a) due to a confluence of the CS and PCS with the superior rostral sulcus (SRS; often anteriorly continuous with the CS or PCS). This was achieved by defining the origin of the CS and PCS as the point at which the sulcus extends posteriorly from an imaginary vertical line running perpendicular to the line passing through the anterior and posterior commissures (AC–PC) and parallel to the VAC. This point is indicated by solid arrow for both the PCS (Fig. 2a) and CS (Fig. 2b).

Raters were instructed to obtain a mid-sagittal section, align the AC–PC line horizontally and move three or four slices laterally from the midline in order to classify AC morphology. Assessing AC morphology over several parasagittal slices is important for two reasons. Firstly, superficial dimples may appear as sulci in the mid-sagittal slice, but disappear within one or two slices from the midline. Secondly, the mid-sagittal image can sometimes be sampled from both the right and left sides of the brain and from the cerebrospinal fluid-filled space between them (Coppola et al., 1995). Thus, partial volume averaging can make a blend of the right and left sulci and contribute differently to the image due to differences in the accuracy of its position or due to the inherent asymmetry of the contributing structure. In such an image it can be difficult to distinguish the right cingulate from the left cingulate. Therefore, to reduce the confounding effects of these two issues, the current set of criteria required the PCS to be clearly evident in at least three slices from the midline. Finally, five MRI data sets that were contaminated with excessive movement artifact were not included in the sample (four male, one female, all right-handed).

Classification Reliability

Intra-rater reliability for the classification of AC morphology was assessed by one rater (M.Y.) on 24 randomly chosen cases. In order to assess inter-rater reliability a second rater (G.S.) was provided with a manual describing the classification criteria before assessing the same 24 cases. Both raters were blind to subject details at all times. From all classifications derived (48 in all), only 2% of classifications were rated differently by the initial rater (M.Y.) and 10% between raters (M.Y. and G.S.). Intra- and inter-rater reliability (weighted kappa) for the classification of AC morphology were = 0.96 and = 0.90 respectively.

Anterior Cingulate Asymmetry Index

An asymmetry index was assigned to each individual based on the combination of left and right AC morphology. For example, the PCS in the ‘prominent’ classification is more explicit (i.e. greater antero-posterior extent of the fissure) than the PCS in the ‘present’ pattern, which in turn is more explicit than that in the ‘absent’ pattern. Thus an asymmetry direction of leftward (left > right), symmetric (left = right), or rightward (left < right), could be assigned to each individual. Further, the magnitude of asymmetry could be by one or two categories in either direction: as absent versus present or present versus prominent (one category difference) or as absent versus prominent (two category difference). In addition there are three ways in which the morphology could be identical (absent versus absent, present versus present, prominent versus prominent). Therefore, the asymmetry index is made up of five values [left > right (2), left > right (1), left = right, left < right (1), left < right (2)] from the nine combinations of left versus right AC morphology. This index is reflective of the relationship between the hemispheres for each individual in terms of the extent of PCS fissurization, indicating whether an individual has a symmetrically or asymmetrically fissured AC, in which direction, and by how much.

The asymmetry index is an important value because it provides an alternative to the ‘unpaired’ comparison that examines patterns of AC morphology across the population. That is, the combination of values from the left and right AC are ‘paired’ for each individual and reduced to a single asymmetry index. As such, the individual's one cerebral hemisphere acts as a control for the other in deriving an asymmetry index value. Also, the use of the asymmetry index as a single dependent variable that is derived from the values from each hemisphere avoids the difficulty of analysing repeated-measures variables (i.e. left versus right) within this categorical analysis framework. Most widely available statistical packages for the analysis of categorical variables assume that the predictor variables are independent. They do not yet allow for the analysis of repeated measures by methods equivalent to those used in the general analysis of variance (ANOVA) framework for continuous dependent variables.

Hemispheric Volume and Hemispheric Fissurization Index

To account for differences in absolute cerebral hemispheric volume, and cerebral hemispheric fissurization, as well as the asymmetry between the two cerebral hemispheres with regards to overall hemispheric fissurization and volume, a semi-automated method was developed from which we were able to estimate these values for each individual. These measures, as well as age, were used both as independent variables as well as covariates in the analyses in order to ascertain the specificity of the results regarding AC fissurization.

Each MRI data set was registered linearly (six-parameter rigid body transform) using the Automated Image Registration package (AIR, version 3.0.8) (Woods et al., 1998) to a high-resolution template MRI (Collins et al., 1998) placed into standard Talairach space (Talairach and Tournoux, 1988). The signal intensity variance artefact often seen using MRI (variously referred to as RF inhomogeneity, shading artifact or intensity non-uniformity) was then corrected using the MNI_N3 non-parametric non-uniformity normalization package (version 1.04) (Sled et al., 1998). Finally, the image dataset was resampled to subvoxel resolution (0.5 mm3, resulting in 368 coronal slices per brain) in order to increase sampling density and minimize sampling error when estimating hemispheric volume and hemispheric fissurization.

Hemispheric Volume and Hemispheric Volume Asymmetry

Having aligned the MRI and corrected the signal intensity artefact, the cerebral hemisphere of interest (excluding the cerebellar lobe and brainstem) was masked out from each brain image. The cerebral hemisphere of interest was then thresholded at a customized level based on the histogram of the image intensities to exclude only ventricular and intra-sulcal cerebrospinal fluid (CSF). The absolute volume of the hemisphere was then calculated.

Hemispheric Fissurization Index and Hemispheric Fissurization Index Asymmetry

To generate a fissurization index the original MR images were segmented at a harsher intensity threshold to remove a small amount of gray matter (in order to open up the sulci) (Magnotta et al., 1999; Nopoulos et al., 2000). Once again, this procedure was customized for each individual brain image as above. A binary transformation was then applied to characterize brain tissue/non-tissue voxels. This enabled the more harshly thresholded volume to be estimated using simple voxel counts. The Roberts edge detection algorithm as implemented in MEDx was then applied to the image to detect the surface voxels of this volume within each coronal slice of the image (see Fig. 3). The ratio of surface voxels to those of the total volume is an indirect measure of the degree of cortical folding, and is the basis of the automated estimate of hemispheric fissurization.

Masks were then applied to the image to isolate each cerebral hemisphere and exclude the cerebellum and brainstem voxels. Surface- to-volume ratios (SVRs) were then calculated separately for each hemisphere, expressed as the total number of surface voxels divided by the total number of volume voxels across the entire 368 slice image. These ratios will be higher when there is greater complexity of cortical folding. However, they will also be affected by the width and depth of cortical sulci. That is, individuals with the same pattern of cortical folding but more prominent sulci may have different SVRs. This will be evident by replacement of brain tissue by CSF. For this reason, the ratio of CSF to grey and white matter was used as a covariate to the SVR in deriving the fissurization index. This control is particularly relevant when there may be increases in SVR produced by brain atrophy, such as in normal aging (Magnotta et al., 1999).

Results

Cerebral Hemispheric Volume and Hemispheric Fissurization Index

We first examined the overall volume and fissurization of the cerebral hemispheres in order to assess the AC morphology in the context of cerebral morphology, as shown in Table 1. As these variables were all continuous, ANOVA was used. The right cerebral volume was significantly larger than the left cerebral volume overall [F(1,169) = 31.29, P < 0.0001], and there was also a significant volume by gender interaction [F(1,169) = 11.46, P < 0.001]. This interaction reflects a larger right/left hemispheric volume difference for males. This difference can also be seen in the asymmetry values. A separate gender comparison for each hemisphere revealed that males had both larger left [F(1,169) = 81.56, P < 0.0001] and right [F(1,169) = 88.6, P < 0.0001] cerebral volumes. Correlational analyses revealed that left and right cerebral volumes were highly associated for both males (r = 0.984, P < 0.0001) and females (r = 0.994, P < 0.0001).

Despite the larger right hemispheric volume, the fissurization index was significantly lower in the right than left hemisphere for all subjects [F(1,169) = 157.32, P < 0.0001]. Again, there was a left versus right fissurization index by gender interaction [F(1,169) = 8.14, P < 0.005], reflecting a larger difference in fissurization between hemispheres in males than females (see also the asymmetry values). A separate comparison for each hemisphere revealed no significant gender difference in either the left or right cerebral fissurization indices. Correlational analyses revealed that left and right cerebral fissurization indices were highly associated for both males (r = 0.823, P < 0.0001) and females (r = 0.919, P < 0.0001).

Anterior Cingulate Morphology

While the CS was present in all hemispheres studied, its morphology was extremely variable within and between hemispheres. Across the entire sample and across both hemispheres, 68% of cingulate cortices showed evidence of a PCS, in either its ‘present’ or ‘prominent’ form. Table 2 (see totals) also shows that, overall, the PCS was more often ‘prominent’ in the left hemisphere (left 49% versus right 28%) and ‘absent’ in the right hemisphere (left 26% versus right 38%).

Asymmetry Index Differences

Table 2 also shows the frequencies for all combinations of left versus right cingulate morphology in the overall sample, and separately for males and females. McNemar's test for symmetry (Siegel and Castellan, 1988) was used to test whether the number of cases of asymmetry in one direction was counterbalanced by an equal number of cases with an asymmetry in the other direction. A significant value indicates an overall lack of symmetry in the sample — a greater explicitness of fissurization in one direction. Overall, the proportion of cases showing a leftward fissurization bias was 42%, compared with only 21% in the opposite direction. This asymmetry is reflected in the significant McNemar statistic [χ2(3) = 16.43, P < 0.001].

An analysis by gender revealed that the proportion of males showing a leftward fissurization bias was 48%, compared with only 19% in the opposite direction. In contrast, females showed a much smaller degree of fissurization asymmetry, with 32% showing a leftward asymmetry and 23% showing a rightward asymmetry. This gender difference is reflected by the McNemar test showing a significant asymmetry for males [χ2(3) = 16.31, P < 0.001] but not for females, suggesting that the significant result obtained in the overall analysis was specific to males. Interaction effects could not be examined within this the analytic framework.

Gender by Hemispheric Differences

The effect of gender on AC morphology was further examined using polychotomous logistic regression (BMDP-PR) (Dixon et al., 1990) in which the AC morphology on the left or right (with three ordered categories) or asymmetry index (with five ordered categories; see Materials and Methods) were the dependent variables. In all analyses, gender and age were used as independent variables. For analyses of left or right AC morphology, the volume and fissurization index of the relevant hemisphere were also used in the analysis. When the AC asymmetry index was the dependent variable, hemispheric volume asymmetry and hemispheric fissurization index asymmetry were included in the analysis. Hemispheric volume asymmetry was derived from the combination of left hemisphere and right hemispheric volumes (left volume – right volume), while the hemispheric fissurization index asymmetry was derived from the combination of left hemisphere and right hemispheric fissurization indices (left fissurization index – right fissurization index). In both cases, positive values were reflective of a leftward asymmetry, while negative values reflected a rightward asymmetry. Note that standard significance tests, such as χ2 and logistic regression, were not used because the dependent variables were inherently ordered (i.e. from absent to prominent), and had more than two categories (three or five).

Table 3 shows the results of the significance tests carried out using polychotomous logistic regression analysis. This table as well as Figure 4 shows that, relative to females (represented as the 0% line), males were significantly more likely to have a ‘prominent-PCS’ and less likely to have an ‘absent-PCS’ in the left hemisphere (by ~22%).

This result was maintained after partialling out the effects of age, overall left hemispheric volume and left hemispheric fissurization index (which itself was significantly related to left AC morphology). Further, there were no significant main effects of age on AC morphology for either the left or right AC, or for AC asymmetry.

Gender by Asymmetry Index Differences

The data in Table 4 represent the morphological relationship of the AC between the hemispheres, indicating whether left and right hemispheres for each individual were symmetrically or asymmetrically fissured, in which direction, and by how much. Examination of gender revealed a significant effect on the explicitness of fissurization between the left and right AC with males tending to have a more leftward pattern of fissurization (males 48% versus females 32%) and females tending to have a symmetric pattern (males 32% versus females 44%). These results were unchanged after controlling for the effects of overall hemispheric fissurization index asymmetry, overall hemispheric volume asymmetry and age.

Effects of Handedness

While the effects of handedness could not be comprehensively assessed due to the small number of left-handed individuals, a restricted analysis with data from only right-handed individuals (n = 118) was carried out. All of the previously reported significant effects remained.

Discussion

The current study extends previous methodologies (Paus et al., 1996a) in investigating the morphology of the AC in a large normative sample. For the whole study sample, the PCS was more commonly ‘prominent’ in the left hemisphere and more often ‘absent’ in the right hemisphere. There was a significant gender difference, such that males showed an asymmetric pattern characterized by increased fissurization of the left AC, while females showed greater symmetry, with less fissurization of the left AC. For overall cerebral morphology, males had larger right than left cerebral volumes, while fissurization showed the reverse asymmetry of greater leftward fissurization. In contrast, females were symmetric in both respects. These findings could not be explained by differences in global measures such as overall hemispheric volume, hemispheric fissurization index, hemispheric volume asymmetry, hemispheric fissurization index asymmetry, or by differences in age and handedness within the population.

To our knowledge, no previous studies have examined AC fissurization asymmetry by comparing the left and right AC in the same individual. Therefore, our finding of a gender effect on AC asymmetry index cannot be compared with other studies. There have been few previous studies that have examined the morphology of the AC in a single hemisphere. Methodological inconsistencies in these studies limit the ability to make direct comparisons, except for the more recent imaging study of Paus et al. (Paus et al., 1996a). Initial studies of AC morphology were based on qualitative judgements of morphological patterns and their frequencies (Eberstaller, 1884; Weinberg, 1905), while recent studies have been limited to a small series of post-mortem brains (Ono et al., 1990; Vogt et al., 1995; Ide et al., 1999). All post-mortem studies that have examined AC morphology have been limited methodologically (Ono et al., 1990; Vogt et al., 1995). In the study by Vogt et al. (Vogt et al., 1995) of brain specimens from older people, laterality could not be investigated as only one hemisphere was examined from each subject (13 left and 10 right hemispheres from different males and females). In the study by Ono et al., asymmetry was not observed in their small sample and no details about age or gender of the subjects were presented (Ono et al., 1990). In addition, not all studies describe the methods used to characterize the AC in sufficient detail. Despite these limitations, our findings of a lateralized pattern of surface morphology in the AC of normal subjects are consistent with the findings of Weinberg (Weinberg, 1905) and more recently those of Ide et al. (Ide et al., 1999). In these post-mortem studies the PCS was more often ‘prominent’ in the left hemisphere. Our findings are also consistent with the only other MRI-based study of AC gross morphology in a large sample of normals (Paus et al., 1996a). Our sample, however, showed a greater proportion of individuals in which the PCS was >20 mm in length, although the nature of the asymmetry was similar across the two studies.

The current study also found gender differences in the gross morphology of the AC. Compared with females, males were significantly more likely to have the most fissured type of AC (prominent-PCS) in the left hemisphere. These results are inconsistent with the study by Ide et al. (Ide et al., 1999), which failed to find any significant gender differences in fissurization patterns. Further, although, Paus et al. (Paus et al., 1996a) found a significant gender by hemisphere effect, the nature of this effect was different, such that females were more likely to be in the two extreme categories (i.e. absent- or prominent-PCS), while males tended to cluster in the middle (i.e. present-PCS). While the hemispheric effect is qualitatively similar to that found by Paus and colleagues, it is difficult to reconcile the differences in frequencies of PCS observed across the three categories, and the nature of the gender effects found. It is possible that methodological issues such as ambiguity in the definition of the anterior origin of the PCS (see Materials and Methods) may explain the differing findings in the two studies. In addition, MR images in the study of Paus et al. (Paus et al., 1996a) were acquired as 2 mm thick slices, resampled by interpolation into Talairach space and resliced into 0.75 mm sagittal images. In contrast, our images were acquired at 1.5 mm, there was no subsequent warping, and the brain volumes were resliced into 1 mm cubic voxels. These differences are likely to have resulted in lower image resolution in the study of Paus et al. (Paus et al., 1996a). While lower image resolution may explain some of the differences in subtyping AC morphological patterns, however, they are unlikely to explain the overall observed discrepancy. Indeed, we found a higher proportion of subjects with a PCS extending >20 mm than did Paus et al. (Paus et al., 1996a), despite our definition of the PCS being more conservative. Thus it is more likely that the hemispheric and gender differences found between the studies are representative of either a sampling bias or true population differences. It should be noted that in a subsequent volumetric study, Paus et al. (Paus et al., 1996b) found significant hemispheric and gender differences in the volume of gray matter buried within the PCS such that there was more intra-sulcal gray matter in the left PCS overall and more so in males. These volumetric effects correspond to our morphometric findings of greater fissurization in the left PCS and particularly in males.

What Might Cause these Systematic Anatomical Variations?

In the present study we found that, relative to females, a higher proportions of males have a prominent left PCS. To the extent that differences in morphology reflect differences in underlying structural organization, our findings suggest that there may exist hemispheric and gender differences in the underlying cytoarchitectonic size and distribution, as well as in the pattern of connectivity in the left and right hemispheres (Welker, 1990; Vogt et al., 1995; Van Essen, 1997). Specifically, Vogt et al. (Vogt et al., 1995) found that the paracingulate gyrus, when present, always contained a large part of Brodmann's area 32 (BA32). When the paracingulate gyrus was absent, BA32 always began in the depths of the cingulate sulcus, occupying the dorsal wall of that sulcus (Vogt et al., 1995). Thus, the characteristics of the underlying cytoarchitecture and connectivity may be different between cingulate cortices with and without a PCS. For example, having a PCS (particularly in its ‘prominent’ form) may be the consequence of stronger internal connectivity within this area, given that it occupies a gyral crown (Welker, 1990). In contrast, not having a PCS may be the consequence of a stronger external connectivity of this area with adjacent areas, given that it occupies the dorsal wall of the sulcus. If these systematic morphological differences have cytoarchitectonic and connectional significance, they are likely to have a neurodevelopmental basis and may also be of evolutionary significance (Zilles et al., 1988, Welker, 1990). It has been suggested that an increase in the cortical folding of a particular brain area is indicative of a progressive evolution of this region in humans (Zilles et al., 1988). Accordingly, the higher incidence of the ‘prominent’ PCS in the left hemisphere of males may reflect the relative expansion in terms of connectivity and/or volume of the left paralimbic (BA32) cortex in males. However, the cognitive and behavioral significance of altered morphology in this region is unknown.

The present study suggests that in the normal male brain there exist morphological asymmetries at both the cerebral and local (AC) levels that are less apparent in the female brain. These asymmetries and gender-related differences are principally determined in utero and may have important implications for understanding the organization, development and functional anatomy the paracingulate region of the AC cortex. It is perhaps relevant that there have been reports of hemispheric and gender differences in the rate of brain maturation indicating that the male brain matures later than that of the female, and the left hemisphere matures later than the right (Geschwind and Galaburda, 1985). Whether or not the general rate of brain maturation is associated with volumetric and fissurization asymmetry is a question that warrants further investigation. However, such gender differences may also be consequent on sex chromosomal differences (Crow, 1990) or the hormonal environment of the fetus, since males are exposed to testosterone in utero (Geschwind and Galaburda, 1985).

Our data also suggest that the fissural pattern of the left and right AC is relatively independent of each other. That is, having a ‘prominent’ pattern of PCS in the left hemisphere has no, or minimal, influence on the pattern in the right hemisphere (i.e. an individual with a ‘prominent-PCS’ in the left is equally likely to have an ‘absent’, ‘present’ or ‘prominent’ pattern on the right). In contrast, left and right fissurization as well as left and right volumes of the cerebral hemispheres were highly interrelated. Thus, the developmental trajectory of the left and right AC morphology, but not the cerebral hemispheres, shows a degree of independence during neurodevelopment. While a genetic contribution to morphological differences is indicated by twin studies (Oppenheim et al., 1989; Biondi et al., 1998; Lohmann et al., 1999), these studies also suggest that there are important environmental influences that affect morphology to a greater degree than volume (Bartley et al., 1997; Steinmetz et al., 1994, 1995). Further, the available studies implicate genetic influences that may explain hemispheric differences in cerebral morphology (Tramo et al., 1995; Bartley et al., 1997), which may be relevant to the observed differences in the AC.

We did not find any relationship between AC morphological complexity and age, consistent with a neurodevelopmental model of the genesis of the AC. Therefore, given that cortical folding has been principally defined during the second and third trimester of embryogenesis (Chi et al., 1977; Huang, 1991; Naidich et al., 1994; Armstrong et al., 1995), the great proportion of environmental influences on the pattern of gyri and sulci must necessarily occur during fetal life. The likely influences during fetal development include nutritional factors, blood supply, hormonal and other factors (Geschwind and Galaburda, 1985). In contrast to gyral/sulcal pattern, gyral size may be modulated considerably postnatally (Kennedy et al., 1998), suggesting that size differences may continue to change during postnatal life. This may be influenced by nutritional factors, and has also been found to be modulated by experience; for example, the size of certain brain regions has been shown to be increased in response to highly over-learned activities (Schlaug et al., 1995a,b; Amunts et al., 1997). However, no studies to date have examined the association between experience or cognition and the size or morphology of the AC.

What Implications for Functional Neuroimaging can be Drawn from these Findings?

In the current study, not only did the arrangement of AC surface morphology differ from person to person, but it also displayed hemispheric and gender differences. Previous studies have also confirmed these marked individual (Ono et al., 1990; Vogt et al., 1995), hemispheric (Paus et al., 1996a,b; Ide et al., 1999) and gender-related differences (Paus et al, 1996a,b). As stated in the introduction, variations in sulcal/gyral pattern correlate with varying degrees to underlying cytoarchitecture (Sanides, 1964; Welker, 1990; Watson et al., 1993; Rademacher et al, 1993), and this has also been demonstrated specifically in the case of the cingulate region (Vogt et al., 1995). To the extent that specific cortical regions are associated with local anatomy, this complicates the goals of combining functional and anatomical data from more than one subject (Rademacher et al., 1993; Roland and Zilles, 1994). At present, most neuroimaging studies use a template or stereotaxic coordinate-based system (Talairach and Tournoux, 1988) that makes no allowance for individual anatomical variation. An alternative method involves ‘unfolding’ the cortical surface to form a ‘flat map’ (Van Essen and Drury, 1997). To the extent that functional regions are associated with specific anatomical landmarks that vary between individuals, unfolding may actually lead to a dispersion of sites of activation in the resulting two-dimensional map.

Another alternative method is to use a probabilistic atlas, where particular locations are assigned statistical probabilities of being located on an anatomical feature (Mazziotta et al., 1995) or within a cytoarchitectonic region. This approach was used by Paus (Paus et al., 1996a, 1998) to identify functional subregions within the cingulate/paracingulate cortex. This approach will be most effective when, in the absence of a paracingulate sulcus, the spatially corresponding region of the cortex carries out the same function. However, in a direct examination of this morpho-functional relationship, Crosson et al. (Crosson et al., 1999) conducted a functional MRI study of the cingulate region of 28 neurologically normal right-handed participants. Twentyone of the 28 in the sample were reported as demonstrating a ‘prominent-PCS’. Activity increases for word generation were centered in the PCS in 18 of these 21 and rarely extended into the cingulate sulcus (CS; 3/21). Remarkably, if there was no ‘prominent PCS’, activity nearly always extended into the CS (6/7 cases). Thus, the balance of evidence indicates that variation in the morphology of the AC usually reflects underlying functional anatomy, consistent with the anatomical findings of Vogt et al. (Vogt et al., 1995).

Therefore, problems may arise using methods that do not allow the possibility that functional regions are associated with specific anatomical features, and which attempt to correct for intersubject variability by anatomical standardization of individual brains. These two factors potentially reduce the spatial resolution of functional images by adding noise in two ways: by standardizing images spatially, and by virtue of a lack of correspondence in activation sites between individuals in the standardized space. This may explain why investigators are still uncertain about which specific areas of the AC are involved in particular functions. There needs to be greater focus on individual morpho-functional correspondences such as intra-subject averaging techniques (Steimetz et al., 1991).

What Implications for Anterior Cingulate Function can be Drawn from these Findings?

The presence of a PCS indicates that BA32 is likely to be located on a gyral crown, whereas the absence of a PCS indicates that this area is likely to be buried within a sulcal wall (Vogt et al., 1995). Given the argument of Van Essen (Van Essen, 1997) that there exists greater connectivity within gyral crowns, this raises the possibility that individuals with a paracingulate gyrus may show qualitative differences in cognitive and neuropsychological function compared to those without a paracingulate gyrus. Similarly, these differences may be apparent between males and females, function may be lateralized more or less among individuals. The nature of these differences remains unclear, mainly because the specific function of the AC and its subregions is also unclear. For example, regions broadly corresponding to BA32′ (posterior–dorsal paracingulate regions) have been activated in various functional studies by tasks involving language, novelty and word generation (Raichle et al., 1994), memory, response selection and conflict detection (Grasby et al., 1993; Carter et al., 1998, Carter et al., 2000), task-related difficulty (Paus et al., 1998), and general cognitive functions. Regions corresponding broadly to BA32 (rostral paracingulate regions) have been activated in functions involving emotion including judgements of affective content (George et al., 1995) [reviewed by Vogt and Devinsky (Vogt et al., 1992; Devinsky et al., 1995)]. It would be interesting to examine whether the paracingulate region involves any gender-related lateralized functions. The left–right asymmetry and gender differences observed in the present study would provide a neuroanatomical substrate for such findings.

Notes

We would like to thank Cathleen Geoghegan for her assistance with data collection and Dr Stephen Wood and the three anonymous reviewers for helping to improve the manuscript with their insightful comments. M.Y. was funded by a Ph.D. scholarship from La Trobe University, Bundoora, Melbourne, Australia. G.W.S. was supported by a NARSAD Young Investigator Award. This research was also supported by the Cognitive Neuropsychiatry Unit at the Mental Health Research Institute (MHRI), the National Health and Medical Research Council, the Australian Communications and Computing Institute, the Jack Brockhoff Foundation, the Ian Potter Foundation, the L.E.W. Carty Trust and the Percy Baxter Charitable Trust, Melbourne, Australia.

Address correspondence to Murat Yücel, Cognitive Neuropsychiatry Unit, c/o Mental Health Research Institute, Locked Bag 11, Parkville, Victoria, Australia 3052. Email: murat@neuro.mhri.edu.au.

Table 1

Hemispheric volume, hemispheric fissurization and hemispheric asymmetry

 Male Female Average 
Data for volume are absolute numbers of voxels for the left and right hemisphere excluding cerebellum and brainstem. Data for fissurization are expressed as a percentage of voxels in the hemisphere that are surface voxels. 
aFissurization index for each hemisphere is derived as follows: SVR × 100 – (CSF-to-brain ratio) + mean of SVR. 
bAsymmetry was derived as follows: volume asymmetry = (left volume – right volume), Fissurization index (FI) asymmetry = (left FI – right FI). 
Volume    
 Overall 1 166 776 1 022 369 1 105 973 
 Left  580 659  510 514  551 124 
 Right  586 117  511 856  554 849 
 Asymmetry  –5 458  –1 342  –3 725 
Fissuration index    
 Overalla 8.81 8.92 8.86 
 Left 9.20 9.18 9.19 
 Right 8.43 8.66 8.53 
 Asymmetryb 0.77 0.51 0.66 
 Male Female Average 
Data for volume are absolute numbers of voxels for the left and right hemisphere excluding cerebellum and brainstem. Data for fissurization are expressed as a percentage of voxels in the hemisphere that are surface voxels. 
aFissurization index for each hemisphere is derived as follows: SVR × 100 – (CSF-to-brain ratio) + mean of SVR. 
bAsymmetry was derived as follows: volume asymmetry = (left volume – right volume), Fissurization index (FI) asymmetry = (left FI – right FI). 
Volume    
 Overall 1 166 776 1 022 369 1 105 973 
 Left  580 659  510 514  551 124 
 Right  586 117  511 856  554 849 
 Asymmetry  –5 458  –1 342  –3 725 
Fissuration index    
 Overalla 8.81 8.92 8.86 
 Left 9.20 9.18 9.19 
 Right 8.43 8.66 8.53 
 Asymmetryb 0.77 0.51 0.66 
Table 2

Hemispheric and gender distribution of anterior cingulate morphological classifications

Right hemisphere morphology Left hemisphere morphology Total 
 Prominent Present Absent  
Data are absolute numbers of AC classification across the left and right hemispheres; proportions (%) are given in parentheses. 
Overall     
 Prominent 28 (16) 10 (6) 10 (6)  48 (28) 
 Present 25 (15) 17 (10) 16 (10)  58 (34) 
 Absent 30 (18) 16 (10) 19 (11)  65 (38) 
 Total 83 (49) 43 (25) 45 (26) 171 (100) 
Males     
 Prominent 20 (20)  6 (6)  4 (4) 30 (30) 
 Present 16 (16)  8 (8)  9 (9) 33 (33) 
 Absent 21 (21) 11 (11)  4 (4) 36 (36) 
 Total 57 (58) 25 (25) 17 (17) 99 (100) 
Females     
 Prominent  8 (11)  4 (6)  6 (8) 18 (25) 
 Present  9 (13)  9 (13)  7 (10) 25 (35) 
 Absent  9 (13)  5 (7) 15 (21) 29 (40) 
 Total 26 (36) 18 (25) 28 (39) 72 (100) 
Right hemisphere morphology Left hemisphere morphology Total 
 Prominent Present Absent  
Data are absolute numbers of AC classification across the left and right hemispheres; proportions (%) are given in parentheses. 
Overall     
 Prominent 28 (16) 10 (6) 10 (6)  48 (28) 
 Present 25 (15) 17 (10) 16 (10)  58 (34) 
 Absent 30 (18) 16 (10) 19 (11)  65 (38) 
 Total 83 (49) 43 (25) 45 (26) 171 (100) 
Males     
 Prominent 20 (20)  6 (6)  4 (4) 30 (30) 
 Present 16 (16)  8 (8)  9 (9) 33 (33) 
 Absent 21 (21) 11 (11)  4 (4) 36 (36) 
 Total 57 (58) 25 (25) 17 (17) 99 (100) 
Females     
 Prominent  8 (11)  4 (6)  6 (8) 18 (25) 
 Present  9 (13)  9 (13)  7 (10) 25 (35) 
 Absent  9 (13)  5 (7) 15 (21) 29 (40) 
 Total 26 (36) 18 (25) 28 (39) 72 (100) 
Table 3

Significance test results

 Left AC Right AC Asymmetry index 
 χ2 P χ2 P χ2 P 
Columns indicate the three dependent variables in the analysis, while rows represent the effects of interest and/or covariates. Degrees of freedom = 1. 
aNote that hemispheric volume and hemispheric fissurization index (FI) with regard to the dependent variable, asymmetry index, refer to hemispheric volume asymmetry and hemispheric FI asymmetry (see Results section). 
Unpartialled       
 Gender 10.85 0.001a 0.53 0.468 3.92 0.047a 
 Age 0.16 0.692 1.72 0.190 0.42 0.517 
 Hemispheric volume 3.44 0.064 0.83 0.362 0.22 0.642 
 Hemispheric FI 3.92 0.047a 1.26 0.262 1.77 0.183 
Partialled       
 Gender controlling for age, hemispheric volume and hemispheric FI 7.19 0.007a 0.10 0.754 3.84 0.049a 
 Left AC Right AC Asymmetry index 
 χ2 P χ2 P χ2 P 
Columns indicate the three dependent variables in the analysis, while rows represent the effects of interest and/or covariates. Degrees of freedom = 1. 
aNote that hemispheric volume and hemispheric fissurization index (FI) with regard to the dependent variable, asymmetry index, refer to hemispheric volume asymmetry and hemispheric FI asymmetry (see Results section). 
Unpartialled       
 Gender 10.85 0.001a 0.53 0.468 3.92 0.047a 
 Age 0.16 0.692 1.72 0.190 0.42 0.517 
 Hemispheric volume 3.44 0.064 0.83 0.362 0.22 0.642 
 Hemispheric FI 3.92 0.047a 1.26 0.262 1.77 0.183 
Partialled       
 Gender controlling for age, hemispheric volume and hemispheric FI 7.19 0.007a 0.10 0.754 3.84 0.049a 
Table 4

Anterior cingulate asymmetry index

 L(2) L(1) Symmetric R(1) R(2) n 
Data represent the morphological relationship of the AC between hemispheres indicating whether left and right hemispheres for each individual are symmetrically or asymmetrically fissured, in which direction, and by how much; (%) are given in parentheses. L(2) = leftward asymmetry with a difference of two categories, L(1) = leftward asymmetry with a difference of one category, R(2) = rightward asymmetry with a difference of two categories, L(1) = rightward asymmetry with a difference of one category. 
Overall 30 (18) 41 (24) 64 (37) 26 (15) 10 (6) 141 
Male 21 (21) 27 (27) 32 (32) 15 (15)  4 (4)  99 
Female  9 (13) 14 (19) 32 (44) 11 (15)  6 (8)  72 
 L(2) L(1) Symmetric R(1) R(2) n 
Data represent the morphological relationship of the AC between hemispheres indicating whether left and right hemispheres for each individual are symmetrically or asymmetrically fissured, in which direction, and by how much; (%) are given in parentheses. L(2) = leftward asymmetry with a difference of two categories, L(1) = leftward asymmetry with a difference of one category, R(2) = rightward asymmetry with a difference of two categories, L(1) = rightward asymmetry with a difference of one category. 
Overall 30 (18) 41 (24) 64 (37) 26 (15) 10 (6) 141 
Male 21 (21) 27 (27) 32 (32) 15 (15)  4 (4)  99 
Female  9 (13) 14 (19) 32 (44) 11 (15)  6 (8)  72 
Figure 1.

Magnetic resonance images of three right hemispheres (top) and their corresponding line drawings (bottom) are shown to illustrate the three categories of AC surface morphology.

Figure 1.

Magnetic resonance images of three right hemispheres (top) and their corresponding line drawings (bottom) are shown to illustrate the three categories of AC surface morphology.

Figure 2.

Two perspectives highlighting the confluence of the superior rostral sulcus (SRS), with the cingulate sulcus (CS) and the paracingulate sulcus (PCS). VAC, imaginary vertical line passing through anterior commissure; VPC, imaginary vertical line passing through the posterior commissure.

Figure 2.

Two perspectives highlighting the confluence of the superior rostral sulcus (SRS), with the cingulate sulcus (CS) and the paracingulate sulcus (PCS). VAC, imaginary vertical line passing through anterior commissure; VPC, imaginary vertical line passing through the posterior commissure.

Figure 3.

The basis of the automated estimate of hemispheric fissurization index. The figures above illustrate two consecutive slices in the frontal lobes of the surface voxels, while the figures below are the same surface voxels superimposed on an inverse greyscale volumetric slice at the equivalent level.

Figure 3.

The basis of the automated estimate of hemispheric fissurization index. The figures above illustrate two consecutive slices in the frontal lobes of the surface voxels, while the figures below are the same surface voxels superimposed on an inverse greyscale volumetric slice at the equivalent level.

Figure 4.

Gender differences in the proportion ‘prominent-PCS’ and ‘absent-PCS’ observed for the left hemisphere and right hemisphere of males relative to females.

Figure 4.

Gender differences in the proportion ‘prominent-PCS’ and ‘absent-PCS’ observed for the left hemisphere and right hemisphere of males relative to females.

References

Amunts K, Schlaug G, Jaencke L, Steinmetz H, Schleicher A, Dabringhaus A, Zilles K (
1997
) Motor cortex and hand motor skills: structural compliance in the human brain.
Hum Brain Map
 
5
:
205
–215
Armstrong E, Curtis M, Buxhoeveden DP, Fregoe C, Zilles K, Casanova MF, McCarthy WF (
1991
) Cortical gyrification in the rhesus monkey: a test of the mechanical folding hypothesis.
Cereb Cortex
 
1
:
426
–32.
Armstrong E, Schleicher A, Omran H, Curtis M, Zilles K (
1995
) The ontogeny of human gyrification.
Cereb Cortex
 
5
:
56
–63.
Bartley AJ, Jones DW, Weinberger DR (
1997
) Genetic variability of human brain size and cortical gyral patterns.
Brain
 
120
:
257
–69.
Biondi A, Nogueira H, Dormont D, Duyme M, Hasboun D, Zouaoui A, Chantome M, Marsault C (
1998
) Are the brains of monozygotic twins similar? A three-dimensional MR study.
Am J Neuroradiol
 
19
:
1361
–1367.
Carter CS, Braver TS, Barch DM, Botvinick MM, Noll D, Cohen JD (
1998
) Anterior cingulate cortex, error detection, and the online monitoring of performance.
Science
 
280
:
747
–749.
Carter CS, Macdonald AM, Botvinick M, Ross LL, Stenger VA, Noll D, Cohen JD (
2000
) Parsing executive processes: strategic vs. evaluative functions of the anterior cingulate cortex.
Proc Natl Acad Sci USA
 
97
:
1944
–8.
Caviness VS Jr (
1975
) Mechanical model of brain convolutional development.
Science
 
189
:
18
–21.
Caviness VS Jr, Filipek PA, Kennedy DN (
1989
) Magnetic resonance technology in human brain science: blueprint for a program based upon morphometry.
Brain Dev
 
11
:
1
–13.
Chi JG, Dooling EC, Gilles FH (
1977
) Gyral development of the human brain.
Ann Neurol
 
1
:
86
–93.
Collins DL, Zijdenbos AP, Kollokian V, Sled JG, Kabani NJ, Holmes CJ, Evans AC (
1998
) Design and construction of a realistic digital brain phantom.
IEEE Trans Med Imag
 
17
:
463
–468.
Coppola R, Myslobodsky M, Weinberger DR (
1995
) Midline abnormalities and psychopathology: how reliable is the midsagittal magnetic resonance ‘window’ into the brain?
Psychiat Res
 
61
:
33
–42.
Crosson B, Sadek JR, Bobholz JA, Gökçay D, Mohr CM, Leonard CM, Maron L, Aurebach EJ, Browd SR, Freeman AJ, Briggs RW (
1999
) Activity in the paracingulate and cingulate sulci during word generation: an fMRI study of functional anatomy.
Cereb Cortex
 
9
:
307
–316.
Crow TJ (
1990
) Temporal lobe asymmetries as the key to the etiology of schizophrenia.
Schizophr Bull
 
16
:
433
–443.
Devinsky O, Morrell MJ, Vogt BA (
1995
) Contributions of anterior cingulate cortex to behaviour.
Brain
 
118
:
279
–306.
Dixon WJ, Brwon, MB, Engllman L, Jenrich RI (1990) BMDP Statistical Software Manual. Berkeley, CA: University of California Press.
Eberstaller O (
1884
) Zur Oberflachenanatomie der Grosshirn- hemispharen.
Wien Med Blaeter
 
7
:
479
–482, 542–582, 644–646.
George MS, Ketter TA, Parekh PI, Horwitz B, Herscovitch P, Post RM (
1995
) Brain activity during transient sadness and happiness in healthy women.
Am J Psychiat
 
152
:
341
–51.
Geschwind N, Galaburda AM (
1985
) Cerebral lateralization. Biological mechanisms, associations, and pathology: II. A hypothesis and a program for research.
Arch Neurol
 
42
:
521
–552.
Goldman-Rakic PS (
1981
) Prenatal formation of cortical input and development of cytoarchitectonic compartments in the neostriatum of the rhesus monkey.
J Neurosci
 
1
:
721
–35.
Grasby PM, Frith CD, Friston KJ, Bench C, Frackowiak RS, Dolan RJ (
1993
) Functional mapping of brain areas implicated in auditory– verbal memory function.
Brain
 
116
:
1
–20.
Huang CC (
1991
) Sonographic cerebral sulcal development in premature newborns.
Brain Dev
 
13
:
27
–31.
Ide A, Dolezal C, Fernandez M, Labbe, E, Mandujano R, Montes S, Segura P, Verschae G, Yarmuch P, Aboitiz F (
1999
) Hemispheric differences in variability of fissural patterns in parasylvian and cingulate regions of human brains.
J Comp Neurol
 
410
:
235
–242.
Kennedy DN, Lange N, Makris N, Bates J, Meyer J, Caviness VS Jr (
1998
) Gyri of the human neocortex: an MRI-based analysis of volume and variance.
Cereb Cortex
 
8
:
372
–84.
Lohmann G, von Cramon DY, Steinmetz H (
1999
) Sulcal variability of twins.
Cereb Cortex
 
9
:
754
–63.
Magnotta VA, Andreasen NC, Schultz SK, Harris G, Cizadlo T, Heckel D, Nopoulos P, Flaum M (
1999
) Quantitative in vivo measurement of gyrification in the human brain: changes associated with aging.
Cereb Cortex
 
9
:
151
–60.
Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J (
1995
) A probabilistic atlas of the human brain: theory and rationale for its development. The International Consortium for Brain Mapping (ICBM).
NeuroImage
 
2
:
89
–101.
Naidich TP, Grant JL, Altman N, Zimmerman RA, Birchansky SB, Braffman B, Daniel JL (
1994
) The developing cerebral surface. Preliminary report on the patterns of sulcal and gyral maturation — anatomy, ultrasound, and magnetic resonance imaging.
Neuroimag Clin N Am
 
4
:
201
–240.
Nopoulos P, Flaum M, O'Leary D, Andreasen C (
2000
) Sexual dimorphism in the human brain: evaluation of tissue volume, tissue composition and surface anatomy using magnetic resonance imaging.
Psychiat Res
 :
98
:
1
–13.
Oldfield RC (
1971
) The assessment and analysis and handedness: the Edinburgh Inventory.
Neuropsychologia
 
9
:
97
–114.
Ono M, Kubik S, Abernathey CD (1990) Atlas of the cerebral sulci. Stuggart: Thieme Verlag.
Oppenheim JS, Skerry JE, Tramo MJ, Gazzaniga MS (
1989
) Magnetic resonance imaging morphology of the corpus callosum in monozygotic twins.
Ann Neurol
 
26
:
100
–4.
Paus T, Tomaiuolo F, Otaky N, MacDonald D, Petrides M, Atlas J, Morris R, Evans AC (
1996
) Human cingulate and paracingulate sulci: pattern, variability, asymmetry, and probabilistic map.
Cereb Cortex
 
6
:
207
–214.
Paus T, Otaky N, Caramanos Z, MacDonald D, Zijdenbos A, D'Avirro D, Gutmans D, Holmes C, Tomaiuolo F, Evans AC (
1996
) In vivo morphometry of the intrasulcal gray matter in the human cingulate, paracingulate, and superior-rostral sulci: hemispheric asymmetries, gender differences and probability maps.
J Comp Neurol
 
376
:
664
–673.
Paus T, Koski L, Caramanos Z, Westbury C (
1998
) Regional differences in the effects of task difficulty and motor output on blood flow response in the human anterior cingulate cortex: a review of 107 PET activation studies.
NeuroReport
 
9
:
37
–47.
Rademacher J, Caviness VS Jr, Steinmetz H, Galaburda AM (
1993
) Topographical variation of the human primary cortices: implications for neuroimaging, brain mapping, and neurobiology.
Cereb Cortex
 
3
:
313
–329.
Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT, Petersen SE (
1994
) Practice-related changes in human brain functional anatomy during nonmotor learning.
Cereb Cortex
 
4
:
8
–26.
Rakic P (
1988
) Specification of cerebral cortical areas.
Science
 
241
:
170
–176.
Richman DP, Stewart RM, Hutchinson JW, Caviness VS (
1975
) Mechanical model of brain convolutional development.
Science
 
189
:
18
–21.
Roland PE, Zilles K (
1994
) Brain atlases — a new research tool.
Trends Neurosci
 
17
:
458
–467.
Sanides F (
1964
) Strucutre and function of the human frontal lobe.
Neuropsychologia
 
2
:
209
–219.
Schlaug G, Jancke L, Huang Y, Staiger JF, Steinmetz H (
1995
) Increased corpus callosum size in musicians.
Neuropsychologia
 
33
:
1047
–55.
Schlaug G, Jancke L, Huang Y, Steinmetz H (
1995
) In vivo evidence of structural brain asymmetry in musicians.
Science
 
267
:
699
–701.
Siegel S, Castellan, NJ (1988) Non-parametric statistics for the behavioral sciences, pp. 65–66. New York: McGraw-Hill.
Sled JG, Zijdenbos AP, Evans AC (
1998
) A non-parametric method for automatic correction of intensity non-uniformity in MRI data.
IEEE Trans Med Imag
 
17
:
87
–97.
Steinmetz H, Seitz RJ (
1991
) Functional anatomy of language processing: neuroimaging and the problem of individual variability.
Neuropsychologia
 
29
:
1149
–1161.
Steinmetz H, Herzog A, Huang Y, Hacklander T (
1994
) Discordant brain-surface anatomy in monozygotic twins.
N Engl J Med
 
331
:
951
–952.
Steinmetz H, Herzog A, Schlaug G, Huang Y, Jancke L (
1995
) Brain (a)symmetry in monozygotic twins.
Cereb Cortex
 
5
:
296
–300.
Talairach J, Tournoux P (1988) Co-planar stereotxic atlas of the human brain. New York: Thieme.
Thompson PM, Schwartz C, Lin RT, Khan AA, Toga AW (
1996
) Three-dimensional statistical analysis of sulcal variability in the human brain.
J Neurosci
 
16
:
4261
–4274.
Tramo M, Loftus W, Thomas C, Green T, Mott L, Gazzaniga M (
1995
) Surface area of human cerebral cortex and its gross morphological subdivisions: in vivo measurements of monozygotic twins suggest differential hemisphere effects of genetic factors.
J Cogn Neurosci
 
7
:
292
–301.
Turner OA (
1948
) Growth and development of the cerebral cortical pattern in man.
Arch Neurol Psychiat
 
59
:
1
–12.
Van Essen DC (
1997
) A tension-based theory of morphogenesis and compact wiring in the central nervous system.
Nature
 
385
:
313
–316.
Van Essen DC, Drury HA (
1997
) Structural and functional analyses of human cerebral cortex using a surface-based atlas.
J Neurosci
 
17
:
7079
–7102.
Vogt BA, Finch DM, Olson CR (
1992
) Functional heterogeneity in cingulate cortex: the anterior executive and posterior evaluative regions.
Cereb Cortex
 
2
:
435
–443.
Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR (
1995
) Human cingulate cortex: surface features, flat maps, and cytoarchitecture.
J Comp Neurol
 
359
:
490
–506.
Watson JD, Myers R, Frackowiak RS, Hajnal JV, Woods RP, Mazziotta JC, Shipp S, Zeki S (
1993
) Area V5 of the human brain: evidence from a combined study using positron emission tomography and magnetic resonance imaging.
Cereb Cortex
 
3
:
79
–94.
Welker W (1990) Why does the cerebral cortex fissure and fold? A review of determinants of gyri and sulci. In: Cerebral cortex, Vol. 8b: Comparative structure and evolution of cerebral cortex, Part II (Jones EG, Peters A, eds), pp. 3–136. New York: Plenum Press.
Weinberg R (
1905
) Die Gehirnform der Polen.
Z Morphol Anthropol
 
8
:
123
–214, 279–424.
Woods RP, Grafton ST, Holmes CJ, Cherry SR, Mazziotta JC (
1998
) Automated image registration: I. General methods and intrasubject, intramodality validation.
J Comput Assist Tomogr
 
22
:
139
–52.
Zilles K, Armstrong E, Schleicher A, Kretschmann HJ (
1988
) The human pattern of gyrification in the cerebral cortex.
Anat Embryol
 
179
:
173
–179.