Skip to main content
NCBI home page
Cerebral Cortex (New York, NY) logoLink to Cerebral Cortex (New York, NY)
. 2019 Oct 30;30(3):1586–1602. doi: 10.1093/cercor/bhz189

Morphology and Spatial Probability Maps of the Horizontal Ascending Ramus of the Lateral Fissure

Trisanna Sprung-Much 1,2,✉, Michael Petrides 1,2
PMCID: PMC7132913  PMID: 31667522

Abstract

The horizontal ascending ramus of the lateral fissure (half) is a characteristic sulcus of the ventrolateral frontal cortex that forms the morphological boundary between the pars triangularis and the pars orbitalis of the inferior frontal gyrus. The present study examined the morphology of this sulcus to provide a means of identifying it accurately with magnetic resonance imaging (MRI). Voxels within the half were labeled in 50 in vivo MRI volumes (1.5 T) that had been linearly registered to the Montreal Neurological Institute stereotaxic space and the morphology of the half was categorized based on relations with neighboring sulci. The spatial variability and extent of the half were then quantified across subjects using volumetric (MINC Toolkit) and surface (FreeSurfer) spatial probability maps. The half could be identified in 95% of hemispheres, and the main morphological patterns were classified into three categories: Types I, II, and III. There were no statistically significant interhemispheric differences in the frequency of the half or its morphological patterns. Understanding the details of the sulcal morphology of this ventrolateral region is critical for an accurate interpretation of the location of activation peaks generated in functional neuroimaging studies investigating language, working memory, and other cognitive processes.

Keywords: inferior frontal gyrus, language, magnetic resonance imaging, ventrolateral frontal cortex, working memory

Introduction

In the language dominant hemisphere, the inferior frontal gyrus is traditionally referred to as “Broca’s area,” or the anterior language zone, that is, it is a region critical for speech production (Broca 1861, 1865; Wernicke 1874; Dejerine 1895). The most common anatomical interpretation of Broca’s area is that it consists of the pars opercularis and the pars triangularis (see Fig. 1) of the inferior frontal gyrus (Amunts et al. 1999; Petrides 2014). Although the clinical syndrome of Broca’s aphasia, a condition characterized by nonfluent, agrammatic, and dysprosodic speech (Mohr 1976; Mohr et al. 1978; Dronkers et al. 2007), involves damage more extensive than Broca’s area, this part of the frontal lobe, which lies anterior to the motor cortex of the precentral gyrus, is clearly necessary for certain aspects of speech function. Electrical stimulation in the language dominant hemisphere in patients undergoing brain surgery has demonstrated that interference with the normal function of the cortex on the pars opercularis and, occasionally, on the pars triangularis, will elicit an aphasic speech arrest or aphasic speech interference (Penfield and Rasmussen 1950; Penfield and Roberts 1959). Functional neuroimaging studies have also provided evidence of increased activity in the pars opercularis and the pars triangularis of the language dominant hemisphere during the performance of tasks requiring speech (Amunts et al. 2004; Heim et al. 2008; see Price 2010).

Figure 1.

Figure 1

Open in a new tab

Sulcal maps of the lateral surface of the human brain by a. Eberstaller (1890), b. Economo and Koskinas (1925), and c. Petrides (2019). The horizontal ramus has been highlighted in purple (Figure 1b reprinted from Economo C, Koskinas GN. 1925. Die Cytoarchitektonik der Hirnrinde des erwachsenen Menchen. Wein: J. Springer, with permission from Springer Nature; Figure 1c reprinted from Petrides M. 2019. Atlas of the morphology of the human cerebral cortex on the average MNI brain. New York: Academic Press, with permission from Elsevier). aalf, anterior ascending ramus of the lateral fissure; c, cap; d, sulcus diagonalis; ds, diagonal sulcus; f2, sulcus frontalis inferior; F3, gyrus frontalis inferior; F3op, pars opercularis; F3pt, pars praetriangularis; F3t, pars triangularis; h, ramus horizontalis fissurae Sylvii; half, horizontal ascending ramus of the lateral fissure; ic, incisura capi; IFG, inferior frontal gyrus; ifs, inferior frontal sulcus; iprs-i, inferior precentral sulcus—inferior branch; Op, pars opercularis; Or, pars orbitalis; p. asc., pars ascendens; p. bas., pars basilaris; pci, sulcus praecentralis inferior; p. orb., pars orbitalis; p. triang., pars triangularis; prts, pretriangular sulcus; r, sulcus radiatus; S2, ramus anterior ascendens fissurae Sylvii; S3, ramus anterior horizontalis fissurae Sylvii; s. pr.i, sulcus praecentralis inferior; Tr, pars triangularis; ts, triangular sulcus; v, ramus verticalis fissurae Sylvii. Note: the above investigators used different names and abbreviations for comparable structures: d and ds are the sulcus diagonalis; f2 and ifs are the inferior frontal sulcus; F3 and IFG both refer to the inferior frontal gyrus; F3op and Op refer to the pars opercularis; F3t, p. triang., and Tr refer to the pars triangularis; F3pt, p. orb., and Or refer to the pars orbitalis; ic and ts are the triangular sulcus; pci, s.pr.i, and iprs-i refer to the inferior precentral sulcus; S2, v, and aalf refer to the anterior ascending ramus of the lateral fissure; S3, h, and half are the horizontal ascending ramus of the lateral fissure. See Table 1.

Anatomists in the first half of the 20th century parcellated the human cerebral cortex into distinct cytoarchitectonic areas (Brodmann 1908, 1909; Economo and Koskinas 1925; Sarkissov et al. 1955). More recent investigations have expanded on these early findings (Petrides and Pandya 1994, 2002; Zilles et al. 2002). Three cytoarchitectonic areas can be identified on the inferior frontal gyrus: area 44, which lies on the pars opercularis; area 45, which occupies the pars triangularis; and area 47/12, a subdivision of Brodmann’s original area 47, which can be found occupying the pars orbitalis (Brodmann 1908, 1909; Petrides and Pandya 1994, 2002). Area 44 is dysgranular cortex, that is, its layer IV is ill-defined and rudimentary, with intruding pyramidal neurons from adjacent cortical layers III and V. Area 45, on the other hand, has a well-developed granular layer IV and very large pyramidal neurons interspersed within the deeper part of layer III (Petrides and Pandya 1994; Amunts et al. 1999; Petrides and Pandya 2002). Some researchers have parcellated area 45 into an anterior and a posterior component based on subtle differences in cytoarchitecture (Sarkissov et al. 1955; Petrides and Pandya 1994). On the anterior ventrolateral frontal cortex, Brodmann (1909) had described a large and heterogeneous region, area 47, that extended onto the orbital surface, approximately to the level of the medial orbital sulcus. Comparative cytoarchitectonic investigations of the macaque monkey and the human brain have subdivided Brodmann’s original area 47 (Petrides and Pandya 1994, 2002). Specifically, the cortex of the pars orbitalis that extends approximately to the level of the lateral orbital sulcus is now referred to as area 47/12 to indicate the part of Brodmann’s area 47 that corresponds, cytoarchitectonically, to the region of the ventrolateral frontal cortex of the macaque monkey that was labeled as area 12 by Walker (1940). Area 47/12 can be differentiated from area 45 by the absence of the very large and deeply stained pyramidal neurons that are characteristic of the deeper part of layer III in area 45. Area 47/12 can also be distinguished from the more caudal orbital cortex, which was originally included in Brodmann’s area 47, by its granular layer IV. Area 47/12 is granular cortex, while the cortex on the caudal orbital surface is dysgranular and is now referred to as area 13 because it corresponds to area 13 as defined by Walker (1940) in the macaque monkey (Petrides and Pandya 1994, 2002).

Functional imaging studies of normal human subjects (Petrides et al. 1995; Cadoret et al. 2001; Kostopoulos and Petrides 2003, 2008, 2016) have demonstrated a specific role for the pars triangularis (i.e., area 45) and pars orbitalis (i.e., area 47/12) in working memory processes. This part of the inferior frontal gyrus is believed to play a role in the active, effortful retrieval of information that has been stored in posterior association cortex (Petrides 1996, 2016). This hypothesis is consistent with the massive direct cortico-cortical connections between the anterior ventrolateral frontal cortex and modality-specific and multimodal cortex in the posterior parietal and temporal lobes (Petrides and Pandya 1984, 1988, 2002, 2009). In other words, areas 45 and 47/12 appear to be involved in the active retrieval of specific information, particularly when competing choices (i.e., interference) are present. It has, therefore, been suggested that area 45 in the language dominant hemisphere, that is, as a component of Broca’s area, is required for the retrieval of verbal and semantic information directly relevant for communication (Petrides 2006, 2014, 2016). This is in contrast to area 44, the dysgranular transition area between granular frontal cortex (i.e., area 45) and agranular motor cortex on the precentral gyrus. Studies of the connectivity of area 44 in the monkey have demonstrated direct input from the anterior part of the posterior parietal cortex involved in the processing of representations of the body, as well from the superior temporal lobe involved in the processing of auditory stimuli (Petrides and Pandya 2002, 2009), making this part of the frontal lobe an ideal candidate for phonological and articulatory regulation (Petrides 2006, 2014, 2016).

The anterior rami of the lateral fissure, namely, the anterior ascending ramus of the lateral fissure (aalf) and the horizontal ascending ramus of the lateral fissure (half), are defining sulci of the ventrolateral frontal cortex (see Table 1 for a list of the terms used to refer to these two sulci in various studies). The aalf delineates the pars opercularis from the pars triangularis (see Fig. 1c), and its morphology has recently been described in detail (Sprung-Much and Petrides 2018). The goal of the present investigation was to characterize the morphology of the half, which separates the pars triangularis from the pars orbitalis (see Fig. 1c), and to quantify its spatial variability in the Montreal Neurological Institute (MNI) stereotaxic space. The morphological data and spatial probability maps presented here can be used to improve the identification of the half in in vivo MRI, which, in turn, will improve the anatomical interpretation of the location of functional activation peaks generated in this part of the brain during language and memory-related tasks. In a similar manner, the anatomical results presented here can aid the localization of this region during brain surgery in the language dominant frontal lobe.

Table 1.

Nomenclature of the anterior rami of the lateral fissure

Authors Horizontal ramus Ascending ramus
Broca (1883) S'': branche antérieure de la scissure de Sylvius s: branche ascendante de la scissure de Sylvius
Eberstaller (1890) S3: ramus anterior horizontalis fissurae Sylvii S2: ramus anterior ascendens fissurae Sylvii
Campbell (1905) R.H.S: ramus horizontalis Sylvii R.A.S: ramus ascendens Sylvii
Cunningham (1905) s1: anterior horizontal limb (Sylvian fissure) s2: ascending limb (Sylvian fissure)
Economo and Koskinas (1925) h: ramus horizontalis fissurae Sylvii v: ramus verticalis fissurae Sylvii
Shellshear (1937) 2h: anterior horizontal limb of the fissure of Sylvius 2a: anterior ascending limb of the fissure of Sylvius
Tomaiuolo et al. (1999) Hr: horizontal ramus of the Sylvian fissure Vr: vertical ramus of the Sylvian fissure
Keller et al. (2007) hr: anterior horizontal ramus of the Sylvian fissure ar: anterior ascending ramus of the Sylvian fissure
Petrides (2012) half: horizontal ascending ramus of the lateral fissure aalf: anterior ascending ramus of the lateral fissure
Open in a new tab

Terms and abbreviations used to refer to the anterior rami of the lateral fissure according to nine different research groups. The abbreviations in Petrides (2012) are used in the current study.

Materials and Methods

Subjects

The morphology of the half was examined in a subset of the 152 MRI scans that were acquired as part of the International Consortium for Brain Mapping (ICBM) project (Mazziotta et al. 1995a, 1995b; Mazziotta et al. 2001). Subjects who participated in this project had no history of neurological and/or psychiatric illness and gave informed consent. A total of 50 MRI scans of right-handed individuals were randomly selected from the database (20 females and 30 males, mean age of 25.1 ± 5.1 years).

Magnetic Resonance Imaging

The MRI scans (1 mm isotropic) were acquired with a 1.5 T Philips Gyroscan with a fast-field echo 3D sequence in the sagittal plane (repetition time (TR) 18 ms; echo delay time (TE) 10 ms; flip angle 30°). All images were corrected for radiofrequency nonuniformities (Sled et al. 1998). For the purpose of the present study, MINC Toolkit (Collins et al. 1994) was used to register the images, using a nine-parameter linear transformation, to the asymmetric version of the most recent MNI template (Fonov et al. 2011), which evolved from the Talairach space (Talairach and Tournoux 1988) and which is the stereotaxic space widely used by the structural and functional neuroimaging community.

Sulcus Identification

The half is defined as an extension, or branch, of the lateral fissure that forms the boundary between the pars triangularis and the pars orbitalis of the inferior frontal gyrus. Additional criteria for its identification in the current study included: 1) a medial extension to the insula such that it fuses with the circular sulcus of the insula, and 2) an approximately 90° angle to its counterpart, the anterior ascending ramus (see Fig. 1). If the half emerged from the lateral fissure at a more obtuse angle, extending ventrally toward the orbital surface of the hemisphere, it was labeled as long as its branching-off point, that is, the point at which it emerged from the lateral fissure, did not lie medial to the level of the lateral orbital sulcus. Any sulcus that lay medial to the lateral orbital sulcus was not interpreted as the half but rather as one of the posterior orbital sulci (referred to as “pos” in Fig. 1c).

For details regarding the method of labeling the half in the 50 MRI volumes, the reader is referred to the methods used by Sprung-Much and Petrides (2018), who carried out a very similar morphological investigation of the anterior ascending ramus and the sulcus diagonalis in 40 MRI volumes that had been acquired as part of the ICBM project.

Probability Maps

The spatial variability and extent of the half were quantified in the form of volumetric and surface spatial probability maps according to the methods described in detail by Sprung-Much and Petrides (2018). MINC Toolkit (Collins et al. 1994) was used to generate the volumetric probability maps, and FreeSurfer (Dale et al. 1999; Fischl et al. 1999a; Fischl et al. 1999b) was used to create the surface probability maps. The reader should be aware that the purpose of generating both types of maps was not to compare the volumetric and surface registration methods, which are inherently quite different, but rather to provide the reader with two complementary analyses of this sulcus.

The volumetric spatial probability maps quantify the likelihood of each voxel in MNI stereotaxic space belonging to the half, from 0% to 100%. The maps shown in Figures 5–7 have been overlaid onto the MNI152 2009c (Fonov et al. 2011) asymmetric template that was used for registration. The MNI152 coordinates of the voxel with the maximum overlap are also provided (Figs 5 and 6).

Figure 5.

Figure 5

Open in a new tab

Volumetric spatial probability map of the half generated from the 48 left hemispheres in which it could be identified. Eleven sagittal sections are shown, with the x-coordinate indicated in the upper right corner of each section. The z and y coordinates are also indicated on the appropriate axes. The probability map has been overlaid onto the MNI152 template used for registration. The color bar indicates the extent of overlap of the labeled voxels, with a maximum overlap of 46% occurring at coordinates x − 45, y + 31, and z − 5.

Figure 7.

Figure 7

Open in a new tab

The volumetric spatial probability maps of the half are shown in the sagittal (x-coordinates), coronal (y-coordinates), and axial (z-coordinates) planes to illustrate that the half extends to the level of the insula. Top panel: volumetric map of the half from 48 left hemispheres. Lower panel: volumetric map of the half from 47 right hemispheres.

Figure 6.

Figure 6

Open in a new tab

Volumetric spatial probability map of the half generated from the 47 right hemispheres in which it could be identified. Ten sagittal sections are shown, with the x-coordinate written in the upper right corner of each section. The z and y coordinates are also indicated on the appropriate axes. The probability map has been overlaid onto the MNI152 template used for registration. The color bar indicates the extent of overlap of the labeled voxels, with a maximum overlap of 47% occurring at coordinates x + 44, y + 31, and z − 6.

The surface probability maps quantify the likelihood of each vertex on the FreeSurfer average surface belonging to the half, from 0% to 100%. The probability maps of the left and right hemispheres, which are illustrated in Figure 8, have been overlaid onto fsaverage, FreeSurfer’s average surface template used for registration. The reader should note that the sulci appear inflated because this is an average surface. Since the coordinates of surface vertices can be expressed volumetrically, the x, y, and z coordinates of the vertex with the maximum overlap are provided in MNI305 space (Evans et al. 1993), the default volumetric space used by FreeSurfer.

Figure 8.

Figure 8

Open in a new tab

Surface spatial probability maps of the half generated from (a) the 48 left hemispheres and (b) the 47 right hemispheres in which it could be identified. The probability maps have been overlaid onto the surface template, fsaverage, used for registration. The color bar indicates the level of overlap of the labeled vertices. The x, y, and z coordinates indicate the position, in MNI305 stereotaxic space, of the vertex with the maximum overlap. cs, central sulcus; ifs, inferior frontal sulcus; iprs, inferior precentral sulcus.

The statistical software package SPSS (IBM SPSS, version 23) was used to conduct the statistical analysis. The graphics design software Adobe Photoshop and Adobe Illustrator (CS6) were used to create all figures.

Results

Morphology

The half could be identified in 95% of subjects (96% of left hemispheres and 94% of right hemispheres). It could be distinguished from neighboring sulci by its medial extension to the insula, which, as Sprung-Much and Petrides (2018) demonstrated, is also the case for the aalf. The morphological patterns of the half could be categorized into three main groups, the frequencies of which are presented in Table 2. In Type I (Figs 2a and 3a), observed in 50% of cases (42% of left hemispheres, 58% of right hemispheres), the half was clearly visible on the lateral surface and did not communicate with any neighboring sulci. This is the typical pattern illustrated in traditional maps of the cerebral cortex (see Fig. 1). In Type II (Figs 2b and 3b), which was observed in 17% of cases (20% of left hemispheres and 14% of right hemispheres), the half shared a common stem with the aalf. In 11 of these hemispheres (six left and five right), this common stem was quite superficial and the rami were clearly separate within the depth of the lateral fissure. In the remaining six hemispheres (four left and two right), however, the half and aalf shared a common stem on the lateral surface and within the depth of the lateral fissure, and this common stem extended medially, to the level of the insula. Thus, in these cases, the half and aalf could never be fully differentiated (see Fig. 3b). Finally, in Type III (Figs 2c and 3c), observed in 16% of hemispheres (20% of left and 12% of right hemispheres, respectively), a third anterior ramus of the lateral fissure could be found lying between the aalf and half and extending medially to reach the circular sulcus of the insula. Of these 16 hemispheres, 11 had a third branch that was obvious on the lateral surface (seven left, four right) and which often formed a 45° angle to both the aalf and half (see Fig. 3c). In the remaining five hemispheres (three left, two right), this third branch did not extend onto the lateral surface but remained hidden within the depth of the lateral fissure.

Table 2.

Frequency of the horizontal ascending ramus of the lateral fissure and its morphological patterns in the left and right hemispheres of the human brain

Total no. LH + RH %100 LH + RH Total no.
LH		 %50
LH		 Total no.
RH		 %50
RH		 Pearson's chi-squared p Value
half identified 95 95 48 96 47 94 – –
Type I 50 50 21 42 29 58 2.560 (1.960) 0.110 (0.162)
Type II 17 17 10 20 7 14 0.638 (0.283) 0.424 (0.594)
Type III 16 16 10 20 6 12 1.190 (0.670) 0.275 (0.413)
Small sulcus ventral to half 11 11 7 14 4 8 0.919 (0.409) 0.338 (0.523)
Bifurcated half 11 11 6 12 5 10 – –
half + neighboring sulcus 4 4 2 4 2 4 – –
Hidden half 3 3 0 0 3 6 3.093 (1.375) 0.079 (0.241)
Outliers 5 5 2 4 3 6 – –
Open in a new tab

The morphological patterns of the half were examined in 50 MRI volumes (i.e., 100 hemispheres). Three major morphological patterns were identified (Types I–III), as well as four less-common configurations. Those hemispheres that had a completely atypical morphology in the ventrolateral frontal cortex were classified as outliers. Refer to the Morphology section of the Results for an explanation of the different classifications. The Pearson's chi-squared test was used to assess the significance of interhemispheric differences in frequency. The values shown in brackets are the adjusted values after applying a continuity correction.

Figure 2.

Figure 2

Open in a new tab

Schematic drawing of the lateral view of the left hemisphere illustrating the main morphological patterns formed by the half, highlighted in purple, with neighboring sulci. (a) Type I: The typical formation in which the half is clearly visible on the lateral surface and is separate from surrounding sulci. (b) Type II: The half shares a common stem, highlighted here in green, with the aalf. (c) Type III: A third anterior ramus of the lateral fissure, highlighted here in orange, can be found lying between the aalf and the half. A, anterior; D, dorsal; ds, sulcus diagonalis; ifs, inferior frontal sulcus; iprs, inferior precentral sulcus; ts, triangular sulcus.

Figure 3.

Figure 3

Open in a new tab

Individual examples of morphological patterns formed by the half with neighboring sulci. Each example is illustrated by a snapshot of a surface extraction with sulci labeled (left), followed by four sagittal sections (lateral to medial) with corresponding x coordinates indicated at the top of each section. For illustrative purposes, the MRI volumes have been de-noised according to the method described by Manjón et al. (2010). (a) Type I, the typical formation in which the half is clearly visible on the lateral surface of the hemisphere and is separate from neighboring sulci. (b) Type II, the half shares a common stem with the aalf. In this particular hemisphere, the two rami are never separate entities, i.e., they share a common stem to the level of the insula. (c) Type III, a third anterior ramus of the lateral fissure, indicated by the yellow arrow, can be found lying between the half and the aalf and traverses medially to the insula. (d) A shallow sulcus, indicated by the yellow arrow, lies ventral to the half. In this particular hemisphere, the small sulcus is visible on the lateral surface. (e) The half is bifurcated at its anterior end. ds, sulcus diagonalis; ifs, inferior frontal sulcus; iprs, inferior precentral sulcus; ts, triangular sulcus.

Additional morphological patterns were identified with lower frequency (see Table 2): 1) In 11% of the hemispheres (14% of left and 8% of right hemispheres, respectively), a small sulcus could be found ventral to the half that also extended to the circular sulcus of the insula (Fig. 3d). This sulcus was rarely observed to emerge onto the lateral surface (only in three left hemispheres) and when it did, it was never as deep or as long as the half above it. It also was observed to lie lateral to or at the level of the lateral orbital sulcus, but never medial to it. 2) In 11% of hemispheres (12% of left and 10% of right hemispheres, respectively), the half was observed to bifurcate at its anterior end (Fig. 3e). This bifurcation occurred on the lateral surface and remained superficial in 10 hemispheres. In one hemisphere, the bifurcation could only be visualized within the depth of the lateral fissure (i.e., only one branch of the bifurcation emerged onto the lateral surface). 3) The half formed a superficial connection with either the triangular sulcus, the pretriangular sulcus, or some other ventrolateral sulcus (see Fig. 1) in four cases (two left and right hemispheres; Fig. 4a). 4) Finally, in three right hemispheres, the half was not easily identifiable on the lateral surface. In two of these hemispheres, it did not extend onto the lateral surface but could be found within the depth of the lateral fissure and extending medially to the insula (Fig. 4b). In the other hemisphere, it was situated very ventrally on the lateral surface and was hidden by the temporal pole. These instances were categorized under “Hidden half” in Table 2.

Figure 4.

Figure 4

Open in a new tab

Additional examples of morphological patterns formed by the half. Each example is illustrated by a snapshot of a surface extraction with sulci labeled (left), followed by four sagittal sections (lateral to medial) with corresponding x coordinates indicated at the top of each section. For illustrative purposes, the MRI volumes have been de-noised according to the method described by Manjón et al. (2010). (a) The half blends, superficially, with a neighboring sulcus, in this case, the triangular sulcus (ts). (b) The half is not visible on the lateral surface of the hemisphere but can be found medial to the lateral surface. (c) A hemisphere with three anterior branches of the lateral fissure creates confusion as the most ventral of the three branches lies at an obtuse angle to the aalf. The middle branch, indicated by the yellow arrow, is interpreted as the additional anterior ramus of the lateral fissure and the more ventral branch is interpreted as the half. This hemisphere is classified as a Type III case. (d) An outlier case. There is only one anterior ramus of the lateral fissure, indicated by the yellow arrow. The orange arrow indicates the sulcus diagonalis (ds), a neighboring sulcus that is much more superficial (see Sprung-Much and Petrides 2018). ifs, inferior frontal sulcus; iprs, inferior precentral sulcus.

The reader should note that five hemispheres in the Type III group were difficult to classify for one of two reasons. In three left hemispheres, the third anterior branch of the lateral fissure, which would normally form a 45° angle to the aalf, formed instead an approximately 90° angle to this sulcus and, thus, took the morphology of the half (see Fig. 4c). As a result, the more ventral of the three branches was displaced even more ventrally, at quite an obtuse angle, toward the orbital surface of the hemisphere. As the half can sometimes form an obtuse angle with the aalf even when a third branch of the lateral fissure is not present, in these three difficult Type III cases, the most ventral of the three sulci was labeled as the half. In this manner, we maintained consistency in sulcal classification, assuming of course that this ventral branch was well defined and did not extend medially, beyond the lateral orbital sulcus (refer to the definition of the half in Methods). The approximate z-coordinate (dorsal–ventral axis) of the half in the contralateral hemisphere was also used as a guide for determining which sulcus should be labeled. In another two hemispheres (one left and one right) of the Type III group, the third anterior branch of the lateral fissure fused with the aalf within the depth of the hemisphere. It was, therefore, difficult to decide whether these cases should be classified as Type III or considered as simply a bifurcated aalf.

Finally, five hemispheres were classified as outlier cases (see Table 2) because of their atypical sulcal morphology in the region of the ventrolateral frontal cortex. In four hemispheres (one left and three right hemispheres, respectively), only a single anterior ramus of the lateral fissure could be identified. In two of these hemispheres, the single branch emerged from the lateral fissure at a 45° angle (see Fig. 4d). In these instances, it was impossible to determine whether this sulcus should be labeled as the aalf or the half. In the other two hemispheres, the single branch satisfied the criteria of the half but an aalf could not be identified. A fifth hemisphere (left) was classified as an outlier because the half appeared to exist in two pieces, a completely uncharacteristic morphology for this sulcus.

A total of 16 hemispheres (eight left and eight right hemispheres, respectively) could be classified into more than one morphology group. Specifically, one left hemisphere of the Type III morphology group also had a bifurcated half. Two right hemispheres of the Type II morphology group had a bifurcated half. One right hemisphere had a bifurcated half that also superficially communicated with the triangular sulcus, and another right hemisphere of the Type III morphology group had a half that was hidden from the lateral surface and that also superficially fused with the pretriangular sulcus. Finally, the 11 hemispheres that had a small sulcus located underneath the half could also be classified into other morphology groups. It is for this reason that the total number of hemispheres in Table 2 reaches 117, instead of 100.

Interhemispheric Differences

A chi-squared test of independence was used to investigate the significance of interhemispheric differences in the frequency with which the morphological patterns of Type I, Type II, Type III, a small sulcus ventral to the half, and a hidden half were identified. Table 2 presents the results. There was no statistically significant difference between the number of left and right hemispheres for any of these comparisons. Although the difference in the number of left and right hemispheres that contained a hidden half did approach significance, this was no longer the case when a more conservative p value was used in a continuity correction, which is required for two-by-two contingency tables.

Spatial Probability Maps

The morphological variability and spatial extent of the half were quantified, separately, for the left and right hemispheres. Volumetric spatial probability maps were created to show the morphological variability and spatial extent of the sulcus along the three axes of MNI stereotaxic space. Surface spatial probability maps were created to quantify the data within a 2D surface space. The reader is once again reminded that the purpose of using both methods was not to compare the two but simply to provide an overview of the variability of the position of this sulcus within the ventrolateral frontal cortex.

Volumetric Maps

Figures 5 and 6 illustrate the volumetric spatial probability maps of the half for the left and right hemispheres, respectively. These maps have been overlaid onto the asymmetric MNI template (Fonov et al. 2011) used for registration. For each map, a series of sagittal sections is used to show the extent of the sulcus from the lateral surface to the insula. The x coordinate of the respective sagittal section is indicated in its top-right corner. A grid has been overlaid onto each section to show the y (anterior–posterior) and z (dorsal–ventral) axes. The color bar at the bottom of each figure indicates the extent of overlap of the labeled voxels. For the purpose of illustration, the maps have been thresholded to a minimum of 10%.

In Figure 5, the morphological variability and spatial extent of the half have been quantified from the 48 left hemispheres in which it could be identified. The sulcus starts, laterally, at an x-coordinate of −53 and terminates, at the insula, at an x-coordinate of −33. The maximum overlap across hemispheres reaches 46%, the voxel of which can be found at MNI coordinates x − 45, y + 31, and z − 5. In a similar manner, the spatial probability map of the half shown in Figure 6 has been generated from the 47 right hemispheres in which it could be identified. The sulcus commences at an x-coordinate of +51 on the lateral surface and finishes, once again, at the level of the insula, at x + 33. The maximum overlap across these hemispheres reaches 47%, the voxel of which can be found at MNI coordinates x + 44, y + 31, and z − 6.

Table 3 summarizes the above-mentioned results.

Table 3.

MNI152 stereotaxic coordinates of the horizontal ascending ramus of the lateral fissure with the highest probability value: volumetric spatial probability maps

x y z Probability (%)
half (lh) −45 +31 −5 46
half (rh) +44 +31 −6 47
Open in a new tab

Coordinates are in MNI152 stereotaxic space. The probability values represent the maximum overlap of the labeled voxels from 0 to 100%.

lh, left hemisphere; rh, right hemisphere.

To illustrate further the medial extension of the half to the level of the insula, Figure 7 shows the same spatial probability maps along the sagittal, coronal, and axial planes. In the top panel, snapshots of the map of the left hemispheres from Figure 5 are shown at coordinates x − 42, y + 28, and z − 1. Similarly, the bottom panel shows the map of the right hemispheres from Figure 6 at x + 40, y + 28, and z − 3. The reader can see that the aalf also extends to the insula.

Surface Maps

Figure 8 illustrates the surface spatial probability maps of the half for both the left and right hemispheres. The data have been overlaid onto the surface template, fsaverage, that was used for registration (Fischl et al. 1999a; Fischl et al. 1999b). The color bar at the top of each image shows the extent of overlap of the labeled vertices of the surfaces. For the purpose of illustration, the maps have been thresholded to a minimum of 20%. The coordinates at the bottom of each image indicate the location, in MNI305 stereotaxic space, of the vertex with the maximum overlap.

In Figure 8a, the morphological variability and spatial extent of the half have been quantified from the 48 left hemispheres in which it could be identified. The maximum overlap reaches 75% and occurs at MNI coordinates x − 38.7, y + 31.2, and z − 4.0. Figure 8b shows the results of a comparative analysis from 47 right hemispheres. Here, the maximum overlap is slightly lower, reaching 67%, at MNI coordinates x + 48.1, y + 29.3, and z − 10.7. The reader will notice that, as in the volumetric maps, the half traverses medially to reach the insula in both hemispheres.

Table 4 summarizes the above-mentioned results.

Table 4.

MNI305 stereotaxic coordinates of the horizontal ascending ramus of the lateral fissure with the highest probability value: surface spatial probability maps

x y z Probability (%)
half (lh) −38.7 +31.2 −4.0 75
half (rh) +48.1 +29.3 −10.7 67
Open in a new tab

Coordinates are in MNI305 stereotaxic space, which is the default volumetric space of the fsaverage template in FreeSurfer. The probability values represent the maximum overlap of the labeled vertices from 0 to 100%.

lh, left hemisphere; rh, right hemisphere.

Discussion

The present study examined the morphology of the horizontal ascending ramus of the lateral fissure, a defining sulcus of the ventrolateral frontal cortex of the human brain. 50 MRI volumes (i.e., 100 hemispheres), registered to MNI stereotaxic space, were examined to determine the frequency of the half in left and right hemispheres, as well as the morphological patterns formed with neighboring sulci. The spatial variability and the extent of this sulcus were then quantified in the left and right hemispheres using spatial probabilistic mapping.

Morphology

The data presented here demonstrate that the half is a frequent sulcus with a comparable incidence in left and right hemispheres (96% and 94%, respectively, refer to Table 2). It could be distinguished from neighboring sulci, such as the triangular sulcus, by its considerable depth within the lateral fissure and its fusing with the circular sulcus of the insula. Previous studies have reported similar incidence rates to those presented here. Keller et al. (2007), who performed a morphological and stereological investigation of Broca’s area in 50 in vivo MRI volumes in an attempt to determine asymmetries in morphology and gray matter volume, identified the horizontal ramus in 98% of left and right hemispheres. Powell et al. (2012), who investigated how sex and handedness affect the morphology and volume of this region, were able to identify this sulcus in 96% of left hemispheres and in all of their right hemispheres (a total of 82 MRI scans). Ono et al. (1990), who studied the major sulci of the entire cerebral cortex in postmortem specimens, reported a horizontal ramus in 84% of left hemispheres and 92% of right hemispheres. This research group restricted most of their examination to the surface of the hemispheres, which might have prevented detection of the horizontal ramus when it remained hidden within the lateral fissure. The reader should also note that, in those hemispheres in which Ono et al. (1990) did not identify a horizontal ramus (i.e., in 16% of left and 8% of right hemispheres, respectively), they always labeled the single anterior ramus on the lateral surface as the anterior ascending ramus. In the current investigation, the hemispheres that consisted of a single anterior ramus (4% of all hemispheres studied, see Results) were grouped as outliers.

Three major morphological patterns of the half were identified, in addition to less frequent patterns. Type I consisted of those hemispheres in which the half was identifiable on the lateral surface and was separate from surrounding sulci (Fig. 2a). This is the typical morphology illustrated in traditional schematic illustrations (see Fig. 1a,b). Type II consisted of hemispheres in which the half shared a common stem with the aalf (Fig. 2b). This common stem was either superficial or extended to the level of the insula. Type III cases were those in which a third anterior branch of the lateral fissure could be identified, lying between the half and aalf (Fig. 2c). This third branch either emerged onto the lateral surface or remained hidden within the depth of the hemisphere, but in all cases extended to the insula, just like the half and aalf. The frequencies of a common stem and a third anterior branch reported here (17% and 16%, respectively, see Table 2) are similar to those described by Sprung-Much and Petrides (2018), who performed an analysis of the morphology of the aalf and its posterior neighbor, the sulcus diagonalis. In the 40 MRI volumes examined in that study, the aalf shared a common stem with the half in 12.5% of hemispheres, and a third anterior branch of the lateral fissure was identified in 17.5% of hemispheres. A single anterior branch of the lateral fissure was observed in 3.75% of all hemispheres (as opposed to 4% in the current study). Keller et al. (2007) identified a common stem in 29% of their cases (30% left and 28% right hemispheres, respectively), which is comparable to the results reported by Ono et al. (1990), who identified a “common trunk” in 26% of hemispheres (24% left and 28% right hemispheres, respectively). Cunningham (1905) and Eberstaller (1890), who carried out extensive examinations of the morphology of the cerebral cortex at the turn of the 20th century, provided detailed descriptions of their observations. With regard to the anterior rami of the lateral fissure, Cunningham (1905) writes: “In many cases, the two anterior limbs spring from a common stem […] and not infrequently both are replaced by a single anterior limb” (p. 556). Although Eberstaller’s work has never been officially translated from the original German, Bailey and Bonin (1951) refer to segments of his manuscripts in their atlas of the human cerebral cortex. When describing the branches of the lateral (Sylvian) fissure, Bailey and Bonin state: “The anterior rami vary considerably. There may be only one ramus (I), there may be two, forming the shape of the letter V, U, or Y, and there occasionally may be three rami” (p. 37). This “I” formation refers to the presence of a single anterior ramus of the lateral fissure, and the “Y” formation to a common stem shared by the anterior rami (Eberstaller 1890, p. 19). Clearly, the anterior rami of the lateral fissure are much more variable than traditional maps would suggest.

In the present investigation, a small sulcus located ventral to the half and which extended to the insula could be identified in 11% of hemispheres (see Table 2 and Fig. 3d). When defining the anterior branches of the Sylvian fissure, Eberstaller (1890) mentions the occasional occurrence of a third anterior branch, which he names “S4” (his “S1” refers to the main branch of the Sylvian fissure, and “S2” and “S3” to the anterior ascending and horizontal rami, respectively. See Figure 1a of the present study). He describes this small sulcus, which he found in 22% of hemispheres, as lying underneath the horizontal ramus and reaching a length of 0.5–1 cm. He interpreted this sulcus in relation to the accessory gyrus brevis of the insula and suggested that its shape on the inner mantel of the frontal operculum is influenced by the accessory gyrus brevis of the insula (Eberstaller 1887, p. 743). His observations differ from those described here, however, in that he noticed that this branch is well developed when the ascending and horizontal rami are not. In the current study, the small sulcus under the half rarely emerged onto the lateral surface. Additionally, Eberstaller stated that the sulcus characteristically lies medial to the lateral orbital sulcus but this was not a defining feature in the current investigation. It is, therefore, not clear whether Eberstaller’s S4 corresponds to the small sulcus observed here or to one of the posterior orbital sulci that are often located on the orbital surface of the hemisphere, lying medial to the lateral orbital sulcus (labeled as “pos” for “posterior orbital sulcus” in Figure 1c).

Spatial Probability Maps

The spatial probability maps presented in Figures 5–8 illustrate the quantified spatial variability and extent of the half across subjects. In Figures 5 and 6, the probability of each voxel of the MNI152 template belonging to the half of the left and right hemisphere, respectively, has been established. In comparing the spatial probability maps in Figures 5 and 6, a few trends become obvious: 1) The maximum overlap (i.e., maximum probability) across hemispheres is comparable for the left and right hemispheres, reaching 46% in 48 left and 47% in 47 right hemispheres, respectively. 2) The MNI152 stereotaxic coordinates of the voxel in which the maximum overlap occurs are also comparable for the left and right hemispheres (refer to Table 3). 3) The spatial probability map of the left hemispheres forms a slightly longer and leaner shape than that of the right hemispheres (compare, for instance, the sagittal sections x − 49, x − 47, x – 45, and x − 43 of Fig. 5 with x + 49, x + 47, x + 45, and x + 43 of Fig. 6), suggesting that the half maintains a somewhat more stable and consistent position within the inferior frontal gyrus in the left hemisphere. It is interesting to note that the maximum probability values of the left and right hemispheres in these volumetric maps are similar to those of the aalf as reported by Sprung-Much and Petrides (2018). In 40 left hemispheres, the maximum overlap of the aalf reached 42%, and in 37 right hemispheres it reached 46% (refer to Figs 5 and 6 in Sprung-Much and Petrides 2018). Finally, the reader will note that the volumetric maps of the present study confirm what the individual morphology analysis demonstrated, namely, that the half extends medially to merge with the circular sulcus of the insula (also illustrated in Fig. 7).

Surface spatial probability maps of the half were generated using a surface registration technique that aligns the cortical folds of an individual surface to those of an average surface (Fischl et al. 1999a; Fischl et al. 1999b). The location of surface vertices can then be described volumetrically, e.g., in MNI stereotaxic space. In Figure 8, the probability of each vertex of the average surface template belonging to the half has been established separately for the left and right hemispheres. The reader can observe a similar trend to the volumetric maps with regard to the shape of the probability map itself, in which the overall shape is longer and leaner in the map of the left hemispheres compared to that of the right hemispheres. This observation, once again, suggests that the half has a more consistent position within the inferior frontal gyrus in the left hemisphere. The maximum overlap across subjects is slightly higher in the left hemispheres compared to the right hemispheres (75% vs. 67%, respectively) and the MNI coordinates of the vertex with the maximum probability are even less comparable: the sulcus in the right hemisphere is most likely to be found at an x coordinate that is 10 mm lateral to the equivalent position in the left hemisphere (refer to Table 4). The maximum probability values are, however, similar to those of the surface spatial probability maps of the aalf (68% in the left hemispheres, 70% in the right hemispheres) presented by Sprung-Much and Petrides (2018). Finally, as in the volumetric maps of Figures 5, 6, and 7 of the present study, one can see in Figure 8 that the half extends medially to the insula.

Hemispheric Differences

Much effort has been expended on explanations of the functional hemispheric dominance of the ventrolateral frontal cortex through an underlying anatomical asymmetry of this region. Studies of surface area, volume, or sulcal morphology of the pars opercularis and pars triangularis, however, have proven inconclusive. Keller et al. (2009) provide an excellent summary of the conflicting results of several in vivo imaging and postmortem investigations, some of which are also reviewed by Sprung-Much and Petrides (2018). As Keller et al. (2009) describe, there are multiple potential reasons for the discrepancies in results across these studies, which include a variability in the anatomical definition of the region of interest, a variability in the actual measure used, and the age and handedness (assuming that this latter factor is considered) of the subjects. Foundas et al. (1998), for instance, measured the surface area of the pars opercularis and pars triangularis in the MRI volumes of 16 right handers and 16 left handers. They found a significantly larger surface area of the pars triangularis in the left hemispheres of right-handed individuals and a smaller leftward asymmetry in the left handers. They also found a leftward asymmetry of the pars opercularis in the right handers, but a rightward asymmetry in the left handers. This research group decided to use the anterior subcentral sulcus as the posterior boundary of the pars opercularis because the inferior precentral sulcus, which is normally used as the posterior boundary, does not always extend below the level of the inferior frontal sulcus. Knaus et al. (2007) measured gray matter volume of the pars opercularis and pars triangularis in 60 MRI volumes of right-handed individuals. They reported a leftward asymmetry for the pars triangularis, but a rightward asymmetry for the pars opercularis. Keller et al. (2007), on the other hand, who also measured the gray matter volume of the pars opercularis and pars triangularis in 45 MRI volumes but whose sample consisted of both right and left handers, reported a statistically significant leftward asymmetry of the pars opercularis and a nonsignificant leftward asymmetry of the pars triangularis. Tomaiuolo et al. (1999) did not find any differences in gray matter volume of the pars opercularis in 50 MRI volumes of right-handed subjects. Finally, Powell et al. (2012) found a rightward volume asymmetry of the pars opercularis in left-handed males and females and a leftward trend in right-handed males. In terms of postmortem investigations, Falzi et al. (1982) measured the extent of the pars opercularis and pars triangularis, together, in photographs of the lateral surfaces of 12 postmortem brains from right-handed individuals. These researchers, who included intrasulcal (i.e., the cortex hidden within a sulcus) data in their analysis by also taking measures from photographs of subsequent sections of the region of interest, found a leftward asymmetry in 75% of their specimens. Conversely, Wada et al. (1975), who performed planimetric measures of the pars opercularis and the posterior pars triangularis on photographs of the lateral surfaces of 100 adult and 85 infant brains and who did not measure intrasulcal cortex, found a slight rightward asymmetry in their sample.

Cytoarchitectonic studies of areas 44 and 45 have reported a general trend of leftwards asymmetry (see Keller et al. 2009). Amunts et al. (1999) found a significant leftward asymmetry in the volume of area 44 in 10 postmortem brains for which the handedness of the subjects was unknown. These researchers reported a nonsignificant leftward volume asymmetry of area 45 from the same brains. Hayes and Lewis (1995) measured the cross-sectional area of magnopyramidal neurons (i.e., the neurons in the deepest part of layer III) in area 45 in 19 specimens and found that the largest of these neurons were significantly larger in the left hemispheres compared to the right hemispheres. Finally, Uylings et al. (2006) reported a larger number of neurons in left area 44 that was significant in the male brains (five specimens) of their sample, and a larger number of neurons in left area 45, although not significant, in the female brains (five specimens). The reader is referred to Keller et al. (2009) for an extensive review of the above and many additional studies.

In the present investigation of 50 MRI volumes of all right handers, interhemispheric differences in the frequency with which the half was identified, as well as in the frequency of its morphological patterns, were determined (chi-squared test of independence). Specifically, interhemispheric differences in the frequency of the morphological patterns of Type I, Type II, Type III, a small sulcus under the half, and a hidden half were analyzed. As can be seen in Table 2, none of the comparisons were significant. Although the difference in frequency of a hidden half across left and right hemispheres did approach significance (p = 0.079), this was no longer the case with a continuity correction, required for two-by-two contingency tables. These results are comparable to those in the study of Sprung-Much and Petrides (2018) who did not find any significant interhemispheric differences in the frequency of the aalf, the sulcus diagonalis, or their morphological patterns in 40 MRI volumes.

Limitations in Sample Size

Manual delineations of anatomical structures require a considerable amount of work and, therefore, the sample size is limited. The current study involved manual delineations of the half in 50 subjects (i.e., 100 hemispheres), which is a sample size comparable to those used in other studies of sulcal morphology, e.g., Tomaiuolo et al. (1999), Huntgeburth and Petrides (2012), Segal and Petrides (2012), and Zlatkina and Petrides (2010). Although none of the statistical interhemispheric comparisons of morphology in the present study were significant, the interhemispheric comparison of the hidden half morphology group showed a trend toward significance (p value of 0.079) before a continuity correction was applied. The effect size of this particular group is small (phi of 0.176) and detecting such a small effect would require a sample size consisting of anywhere between 196 and 785 hemispheres at an appropriate statistical power of 0.8 (Cohen 1988). Such a large sample size is beyond the scope of studies that use manual segmentations. One solution for this is to implement machine learning techniques in future studies of sulcal morphology to automate the detection of sulci of interest in hundreds of subjects using an appropriate sample of manually labeled training data.

Sulci as Morphological Landmarks

Anatomical investigations in the early 20th century demonstrated strong correlations between the location of specific folds of the cerebral cortex and the location of particular cytoarchitectonic areas involved in basic functions (Campbell 1905; Smith 1907; Brodmann 1909; Economo and Koskinas 1925). The general consensus has been that certain sulci, such as the central sulcus or the calcarine sulcus, which maintain a relatively stable morphology across hemispheres and individuals, can serve as morphological landmarks for sensory and motor cortical areas (Fischl et al. 1999b; Fischl et al. 2007). More recently, several studies have established additional anatomo-functional relationships involving more variable cortical folds. Segal and Petrides (2013), for instance, studied the relationship between the morphology of the sulci of the angular gyrus and the location of functional activation peaks in the left hemispheres of nine right-handed subjects performing reading-related tasks. These investigators were able to demonstrate that certain functional peaks of interest were consistently located between the central and posterior branches of the caudal superior temporal sulcus. In eight subjects, Amiez et al. (2006) compared the part of the dorsal premotor cortex that is activated during saccadic eye movements (i.e., the frontal eye fields) with a part of the same region of cortex that is activated during visuomotor conditional tasks involving the hand. Until this point, these two very different functional loci had been reported as having overlapping locations at the junction of the caudal superior frontal sulcus with the superior precentral sulcus, results that did not match the functional organization of the homologous region in the monkey, where saccadic eye movements and visuomotor conditional tasks of the hand engage distinct regions of the premotor cortex. Rather than averaging the functional data across subjects as previous studies had done, Amiez et al. (2006) performed the functional analysis at the individual subject level. In doing so, they were not only able to establish that these loci have separate positions within the dorsolateral frontal cortex, but their locations could actually be predicted by local sulcal folds. Huntgeburth and Petrides (2016) studied the morphology of the rhinal sulcus in 40 in vivo MRI volumes and compared their results to the location of the entorhinal cortex. By overlapping their spatial probability map of the rhinal sulcus with a spatial probability map of ex vivo data of the entorhinal cortex in the human brain, these investigators were able to demonstrate what had been known in the monkey for some time, namely, that the rhinal sulcus delineates the lateral/ventral border of the entorhinal cortex.

The general relationship between the location of the half and that of cytoarchitectonic area 45 was examined by overlaying surface labels of area 45, the cytoarchitectonic data of which were derived from the histological analysis of 10 postmortem brains at the Jülich Research Centre (see Amunts et al. 1999), with the surface spatial probability maps of the horizontal ramus provided in the present study. The cytoarchitectonic data, which were originally represented as volumetric spatial probability maps (Amunts et al. 2004), and later reproduced as surface spatial probability maps using FreeSurfer (Fischl et al. 2007), are available as surface labels in the FreeSurfer pipeline. In Figure 9, a thresholded version of the surface label of area 45, also available in FreeSurfer, has been incorporated, such that the surface area of this average label best represents the average surface area of the label in each of the 10 brains. Traditional schematic cytoarchitectonic maps of the human cortex suggest that the half approximately delineates the ventral border of area 45, which lies on the pars triangularis. Figure 9 confirms this suggestion using quantitative data derived from multiple subjects. For relationships between area 45, area 44, and the aalf, the reader is referred to the recent study by Sprung-Much and Petrides (2018).

Figure 9.

Figure 9

Open in a new tab

The surface spatial probability maps of the half are shown in conjunction with an outline of surface labels of area 45, the cytoarchitectonic data of which were generated from the histological analysis of 10 postmortem brains at the Institute of Neuroscience and Medicine (INM-1) in Jülich, Germany (see Amunts et al. 1999). The cytoarchitectonic data have been reproduced as surface labels registered to fsaverage (see Fischl et al. 2007) and a thresholded version of the labels, freely available in the FreeSurfer pipeline, has been incorporated here. The outline of the labels has been filled for clarity using open circles. All data have been overlaid onto the fsaverage template used for registration.

Future histological investigations of postmortem specimens will be necessary to examine further the relation between area 45 and the various morphological patterns described here by comparing the location of the cytoarchitectonic borders of area 45 to neighboring cortical folds. To do this, sulci will have to be sectioned optimally, that is, at a 90° angle, in order to capture the cortical layers in their entirety and determine accurately the precise relationship of cytoarchitecture to sulci (Novek et al. 2019). Economo and Koskinas (1925) emphasized the importance of optimal sectioning in cytoarchitectonic analysis because of the convoluted nature of the cortex and stated that the standard procedure of sectioning whole specimens along a single plane (e.g., the coronal plane) provides a distorted picture of the cytoarchitecture in regions of cortex that do not lie perpendicular to that plane. Optimal sectioning of sulci with highly atypical morphology in the region of the inferior frontal gyrus, such as the ventral half and the outlier cases described here, will be necessary to establish precise relations between cytoarchitecture and variations in sulcal morphology.

Concluding Remarks

The present study provides an overview of the morphology of the horizontal ascending ramus of the lateral fissure, which is a characteristic sulcus of the inferior frontal gyrus. In the language dominant hemisphere, this region is known as the anterior language zone (Broca’s area). The anatomical definition of the anterior language zone varies across studies, but the most prevalent interpretation refers to a combination of the cortex of the pars opercularis and the pars triangularis (see Tremblay and Dick 2016 for a review of the variable definitions of this region). Broca (1865) originally emphasized the importance of the left posterior inferior frontal gyrus in speech, but scans of the brains of his classic patients LeBorgne and LeLong, 150 years later, revealed extensive damage to neighboring cortex, underlying white matter, and even subcortical structures in the case of the brain of LeBorgne (Dronkers et al. 2007). Functional imaging has demonstrated an undeniable role of surrounding cortex in speech functions, including the cortex of the pars orbitalis at the anterior inferior frontal gyrus, and also the dorsolateral frontal cortex (see Ardila et al. 2016). There is also evidence for a role of the insula and the supplemental motor cortex in speech processes. For example, in a study of 25 stroke patients who suffered from apraxia of speech, Dronkers (1996) found that the area of overlapping damage across these patients was the anterior insula in the left hemisphere, and this region was completely spared in patients who did not have speech apraxia. Chapados and Petrides (2013) demonstrated that dorsomedial frontal lesions in the left hemisphere that include the supplementary motor cortex can result in deficits in verbal fluency.

Regardless of the precise anatomical definition of the anterior language zone, the cortex within and surrounding the half in the language dominant hemisphere is clearly important for some aspect of speech function. In both hemispheres, the cortex of the pars triangularis and that of the pars orbitalis has been shown to be engaged during active, effortful retrieval of specific information that has likely been stored in posterior cortex (Petrides et al. 1995; Cadoret et al. 2001; Kostopoulos and Petrides 2003, 2008, 2016). In the left hemisphere, this retrieval seems to be related to semantic information, consistent with recent research that has emphasized the importance of a ventral pathway for language comprehension connecting the middle temporal lobe and the ventrolateral prefrontal cortex via the extreme capsule (Saur et al. 2008). The morphological descriptions presented in the current investigation can be used to improve the identification of the half in histological studies of the inferior frontal gyrus, and to aid neuronavigation during brain surgery, where it is imperative to avoid damage to speech areas. Finally, the stereotaxic coordinates provided here can facilitate the identification of this sulcus in MRI to improve the anatomical interpretation of functional imaging studies that engage language-related and nonlinguistic memory processes.

Funding

Canadian Institutes of Health Research (CIHR) Foundation Grant (FDN-143212 awarded to M.P.) and Fonds de Recherche du Québec - Santé (scholarship awarded to T.S.-M.).

Notes

We thank Kristina Drudik for her assistance with labeling the horizontal ramus in the MRI scans. We thank Guido Guberman and Rhonda Amsel for statistical advice. We also thank Philip Novosad for technical assistance with Matlab and MINC Toolkit, as well as for providing helpful feedback during manuscript revision, and Guy Sprung for assistance in translating from German pertinent sections of Eberstaller’s manuscript.

Conflict of Interest: None declared.

References

  1. Amiez C, Kostopoulos P, Champod A-S, Petrides M. 2006. Local morphology predicts functional organization of the dorsal premotor region in the human brain. J Neurosci. 26:2724–2731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Amunts K, Schleicher A, Bürgel U, Mohlberg H, Uylings H, Zilles K. 1999. Broca's region revisited: cytoarchitecture and intersubject variability. J Comp Neurol. 412:319–341. [DOI] [PubMed] [Google Scholar]
  3. Amunts K, Weiss PH, Mohlberg H, Pieperhoff P, Eickhoff S, Gurd JM, Marshall JC, Shah NJ, Fink GR, Zilles K. 2004. Analysis of neural mechanisms underlying verbal fluency in cytoarchitectonically defined stereotaxic space—the roles of Brodmann areas 44 and 45. Neuroimage. 22:42–56. [DOI] [PubMed] [Google Scholar]
  4. Ardila A, Bernal B, Rosselli M. 2016. How localized are language brain areas? A review of Brodmann areas involvement in oral language. Arch Clin Neuropsychol. 31:112–122. [DOI] [PubMed] [Google Scholar]
  5. Bailey P, Bonin G. 1951. The Isocortex of Man. Urbana: University of Illinois Press. [Google Scholar]
  6. Broca P. 1861. Nouvelle observation d’aphémie produite par une lésion de la moitié postérieure des deuxième et troisième circonvolutions frontales. Bull Soc Anatom Paris. 36:398–407. [Google Scholar]
  7. Broca P. 1865. Sur le siège de la faculté du langage articulé. Bull Soc Anthropol Paris. 6:377–393. [Google Scholar]
  8. Broca P. 1883. Description élémentaire des circonvolutions cérébrales de l'homme. Rev Anthropol. 6:1–34. [Google Scholar]
  9. Brodmann K. 1908. Beiträge zur histologischen Lokalisation der Grosshirnrinde. VI. Mitteilung: die Cortexgliederung des Menschen. J Psychol Neurol. 10:231–246. [Google Scholar]
  10. Brodmann K. 1909. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: Barth. [Google Scholar]
  11. Cadoret G, Pike GB, Petrides M. 2001. Selective activation of the ventrolateral prefrontal cortex in the human brain during active retrieval processing. Eur J Neurosci. 14:1164–1170. [DOI] [PubMed] [Google Scholar]
  12. Campbell AW. 1905. Histological Studies on the Localisation of Cerebral Function. Cambridge: Cambridge University Press. [Google Scholar]
  13. Chapados C, Petrides M. 2013. Impairment only on the fluency subtest of the Frontal Assessment Battery after prefrontal lesions. Brain. 136:2966–2978. [DOI] [PubMed] [Google Scholar]
  14. Cohen J. 1988. Statistical Power Analysis for the Social Sciences. New Jersey: Lawrence Erlbaum Associates. [Google Scholar]
  15. Collins DL, Neelin P, Peters TM, Evans AC. 1994. Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr. 18:192–205. [PubMed] [Google Scholar]
  16. Cunningham DJ. 1905. Textbook of Anatomy. New York: W. Wood and company. [Google Scholar]
  17. Dale AM, Fischl B, Sereno MI. 1999. Cortical surface-based analysis: I. Segmentation and surface reconstruction. Neuroimage. 9:179–194. [DOI] [PubMed] [Google Scholar]
  18. Dejerine J. 1895. Anatomie des Centres Nerveux. Paris: Rueff et Cie. [Google Scholar]
  19. Dronkers NF. 1996. A new brain region for coordinating speech articulation. Nature. 384:159–161. [DOI] [PubMed] [Google Scholar]
  20. Dronkers NF, Plaisant O, Iba-Zizen MT, Cabanis EA. 2007. Paul Broca's historic cases: high resolution MR imaging of the brains of Leborgne and Lelong. Brain. 130:1432–1441. [DOI] [PubMed] [Google Scholar]
  21. Eberstaller O. 1887. Zur Anatomie und Morphologie der Insula Reilii. Anat Anz. 2:739–750. [Google Scholar]
  22. Eberstaller O. 1890. Das Stirnhirn. Wein: Urban & Schwarzenberg. [Google Scholar]
  23. Economo C, Koskinas GN. 1925. Die Cytoarchitektonik der Hirnrinde des erwachsenen Menschen. Wein: J. Springer. [Google Scholar]
  24. Evans AC, Collins DL, Mills S, Brown E, Kelly R, Peters TM. 1993. 3D statistical neuroanatomical models from 305 MRI volumes. Nuclear Science Symposium and Medical Imaging Conference: IEEE Conference Record. p 1813–1817.
  25. Falzi G, Perrone P, Vignolo LA. 1982. Right-left asymmetry in anterior speech region. Arch Neurol. 39:239–240. [DOI] [PubMed] [Google Scholar]
  26. Fischl B, Rajendran N, Busa E, Augustinack J, Hinds O, Yeo BT, Mohlberg H, Amunts K, Zilles K. 2007. Cortical folding patterns and predicting cytoarchitecture. Cereb Cortex. 18:1973–1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Fischl B, Sereno MI, Dale AM. 1999a. Cortical surface-based analysis: II. Inflation, flattening, and a surface-based coordinate system. Neuroimage. 9:195–207. [DOI] [PubMed] [Google Scholar]
  28. Fischl B, Sereno MI, Tootell RB, Dale AM. 1999b. High-resolution intersubject averaging and a coordinate system for the cortical surface. Hum Brain Mapp. 8:272–284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Fonov V, Evans AC, Botteron K, Almli CR, McKinstry RC, Collins DL, Brain Development Cooperative Group . 2011. Unbiased average age-appropriate atlases for pediatric studies. Neuroimage. 54:313–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Foundas AL, Eure KF, Luevano LF, Weinberger DR. 1998. MRI asymmetries of Broca's area: the pars triangularis and pars opercularis. Brain Lang. 64:282–296. [DOI] [PubMed] [Google Scholar]
  31. Hayes TL, Lewis DA. 1995. Anatomical specialization of the anterior motor speech area: hemispheric differences in magnopyramidal neurons. Brain Lang. 49:289–308. [DOI] [PubMed] [Google Scholar]
  32. Heim S, Eickhoff SB, Amunts K. 2008. Specialisation in Broca's region for semantic, phonological, and syntactic fluency? Neuroimage. 40:1362–1368. [DOI] [PubMed] [Google Scholar]
  33. Huntgeburth SC, Petrides M. 2012. Morphological patterns of the collateral sulcus in the human brain. Eur J Neurosci. 35:1295–1311. [DOI] [PubMed] [Google Scholar]
  34. Huntgeburth SC, Petrides M. 2016. Three-dimensional probability maps of the rhinal and the collateral sulci in the human brain. Brain Struct Funct. 221:4235–4255. [DOI] [PubMed] [Google Scholar]
  35. Keller SS, Crow T, Foundas A, Amunts K, Roberts N. 2009. Broca’s area: nomenclature, anatomy, typology and asymmetry. Brain Lang. 109:29–48. [DOI] [PubMed] [Google Scholar]
  36. Keller SS, Highley JR, Garcia-Finana M, Sluming V, Rezaie R, Roberts N. 2007. Sulcal variability, stereological measurement and asymmetry of Broca's area on MR images. J Anat. 211:534–555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Knaus TA, Corey DM, Bollich AM, Lemen LC, Foundas AL. 2007. Anatomical asymmetries of anterior perisylvian speech-language regions. Cortex. 43:499–510. [DOI] [PubMed] [Google Scholar]
  38. Kostopoulos P, Petrides M. 2003. The mid-ventrolateral prefrontal cortex: insights into its role in memory retrieval. Eur J Neurosci. 17:1489–1497. [DOI] [PubMed] [Google Scholar]
  39. Kostopoulos P, Petrides M. 2008. Left mid-ventrolateral prefrontal cortex: underlying principles of function. Eur J Neurosci. 27:1037–1049. [DOI] [PubMed] [Google Scholar]
  40. Kostopoulos P, Petrides M. 2016. Selective memory retrieval of auditory what and auditory where involves the ventrolateral prefrontal cortex. Proc Natl Acad Sci USA. 113:1919–1924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Manjón JV, Coupé P, Martí-Bonmatí L, Collins DL, Robles M. 2010. Adaptive non-local means denoising of MR images with spatially varying noise levels. J Magn Reson Imaging. 31:192–203. [DOI] [PubMed] [Google Scholar]
  42. Mazziotta JC, Toga AW, Evans AC, Fox PT, Lancaster JL. 1995a. A probabilistic atlas of the human brain: theory and rationale for its development: The International Consortium for Brain Mapping (ICBM). Neuroimage. 2:89–101. [DOI] [PubMed] [Google Scholar]
  43. Mazziotta JC, Toga AW, Evans AC, Fox PT, Lancaster JL. 1995b. Digital brain atlases. Trends Neurosci. 18:210–211. [DOI] [PubMed] [Google Scholar]
  44. Mazziotta JC, Toga AW, Evans AC, Fox PT, Lancaster JL, Zilles K, Woods R, Paus T, Simpson G, Pike B. 2001. A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philos Trans R Soc Lond, Ser B. 356:1293–1322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Mohr J. 1976. Broca's area and Broca's aphasia. Stud Neurolinguis. 1:201–236. [Google Scholar]
  46. Mohr J, Caplan LR, Melski JW, Goldstein RJ, Duncan GW, Kistler J, Pessin MS, Bleich HL. 1978. The Harvard Cooperative Stroke Registry: a prospective registry. Neurology. 28:754–762. [DOI] [PubMed] [Google Scholar]
  47. Novek J, Sprung-Much T, Nolan E, Frey S, Petrides M. 2019. Optimal cutting of the cortex for cytoarchitectonic visualization: a neuronavigation-based approach. Rome, Italy: Organization for Human Brain Mapping. [Google Scholar]
  48. Ono M, Kubik S, Abernathey CD. 1990. Atlas of the Cerebral Sulci. Stuttgart: Thieme. [Google Scholar]
  49. Penfield W, Rasmussen T. 1950. The Cerebral Cortex of Man: a Clinical Study of Localization of Function. New York: Macmillan. [Google Scholar]
  50. Penfield W, Roberts L. 1959. Speech and Brain Mechanisms. New Jersey: Princeton University Press. [Google Scholar]
  51. Petrides M. 1996. Specialized systems for the processing of mnemonic information within the primate frontal cortex. Philos Trans R Soc Lond, Ser B. 351:1455–1462. [DOI] [PubMed] [Google Scholar]
  52. Petrides M. 2006. Broca’s area in the human and the non-human primate brain In: Grodzinsky Y, Amunts K, editors. Broca’s Region. Oxford: Oxford University Press, pp. 31–48. [Google Scholar]
  53. Petrides M. 2012. The Human Cerebral Cortex: an MRI Atlas of the Sulci and Gyri in MNI Stereotaxic Space. Chicago: Academic Press. [Google Scholar]
  54. Petrides M. 2014. Neuroanatomy of Language Regions of the Human Brain. Chicago: Academic Press. [Google Scholar]
  55. Petrides M. 2016. The ventrolateral frontal region In: Hickok G, Small SL, editors. Neurobiology of Language. London: Academic Press, pp. 25–33. [Google Scholar]
  56. Petrides M. 2019. Atlas of the Morphology of the Human Cerebral Cortex on the Average MNI Brain. New York: Academic Press. [Google Scholar]
  57. Petrides M, Alivisatos B, Evans AC. 1995. Functional activation of the human ventrolateral frontal cortex during mnemonic retrieval of verbal information. Proc Natl Acad Sci USA. 92:5803–5807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Petrides M, Pandya DN. 1984. Projections to the frontal cortex from the posterior parietal region in the rhesus monkey. J Comp Neurol. 228:105–116. [DOI] [PubMed] [Google Scholar]
  59. Petrides M, Pandya DN. 1988. Association fiber pathways to the frontal cortex from the superior temporal region in the rhesus monkey. J Comp Neurol. 273:52–66. [DOI] [PubMed] [Google Scholar]
  60. Petrides M, Pandya DN. 1994. Comparative cytoarchitectonic analysis of the human and the macaque frontal cortex In: Boller F, Grafman J, editors. Handbook of Neuropsychology. Amsterdam: Elsevier, pp. 17–58. [Google Scholar]
  61. Petrides M, Pandya DN. 2002. Comparative cytoarchitectonic analysis of the human and the macaque ventrolateral prefrontal cortex and corticocortical connection patterns in the monkey. Eur J Neurosci. 16:291–310. [DOI] [PubMed] [Google Scholar]
  62. Petrides M, Pandya DN. 2009. Distinct parietal and temporal pathways to the homologues of Broca's area in the monkey. PLoS Biol. 7:e1000170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Powell JL, Kemp GJ, Roberts N, García-Finaña M. 2012. Sulcal morphology and volume of Broca’s area linked to handedness and sex. Brain Lang. 121:206–218. [DOI] [PubMed] [Google Scholar]
  64. Price CJ. 2010. The anatomy of language: a review of 100 fMRI studies published in 2009. Ann N Y Acad Sci. 1191:62–88. [DOI] [PubMed] [Google Scholar]
  65. Sarkissov S, Filimonoff I, Kononowa E, Preobraschenskaja I, Kukuew L. 1955. Atlas of the Cytoarchitectonics of the Human Cerebral Cortex. Moscow: Medgiz. [Google Scholar]
  66. Saur D, Kreher BW, Schnell S, Kümmerer D, Kellmeyer P, Vry M-S, Umarova R, Musso M, Glauche V, Abel S, Huber W, Rijntjes M, Hennig J, Weiller C. 2008. Ventral and dorsal pathways for language. Proc Natl Acad Sci USA. 105:18035–18040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Segal E, Petrides M. 2012. The morphology and variability of the caudal rami of the superior temporal sulcus. Eur J Neurosci. 36:2035–2053. [DOI] [PubMed] [Google Scholar]
  68. Segal E, Petrides M. 2013. Functional activation during reading in relation to the sulci of the angular gyrus region. Eur J Neurosci. 38:2793–2801. [DOI] [PubMed] [Google Scholar]
  69. Shellshear JL. 1937. The brain of the aboriginal Australian: a study in cerebral morphology. Philos Tran R Soc Lond, Ser B, Biol Sci. 227:293–409. [Google Scholar]
  70. Sled JG, Zijdenbos AP, Evans AC. 1998. A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging. 17:87–97. [DOI] [PubMed] [Google Scholar]
  71. Smith GE. 1907. A new topographical survey of the human cerebral cortex, being an account of the distribution of the anatomically distinct cortical areas and their relationship to the cerebral sulci. J Anat Physiol. 41:237–254. [PMC free article] [PubMed] [Google Scholar]
  72. Sprung-Much T, Petrides M. 2018. Morphological patterns and spatial probability maps of two defining sulci of the posterior ventrolateral frontal cortex of the human brain: the sulcus diagonalis and the anterior ascending ramus of the lateral fissure. Brain Struct Funct. 223:4125–4152. [DOI] [PubMed] [Google Scholar]
  73. Talairach J, Tournoux P. 1988. Co-planar Stereotaxic Atlas of the Human Brain. New York: Thieme. [Google Scholar]
  74. Tomaiuolo F, MacDonald J, Caramanos Z, Posner G, Chiavaras M, Evans AC, Petrides M. 1999. Morphology, morphometry and probability mapping of the pars opercularis of the inferior frontal gyrus: an in vivo MRI analysis. Eur J Neurosci. 11:3033–3046. [DOI] [PubMed] [Google Scholar]
  75. Tremblay P, Dick AS. 2016. Broca and Wernicke are dead, or moving past the classic model of language neurobiology. Brain Lang. 162:60–71. [DOI] [PubMed] [Google Scholar]
  76. Uylings HB, Jacobsen AM, Zilles K, Amunts K. 2006. Left-right asymmetry in volume and number of neurons in adult Broca's area. Cortex. 42:652–658. [DOI] [PubMed] [Google Scholar]
  77. Wada JA, Clarke R, Hamm A. 1975. Cerebral hemispheric asymmetry in humans: cortical speech zones in 100 adult and 100 infant brains. Arch Neurol. 32:239–246. [DOI] [PubMed] [Google Scholar]
  78. Walker AE. 1940. A cytoarchitectural study of the prefrontal area of the macaque monkey. J Comp Neurol. 73:59–86. [Google Scholar]
  79. Wernicke C. 1874. Der aphasische Symptomencomplex. Eine psychologische Studie auf anatomischer Basis. Breslau: M. Cohn and Weigert. [Google Scholar]
  80. Zilles K, Schleicher A, Palomero-Gallagher N, Amunts K. 2002. Quantitative analysis of cyto-and receptor architecture of the human brain In: Toga AW, Mazziotta JC, editors. Brain Mapping: The Methods. 2nd ed. San Diego: Elsevier, pp. 573–602. [Google Scholar]
  81. Zlatkina V, Petrides M. 2010. Morphological patterns of the postcentral sulcus in the human brain. J Comp Neurol. 518:3701–3724. [DOI] [PubMed] [Google Scholar]

Articles from Cerebral Cortex (New York, NY) are provided here courtesy of Oxford University Press

RESOURCES