Neural Structures and Mechanisms for Speech, Language, and Hearing


Neural Structures and Mechanisms for Speech, Language, and Hearing


This chapter introduces the role of the central and peripheral nervous systems in speech, language, and hearing. Although other chapters in this text are focused on the speech apparatus (Chapters 2–6), the speech signal it produces (Chapters 7–11), and the perception of that signal (Chapter 12), the perspective is broadened in the current chapter to include selected aspects of language as well as speech production and perception. This recognizes the role of the nervous system in all aspects of normal and disordered communication and emphasizes the likelihood of overlap and interactions among speech, language, and hearing functions of the brain. Reference is made to early studies of the brain and human communication. These were primarily concerned with aphasia, the loss of language performance and possibly competence resulting from damage to specific brain locations.

The chapter begins with an overview of major concepts. Next, gross neuroanatomy is presented for the cerebral hemispheres and cerebral white matter, subcortical nuclei (e.g, basal ganglia and thalamus), cerebellum, brainstem and cranial nerves, cortical innervation patterns, and the spinal cord and its peripheral nerves. Cells within the nervous system and selected aspects of their function are discussed, followed by descriptions of meninges, ventricles, and the blood supply to the brain. Aspects of neurophysiology are interwoven with neuroanatomy throughout the chapter, along with selected clinical implications for the speech-language pathologist and audiologist. The chapter concludes with a well-known model of speech production that shows how known anatomy and physiology of the brain can be translated to the understanding of communication disorders.


In this chapter, the term “brain structures” refers to the anatomy of the central nervous system. Coverage of nervous system anatomy is focused primarily on gross anatomy—structures easily observable when handling and dissecting a whole brain, or when viewed using modern imaging techniques. More limited information is provided on cellular and even molecular anatomical levels, to provide a foundation for understanding neurological diseases affecting these fine-structure components of the nervous system. The term “brain mechanisms” refers to the physiology of the brain, at the molecular, cellular, neurochemical, and system levels. “System” levels of brain function are presumably the ones associated with observable behaviors, such as producing a sentence or providing behavioral evidence for understanding spoken language (as in following an instruction).

The major concepts provide a framework and set of reference terms that can be consulted throughout the chapter. They include central versus peripheral nervous system, autonomic nervous system, anatomical planes and directions, white versus gray matter, tracts versus nuclei, nerves versus ganglia, efferent and afferent, and lateralization and specialization of function.

Central Versus Peripheral Nervous System

The concepts of central nervous system (CNS) and peripheral nervous system (PNS) are familiar to most readers of this textbook. The CNS includes the cerebral hemispheres and its contents, the cerebellum, and the brainstem and spinal cord. The PNS includes the nerves issued from the brainstem and spinal cord, plus clusters of sensory nerve cells, called ganglia, located in close proximity to, but outside, the brainstem and spinal cord. In Figure 15–1, the major components of the CNS are labeled, as are nerves of the PNS, extending from the brainstem and spinal cord. Ganglia, located close to the brainstem and spinal cord, are not shown here but are discussed below.

Later in this chapter the structure of nerve cells is presented in some detail. For the current discussion of CNS and PNS, nerve cells (neurons)—consist of a cell body and an axon. The axon conducts electrical impulses away from the cell body to its endpoint, where the electrical energy is converted into chemical energy. Neurons not only conduct electrical impulses that are converted to chemical energy, but also control contraction of voluntary (striated) and involuntary (smooth) muscle, as well as secretions of glands. Bundles of axons are found in abundance in both the CNS and PNS. A bundle of axons in the CNS is called a tract; a bundle of axons in the PNS is called a nerve.

Autonomic Nervous System

The autonomic nervous system (ANS) controls involuntary actions, sometimes called subconscious behaviors. Examples of these behaviors include control of heartbeat, vegetative breathing, taste, and contraction of smooth muscle of the heart, digestive system, and eye (for control of pupil size), as well as other systems. Secretions of glands are also controlled by the ANS. Selected aspects of ANS function are mentioned in the section on cranial nerves.

Anatomical Planes and Directions

The terminology for planes and directions varies depending on the part of the CNS under discussion. The left part of Figure 15–2 shows an artist’s rendition of three anatomical planes as they are applied to the cerebral hemispheres. The coronal plane cuts the cerebral hemispheres into front and back sections, the sagittal plane into left and right sections, and the horizontal plane (also called axial or transverse) into upper and lower sections. Each of these planes can be moved along an axis perpendicular to the plane to cut through the cerebral hemispheres at different locations. For example, the horizontal plane can be moved up or down to obtain higher or lower cuts through the hemispheres. Similarly, the sagittal plane can be moved left or right, away from the midline (called the midsagittal plane), which divides the brain into equal left and right halves. Sagittal cuts away from the midline are referred to as parasagittal planes.

Several structures within the cerebral hemispheres have complex, curved shapes. Some structures are buried within the cerebral hemispheres and are difficult to visualize without multiplane views. Views of the brain in all three of the “standard” planes—coronal, sagittal, horizontal—are necessary to appreciate the form and location of these curved structures. A good example of the varying appearance of a single structure, the corpus callosum, is shown on the right side of Figure 15–2. Coronal (top), midsagittal (middle), and horizontal (bottom) slices through the structure are shown in magnetic resonance images (MRI). The corpus callosum is a massive bundle of fibers (a tract) linking structures in the left and right cerebral hemispheres. The top image is a coronal slice, roughly at the midway point between the front and back of the cerebral hemispheres. The red arrow points to a gently concave, white band of tissue located above two black “horns.” This coronal plane image intersects the corpus callosum at a single location, where the structure extends laterally from the midline into the two hemispheres (the arrow is just off the midline, in the right hemisphere, which appears on the left from the reader’s view). The appearance of the corpus callosum in the coronal plane varies depending on where the slice is located along the front-to-back extent of the hemispheres. This is better appreciated by examining the midsagittal plane image (see Figure 15–2, middle). The white band of callosal (“callosal” = of the corpus callosum) tissue extends along a good part of the front-to-back length of the brain, and has a flattened, archlike shape with relatively complex form at its front and back ends. The coronal slice shown above the sagittal slice was taken just in front of the red arrow on the sagittal slice; this slice interrupts the corpus callosum in the middle of the flattened arch. The bottom image in Figure 15–2 shows a horizontal slice in which the white bands of corpus callosum tissue are indicated by two red arrows, one toward the front (lower part of image) and one toward the back (upper part of image) of the hemispheres. Between these two regions of corpus callosum tissue, other callosal fibers are not apparent. This is because the horizontal slice is placed below the highest point of the arch seen in the midsagittal slice, which means it does not intersect this part of the corpus callosum. In fact, the horizontal slice cuts through two parts of the corpus callosum, as indicated by the two arrows in Figure 15–2.

Locations of brain structures are often identified relative to locations of other brain structures. Figure 15–3 shows an MR image in the coronal plane, slightly posterior to the halfway point between the anterior and posterior “poles” (the anterior and posterior tips of the cerebral hemispheres, seen in a midsagittal slice) of the cerebral hemispheres. Along the right-hand side of the image is a vertical dimension labeled dorsal at the top of the section, and ventral at the bottom; across the left-to-right extent of the image is a dimension labeled lateral (left), medial (center), and lateral (right). Within the cerebral hemispheres, dorsal means toward the top and ventral toward the bottom; therefore, the dorsal surface of the brain is the top surface and the ventral surface the underside of the brain. The side-to-side dimension within the cerebral hemispheres uses the term medial for the midline of the coronal view, and lateral for the sides of the brain, away from the midline.

The location of one structure relative to another is expressed by combinations of the ventral-dorsal and medial-lateral terms. In Figure 15–3, the cortex is the outermost tissue layer of the cerebral hemispheres, shown as a dark “rind” of tissue around the edge of the coronal image. Inside the cortical layer there is a good deal of white tissue, as well as several clusters of grayish tissue. Three of these grayish masses are labeled here: the putamen, caudate, and thalamus. In addition, the insula, a “hidden” part of the cortex beneath the lateral surface of either cerebral hemisphere, is labeled in the image (note how the insula is covered by two folds of the outer rind of cortex).

These four labeled structures are described more specifically below. They are included here to show how the terms dorsal, ventral, medial, and lateral are used to locate one structure relative to another within the cerebral hemispheres. For example, in the coronal section shown in Figure 15–3, the insula is lateral to the putamen, and the caudate is dorsal to both the putamen and the thalamus. The position of the caudate relative to the thalamus and putamen, however, is somewhat more complex than simply “dorsal.” The caudate is dorsal to the putamen, but also medial to it (closer to the mid-line), so a more precise statement is that the caudate is dorsomedial to the putamen. Similarly, the caudate is not only dorsal to the thalamus, but somewhat lateral as well. The caudate is therefore dorsolateral to the thalamus (or, reversing the reference structure, the thalamus is ventromedial to the caudate). These combinations can also include the terms anterior and posterior, as in one structure being anterolateral to another (see horizontal slice in Figure 15–4). This directional terminology is important because of its frequent use in descriptions of neuroanatomical structures, both in basic anatomical study and in imaging studies performed for diagnostic purposes.

Figure 15–4 shows a horizontal MR image on the left, and a sagittal image on the right that includes the brainstem. Toward the top of the horizontal image—in this case, the front of the cerebral hemispheres—is the anterior direction, toward the bottom the posterior direction. The anterior-posterior dimension is also shown in the sagittal image on the right of Figure 15–4; here the directions are obvious because of the image of facial features.

The sagittal image, which includes the lateral aspect of one cerebral hemisphere and the brainstem, introduces an interesting change in the use of direction terms. In the cerebral hemispheres there is a distinction between the dorsal-ventral and anterior-posterior dimensions. For neuroanatomical structures from the top of the brainstem and below, however, the dorsalventral and posterior-anterior dimensions are one and the same. The top of the brainstem is indicated in the sagittal image by a horizontal dashed line. Below this line, dorsal and posterior indicate the same direction—toward the back of the body—and ventral and anterior indicate toward the front of the body. The identity between the terms dorsal/posterior and ventral/anterior applies to spinal cord structures as well.

The terms deep and superficial are also used to locate one structure relative to another. These terms are typically used to designate the relative locations of structures on a path from “outside to inside” (or the reverse). The question, “What is beneath this surface?” is equivalent to asking, “What is deep to this surface?” For example, the white matter of the cerebral hemispheres is deep to the cortex (or the cortex is superficial to the white matter).

White and Gray Matter, Tracts and Nuclei, Nerves and Ganglia

When a brain is removed from the skull with the intention of preserving it for later study, it is “fixed” in a solution of formalin, a fluid that hardens biological tissue. A formalin-preserved brain shows certain regions having a grayish-brown appearance, and other regions with a pale, near-white appearance. Similarly, in conventional MR images of the brain, some regions appear grayish, and some appear whitish. The coronal plane MR images in Figures 15–2 through 15–4 show both gray and white matter. Gray matter consists of clusters of neuron cell bodies (somata, the plural of soma, which means cell body). White matter is formed from myelinated axons issued by those cell bodies. Bundles of axons are called tracts in the CNS, and nerves in the PNS.

Gray Matter and Nuclei

The cortex, the outermost thick covering of the cerebral hemispheres, consists of densely packed cell bodies; the cortex is a major part of gray matter in the cerebral hemispheres. Figure 15–3 shows the cortical “rind” of gray matter enclosing extensive white matter. Within this white matter are several regions of additional gray matter (e.g., the structures labeled “caudate,” “putamen,” and “thalamus”). A specific cluster of cell bodies inside the cerebral hemispheres, or within the brainstem or spinal cord, is referred to in the singular as a nucleus (as in “caudate nucleus”) or in some cases as a group of nuclei (as in “cranial nerve nuclei,” which are clusters of cell bodies within the brainstem). The clusters of cell bodies deep to the cortex but within the cerebral hemispheres, such as the caudate, putamen, and thalamus, are subcortical nuclei. Some subcortical nuclei, such as the thalamus, are collections of many smaller nuclei but are often referred to jointly as a single structure. The term subcortical nuclei does not include nuclei within the brainstem, cerebellum, and spinal cord, but is reserved for clusters of cell bodies within the cerebral hemispheres. Nuclei are also found in the cerebellum; these are referred to as cerebellar nuclei.1

Neuronal cell bodies, whether in the cortex, sub-cortical region, cerebellum, brainstem, or spinal cord, most often cluster together for a common purpose; they do not aggregate randomly. In a specific region of cortex, for example, cell bodies related to eye movements, or to auditory perception, are likely to cluster together. A region of cortex, or within a subcortical or brainstem nucleus in which cells have a common function, may have a more fine-grained, systematic grouping of cells. For example, the most posterior gyrus (where gyrus = a hill of tissue on the cortical surface, separated from other gyri by deep fissures, or sulci) of the frontal lobe is the primary motor cortex, containing cells that have direct control over the timing, force, and duration of muscle contractions. Within this primary motor cortex, cells associated with particular parts of the body aggregate together. As one example, hand and finger cells are found in close proximity in the primary motor cortex. The cellular representation according to body parts within a specific cortical area is called somatotopic organization. Somatotopic organization is a maplike projection of the body plan onto the cortex as well as on subcortical nuclei, the cerebellum, the brainstem, and even in the organization of fiber tracts. Somatotopy is characteristic of both motor and sensory systems in the CNS (see below, Figure 15–6).

Somatotopic Representation Is Not Always “Clean”

Many studies have demonstrated the “truth” of somatotopic representation in primary motor and sensory cortices. These studies include correlations between highly localized lesions and affected body parts, electrical stimulation of very discrete locations along the cortex and observation of the resulting movements or reported sensations, and functional MRI (fMRI) studies in which very local movements, as in raising of a single finger, result in the “lighting up” of a small region of the primary motor cortex. Somatotopy, however, is not always neat and clean. For example, fMRI studies of representation of speech apparatus structures in the primary motor cortex have revealed that cells for muscles of the pharynx, tongue, and lips “share” space close to the bottom of the central fissure, where the primary motor cortex meets the sylvian fissure (Takai, Brown, & Liotti, 2010). These cortical cells are not perfectly segregated by structure as might be expected by a strict somatotopy. Takai and colleagues call the mixing of cells for these three speech apparatus structures “somatotopy with overlap.” As argued by these authors, the overlap makes functional sense for the control of coordinated behaviors such as swallowing and speaking.

The concept of somatotopic representation is important not only for understanding brain anatomy, but also for gaining clinical perspectives on the effects of lesion location on function. For example, somatotopic organization explains (in part) how a stroke can affect a client’s ability to walk but leave speech and language unaffected; or how a stroke can affect speech and language in the absence of other obvious problems.

White Matter and Fiber Tracts

Neurons have a cell body, an axon, and a terminal (end) structure. The primary function of a neuron is to conduct electrical impulses from the cell body along the axon to the terminal structure, also called the terminal segment or button. As discussed in more detail below, the majority of axons in the brain are wrapped in a fatty substance called myelin. Myelin insulates axons and in so doing makes neuronal conduction of electrical impulses faster and more efficient. Myelin gives axons the whitish appearance in fixed brains and brain images. The coronal section shown in Figure 15–3 shows a good deal of white matter, and therefore many, many myelinated axons.

Axons connect different areas of gray matter. These connections are referred to as pathways, fiber tracts, fiber bundles, fasciculi (singular = fasciculus), and lemnisci (singular = lemniscus). Bundles of axons connect different cortical regions to one another, cortical cells to subcortical and brainstem nuclei, spinal nuclei to brainstem and subcortical nuclei, cerebellar nuclei to many different nuclei in the brain—in short, myelinated pathways are everywhere in the brain, connecting and interconnecting different masses of gray matter. These pathways are typically organized somatotopically, in much the same way as the areas of gray matter they connect.


The term “ganglia” (singular = ganglion) is almost exclusively reserved for clusters of nerve cell bodies located outside the CNS. Ganglia are technically part of the PNS. They receive sensory fibers coming from receptors in the body (such as tactile receptors, or from receptors in the cochlea or retina) where a first synapse (connection with another neuron) is made prior to entry of the information into the CNS. For example, tactile receptors in the hand are special end organs of the PNS, embedded in skin or muscle and connected by a sensory fiber to the CNS. When the end organ is stimulated (e.g., by the compression of touch), it “fires,” sending electrical impulses to the nervous system via the sensory fibers. These fibers make first contact—that is, make first synapses—with cells in a ganglion immediately outside the spinal cord. The ganglion cells receiving this information deliver it, via its own axons, to sensory cells in the spinal cord. Similar receptors are found in muscles of the speech apparatus, including muscles of the respiratory system, larynx, and upper airway (pharyngeal, oral, and nasal passageways).

The spinal cord is associated with a series of dorsal root ganglia, arranged from the top to bottom segments of the cord. These ganglia serve as first synapses for much of the sensory information delivered from the limbs and trunk to the CNS. There are also ganglia immediately outside the brainstem that serve as the first synapses for sensory information from head and neck structures. For example, the hair cells of the cochlea are the specialized sensory endings for hearing, and when they are deformed by motion of the basilar membrane they fire, sending impulses to the spiral ganglion, a bundle of nerve cells outside the brainstem. First synapses for auditory neurons are made in the spiral ganglion cells, which send axons into the brainstem to make synapses within the first set of nuclei along the auditory pathway (see Chapter 13).

Efferent and Afferent

The terms efferent and afferent are used in two different ways to describe information flow in the nervous system. The first use of the terms concerns overall information flow for production of muscular effort (efferent) in contrast to the information flow for the sensation of an environmental event (afferent). When motor commands are issued from cortical tissue and travel along descending pathways through one or more motor nuclei before eventually being directed by peripheral nerves to muscles, the pathways are referred to as efferent. Sometimes the term efference is equated with the basic components—brain-directed muscle contractions and their patterns in space and time—of motor control. In contrast, when a sensory receptor (such as a tactile receptor in a finger, or the hair cells of the cochlea) is stimulated and the resulting signal travels via peripheral nerves to the spinal cord or brainstem and then through ascending pathways to a final destination in the cortex, the pathways are said to be afferent. Afference is often taken to mean “sensory.”

A second use of the terms efferent and afferent indicates the inputs and outputs of nuclei within the CNS. This usage does not imply a motor act or sensation to or from a body structure, but rather the flow of information from one nucleus to another, or from nuclei to cortex or cortex to nuclei. For example, the substantia nigra (SN) is a nucleus in the midbrain that sends a neurotransmitter called dopamine to several subcortical and brainstem cells. The cells of the SN send information via a fiber tract to the striatum, a subcortical nucleus group made up of the caudate and putamen. The output of the SN to the striatum is therefore referred to as an efferent projection of the SN. On the other hand, the SN receives efferent projections from the subthalamic nucleus (STN), another subcortical nucleus. These inputs to the SN from the STN are afferent projections. The SN, and many other cells in both subcortical nuclei and the cortex, issue efferent projections to other nuclei (or cortical cells) and receive afferent input from other cell groups.

For the present purposes, the specific nuclei sending information to, or receiving information from, other nuclei are not important. Rather, the concept of efferent and afferent projections within the CNS is crucial. Note the importance of a reference structure in naming the efferent and afferent projections of a nucleus: a nucleus’ efferent projections are pathways sent to another nucleus (or the cortex), whereas the nucleus’ afferent projections are pathways received from other nuclei (or the cortex). A single nucleus, such as the SN or STN, typically has both efferent and afferent projections—the nucleus influences other cell bodies, and is influenced by other cell bodies. Sometimes these influences are in a loop, so two nuclei may communicate with each other via both efferent and afferent projections.

Most nuclei, and cortical cells as well, receive and send information to multiple locations within the brain. Each cortical region or nucleus is likely to have multiple efferent and afferent projections. The dense and overlapping interconnections within the brain, including the efferent projections of a single nucleus (or cortical area) to many different nuclei (or areas), plus the variety of afferent inputs to a single nucleus (or cortical area), make the simple distinction of efferent versus afferent less useful than once thought. For example, motor systems are not isolated from, or independent of, sensory systems. In fact, motor and sensory systems are highly interdependent, with certain brain “circuits”—interconnected nuclei—performing sensorimotor integration for the most efficient and skilled actions. Sensorimotor control is often the preferred term to represent action such as the coordinated motions of the articulators (pharyngeal-oral and velopharyngeal-nasal structures), larynx, and breathing system to produce the acoustic signal we call speech.

Neurons and Synapses

Neurons are the “firing” cells of the CNS and PNS. Under the right circumstances, neurons generate electrical potentials that allow them to cause other neurons to “fire.” A synapse includes the anatomical structures serving the transfer of information from one neuron to another. The transfer information across neurons is accomplished by the conversion of electrical energy to neurochemical energy, which in turn is converted back to electrical energy.

Lateralization and Specialization of Function

“Lateralization” and “specialization” may sometimes be used interchangeably, but the terms have different technical meanings. When a function is said to be “lateralized” in the brain, the function is primarily controlled by one hemisphere relative to the other. Lateralization is also referred to as “hemispheric asymmetry” for function and can be applied strictly to anatomical structure (as in the case of similar locations in the two hemispheres having different volumes of neural tissue). Good examples of lateralized functions include (of course) speech and language (thought to be controlled by left hemisphere structures in about 95% of the population), handedness (also controlled by the left hemisphere in about 95% of the population), and emotions (thought to be controlled primarily by right hemisphere structures). Lateralization of speech and language function to the left hemisphere has been demonstrated by clinical cases and research studies, including the common observation of disrupted speech and language function when a stroke affects the blood supply to the left hemisphere but not the right hemisphere. Lateralization of speech and language has also been shown by the ability to elicit speech and language behaviors during surgical procedures, when an electrical current stimulates regions of the left hemisphere’s cortical surface. Stimulation in the same cortical regions of the right hemisphere does not typically elicit these speech and language behaviors.

A test of brain lateralization of speech and language functions is to inject a drug, called amobarbital, into either the left or right carotid artery so that the left or right hemisphere is anesthetized for a brief period of time. The carotid arteries are the main source of blood flowing from the heart to the cerebral hemispheres. An injection of amobarbital into the left carotid artery initially distributes the drug to the left hemisphere, leaving the right hemisphere unaffected for a short time. The drug temporarily blocks neural activity in the anesthetized cerebral hemisphere, and therefore provides a technique to determine which hemisphere is implicated for speech and language function. If injection into the left carotid artery results in speech and language deficits, but injection into the right carotid artery does not, this is good evidence for lateralization of speech and language functions to the left hemisphere.

The amobarbital test, called the Wada test after the Japanese-Canadian neurologist Juhn Wada, who developed the procedure in the late 1940s, is used in patients undergoing resection (extraction) of parts of the brain to relieve chronic, severe epilepsy. Before the Wada test, clinical cases of stroke affecting either the left or the right cerebral hemisphere led physicians and scientists to regard speech and language functions of the brain as lateralized to the left hemisphere. This suggested that the same lateralization of speech and language functions was typical of neurologically normal individuals. In epilepsy, however, there was (and is) the sense that a seizure-causing lesion in the left hemisphere may result in speech and language functions being “transferred” to the right hemisphere, or to be split more equally between the two hemispheres. The Wada test is used in these patients to identify the hemisphere to which speech and language function is lateralized, to indicate areas where surgical resection should be avoided or minimized for maximal preservation of communication ability.

Many studies using the Wada test have been reported in the clinical literature (for two good examples, both of which contain reviews of much of the pertinent literature, see Lee et al. [2008], and Springer et al. [1999]). Use of the Wada test in neurologically normal individuals to determine if laterality varies by handedness has been challenged by the difficulty of assembling a large enough group of left-handers to compare with more easily recruited right-handers. When the Wada test has been used to compare laterality of left- and right-handers, it is usually among presurgical patients with epilepsy. This work shows about 96% of right-handed people to have left hemisphere lateralization for speech and language, compared with 85% of left-handers (or “mixed” handers; see Rasmussen & Milner, 1977). When contemporary brain imaging methods are used to estimate lateralization of speech and language, estimates for right-handers almost match the data from Rasmussen and Milner, but more left-handers show lateralization of speech and language to the right hemisphere (Swanson, Sabsevitz, Hammeke, & Binder, 2007). Using an imaging technique similar to fMRI (see sidetrack on this page), Knecht et al. (2000) obtained results showing a relationship between degree of left-handedness and the likelihood of speech and language being lateralized to the right hemisphere. Hard-core righties were almost all left-lateralized (~96%), whereas only about 73% of hard-core “southpaws” were left-lateralized. The sample of people with various degrees of “mixed” handedness fell, as a group, between these two numbers.

This brief review of the literature on lateralization of speech and language function suggests at least two broad conclusions, as well as one additional thought concerning the concept of lateralization. First, most people, either left or right-handed, have speech and language lateralized to the left hemisphere. Second, there is a greater chance for language to be lateralized to the right hemisphere, or for its representation to be split between the hemispheres, in people who are clearly left-handed. Still, the majority of left-handers are “left-dominant” for speech and language, even though they are “right-dominant” for handedness. Finally, the concept of lateralization for speech and language to one hemisphere or the other is not an absolute, either-or concept. Lateralization for speech and language appears to be a continuous phenomenon, not only because of the demonstration of greater likelihood for right-hemisphere dominance with increasing left-handedness (Knecht et al., 2000), but also because certain aspects of speech and language—such as prosody—have been shown to be represented primarily in the right hemisphere. Even though associations exist between handedness and lateralization of speech and language function, the association is not necessarily due to a common underlying mechanism; handedness and lateralization of speech and language function are related only partly in a causal way (Ocklenburg, Beste, Arning, Peterburs, & Güntürkün, 2014). When the term “dominant hemisphere for speech and language” is used, these qualifying thoughts should be kept in mind.


MR images are obtained by placing a body structure within a strong magnetic field and then, by application and withdrawal of a second magnetic field, causing cell nuclei to generate magnetic properties that are sensed by a coil. The coil detects different amounts of energy generated by the cells, depending on cell properties, cell locations, and other factors. Software processes the signals generated by the cells and assembles them into an image of the target structure (for example, see right side of Figure 15–2). Functional MRI (fMRI) makes use of the same general principle, but with a twist (that’s a pun for those of you familiar with the magnetized behavior of brain cell nuclei). When neurons are active, the active region attracts blood flow from arteries; such arterial blood flow is oxygen rich, and oxygen-rich blood has different magnetic properties than run-of-the-mill blood. When an area of the brain is being used for a specific task, like speaking, that area generates a different signal than an area not being used for that task. The software is designed to make that active area “light up” because of increased blood flow. So, now you know that you can actually light up a room when you speak words of cheer!

Genes, Hands, and Brains

Scientists continue to be interested in the relationship between handedness and lateralization of speech and language function. In a recent study, Schmitz, Lor, Klose, Güntürkün, and Ocklenburg (2017) performed an analysis of the genetic overlap between handedness and lateralization of speech and language. In the past, handedness was thought to be determined by a single gene, but more recent research has demonstrated that it is more likely determined by a group of genes. Schmitz et al. found only a small overlap between sets of genes for handedness and sets for speech and language lateralization. In lay language, handedness and lateralization of speech and language do not develop solely as a result of common genetic mechanisms. Why is this important? In many studies of speech and language function, handedness is often used as a control variable to rule out possible messiness in the data due to different degrees of lateralization among participants. The work by Schmitz et al. suggests that restricting a participant sample to right-handers only may not guarantee the degree of left lateralization of speech and language skills.

How is “specialization” different from “lateralization”? Although it may seem reasonable to say that the left hemisphere is specialized for speech and language (in the same way it is lateralized for speech and language), the term “specialization” is used in a more specific—pun intended—sense. Specialization means that certain brain regions have evolved to serve distinct functions, whether lateralized or not. For example, a well-known claim for specialization is that portions of the frontal lobe are specialized for executive function (see Alvarez & Emory, 2006 for a review of evidence for and against this claim). “Executive function” is, broadly speaking, the ability to organize behavior to achieve a goal or set of goals and to connect current behavior with future outcomes. If the brain is a movie in production, executive function is the director of the production. Executive function coordinates brain function, matches behaviors to desired outcomes, and regulates actions. In neurologically intact individuals, tasks requiring organization of complex material to achieve a certain outcome often cause parts of the frontal lobes to “light up” in fMRI studies. People with damage to the frontal lobes are likely to show impaired ability to deal with complex decision making and demonstrate poor regulation of behavior. Some scientists and clinicians regard the frontal lobes of both hemispheres, or at least certain parts of them, to be specialized for executive function. Here we have a case of specialization without obvious lateralization.

Another example of hypothesized specialization is in regions of the brain thought to be critical for the representation and programming of articulatory gestures for speech. Some scientists place these two processes in a very small portion of the posterior, ventral edge of the left frontal lobe (more is said about this below). The actual neural tissue for the execution of speech sounds—where execution means transforming the represented and programmed sounds into movements—is thought to be located slightly posterior to this programming tissue, but still in the frontal lobe. This is an example of specialization within lateralization: speech is left-lateralized, and within this lateralized function there are finer degrees of tissue specialization for the production of speech and language.

A popular view of specialization, one that takes the concept to a logical (but in some views, extreme) conclusion, is found in the idea of brain modules. Modules are thought to be regions of brain tissue that are specialized for particular tasks—“dedicated,” in computer terminology—and in fact insulated from other tasks. Some scientists and clinicians believe, for example, that in humans a very specific region on the underside of the temporal lobe contains a module for human face recognition (matching a person’s identity with his or her face; see Said, Haxby, & Todorov, 2011). This brain region is thought to be “insulated” from other tasks because it is not active for recognition of nonhuman/nonface objects such as cars, dogs, and so forth. Closer to home, some speech scientists believe there is a module for speech perception in the left hemisphere, a species-specific (human) collection of neural tissue activated only by human speech or simulations of it (i.e., computer generated speech) (see Chapter 12).


The cerebral hemispheres are part of the CNS (see Figure 15–1). Each hemisphere contains four lobes, each lobe having gray and white matter.

Cerebral Hemispheres

Figure 15–5 shows the cerebral hemispheres in four views. The top-left view is from above the brain, looking down to the dorsal surfaces of the two hemispheres. The front of the brain is toward the top of the image. This view shows the left and right cerebral hemispheres, separated by a long, front-to-back fissure called the longitudinal fissure (also called the interhemispheric or sagittal fissure). The visible surface tissue is cortex, regarded as the most complex and “sophisticated” part of the brain. The cortex covers the entire surface of the cerebral hemispheres. Note the ridges or “hills” of the cortex, and the “dips” between them. The ridges are called gyri (singular = gyrus) and the “dips” sulci (singular = sulcus) or fissures (the term “fissure” is typically used to mean a particularly deep sulcus, such as the longitudinal fissure). One notable difference between the human brain and the brain of animals such as sheep, cats, and dogs is that humans have relatively deep and numerous sulci defining the cortical surface. These deep sulci are infoldings of cortical surface forming unseen “walls” of tissue that contribute to a greater volume of cortical cells in the human brain relative to other animals. The hidden cortical surface area and its corresponding thickness add to the cognitive and performance power of humans. By gently separating any sulcus on the surface of a prepared (formalin-hardened) brain, the walls of hidden cortical tissue are revealed. Some authors have estimated that close to two-thirds of the human cortex is hidden inside sulcal walls (Zilles, Armstrong, Schleicher, & Kretschmann, 1988). This unique feature of the human brain appears to be an evolutionary solution to packing lots of cortical tissue into a container—the skull—of limited size.

A dramatic view of “hidden” cortical tissue can be gained by putting your thumbs inside the longitudinal fissure of a prepared brain, your two hands resting on the two hemispheres, and gently separating the hemispheres without tearing the tissue. This exposes the deep inside (medial) walls of the two hemispheres. The medial wall of the right cerebral hemisphere is shown in the top-right view of Figure 15–5. A prominent feature of this midsagittal view is the corpus callosum, the massive bundle of tissue that connects structures across the two hemispheres.

Figure 15–5, bottom-right, shows a side view of the left hemisphere of the brain. The front of the brain is toward the left of the image. This view shows the four lobes of the brain, their boundary landmarks, plus additional regions important to speech, language, and hearing. The image shows the frontal (green), parietal (light brown), temporal (blue), and occipital (purple) lobes. This color code is modified at specific locations to highlight important cortical regions within the hemisphere. What follows is a more detailed consideration of each of these lobes. The structure and function of two additional cerebral regions, the insula and limbic system, are also discussed.

Frontal Lobe

The frontal lobe is bounded posteriorly by the central fissure (also called the fissure of Rolando), and below by the anterior part of the lateral sulcus or sylvian fissure (see Figure 15–5, top left and right for central fissure, bottom right for sylvian fissure). The gyrus immediately in front of the central sulcus, and therefore within the frontal lobe, is the primary motor cortex (shown as a lighter shade of green in Figure 15–5). The primary motor cortex contains cells called motor neurons that send signals to motor neurons in the brainstem and spinal cord, which, in turn, send axons to muscles to control their contraction patterns. The pathway: primary motor cortex neurons → brainstem/spinal cord motor neurons → muscles can be thought of as the route of direct nervous system control of the timing, strength, and speed of muscle contractions for head, neck, limb, and torso structures. As discussed later in this chapter, this direct route to muscle control is modulated and fine-tuned by activity in several different parts of the CNS, including other cortical regions (such as the supplementary motor area, primary somatosensory cortex, and Broca’s area, shown in Figure 15–5, top and bottom right), the basal ganglia and cerebellum to produce everyday movement as well as highly skilled, specialized movement.

Primary Motor Cortex. The somatotopic arrangement of cells (motor neurons) in the primary motor cortex is like an inverted map of the body. Figure 15–6 shows that the map extends from the top of the brain (at the longitudinal fissure), down the lateral surface of the hemisphere to the sylvian fissure. The map is found in both hemispheres. The map reflects not only which body parts are represented in a specific region of the primary motor cortex, but the amount of cortex devoted to the control of specific body parts, as shown in the bottom left image of Figure 15–6. A notable characteristic of the body map of the primary motor cortex is the upside-down representation of the body: cells that control muscles of the lower part of the body (such as muscles of the hip and knee) are at the top of the primary motor cortex (or even along the medial wall of the cortex—note the location of cells for muscles of the feet), whereas control of muscles of the face, tongue, and larynx are toward the bottom of the gyrus, just above the sylvian fissure. Larger representation of body parts on the map indicate a greater number of cells devoted to control of the body part. Note the very large size of the face and its associated structures (the tongue, lips, larynx) compared with the size of the feet or the trunk. A disproportionate number of cells in the primary motor cortex is associated with control of the structures of primary interest to speech-language pathologists. The auditory cortex, not represented here, has its own systematic plan for cells, arranged according to signal frequency (Chapter 13).

The disproportionate representation of cells that control movements of orofacial structures suggests their great relevance to the lives of humans. Not much imagination is required to make the case for the centrality of eating and breathing in human function, and the need for sophisticated muscular control to support these behaviors. The same case can be made for any mammal, but the disproportionate representation of these structures within the primary motor cortex of humans is very much a function of our unique ability to generate spoken language. Notice the phrase “spoken language”; it is not just the production of sounds ​—​many animals do this for simple communicative purposes​—but the extensive use of a signal system (the acoustic signal emerging from the vocal tract) to give meaning to abstract, complex ideas, to convey the same idea in different ways, even to create ideas (Deacon, 1997).

The location of orofacial and laryngeal cells at the “bottom” (ventral aspect) of the primary motor cortex is of interest because they are close to Broca’s area, a premotor cortical region located in the left hemisphere on the ventrolateral surface of the frontal lobe (Figure 15–5, bottom right). Broca’s area is immediately adjacent to the primary motor cortical representation of orofacial and laryngeal control. Surely this is not a coincidence.

Broca’s Area. The inferior frontal gyrus, on the lateral surface of the left frontal lobe and immediately above the anterior end of the sylvian fissure, is called Broca’s area. Broca’s area is shown in Figure 15–5, bottom right, as the frontal lobe region colored dark green. Broca’s area (or at least some of it), along with several other related areas of the frontal lobe, is premotor cortex, the latter term indicating a role in motor control different from the direct control of muscles associated with cells in the primary motor cortex. Broca’s area is believed to have a central role in the planning and organization of motor behavior required for speech production. This conclusion was first formulated by Paul Broca (1824–1880), the famous French physician who between 1861 and 1865 reported on a few patients with primarily expressive speech-language disorders resulting from neurological disease (usually a stroke). On autopsy, Broca noted lesions in and around the third frontal gyrus (the lower part of the inferior gyrus) of the left hemisphere. The patients examined by Broca had expressive speech problems but largely unaffected speech and language comprehension. Because the autopsy revealed damage in the lower, posterior frontal lobe, Broca concluded that this region of the brain was responsible for speech expression. The conclusion was reinforced by observations that lesions in the inferior gyrus of the right frontal lobe did not produce speech or language disorders. The emerging picture was of specialization for expressive control of speech and language in the left hemisphere, in the third frontal convolution of the frontal lobe. This brain region has since been known as “Broca’s area.”

We know now that historical and contemporary interpretations of Broca’s observations are too simplistic. First, damage to the brains of people with expressive disorders like those described by Broca (often called “Broca’s aphasia”) typically extends to regions well beyond the third frontal gyrus (Keller, Crow, Foundas, Amunts, & Roberts, 2009). In fact, the actual brains autopsied by Broca were examined again almost 20 years ago with imaging techniques and shown to have widespread damage in addition to the obvious lesion to the inferior convolution of the left frontal lobe (Dronkers, Plaisant, Ibas-Zisen, & Cabanis, 2007). Second, there is evidence that Broca’s area is involved in language comprehension, specifically of utterances with relatively complicated syntax (Grodzinsky & Santi, 2008) and semantics (Willems & Hagoort, 2009). Third, tissue in and around Broca’s area has been shown to respond to non-linguistic events, two of which are watching finger and/or mouth movements. Broca’s area “lights up” when participants observe these movements (Lindenberg, Fangerau, & Seitz 2007). Finally, for the great majority of people, speech and language control in Broca’s area is both lateralized to one hemisphere and specialized by virtue of a specific area of left hemisphere cortex devoted to speech and language function. There is asymmetry of function among most people, but a small number of people have lateralization of speech and language function to the right hemisphere, or no lateralization at all. Lesions to the right hemisphere may produce the same speech and language signs as lesions to the left hemisphere, at least in a small number of people. Broca’s area is the term used to designate a specialized region in the left hemisphere for speech expression, but in cases of right hemisphere lateralization for these functions, the relationship between anatomy and function is not so clear.

Broca’s area is clearly important for speech and language behavior, but is not devoted exclusively to speech expression. Instead, Broca’s area is part of a network involving extensive interconnections and shared functions, serving the complexity of all aspects of human communication. It is not productive to expect simple matches between damage to specific areas of the brain and specific functions.

Premotor and Supplementary Motor Area. The gyrus just forward of, and parallel to, the primary motor cortex is called premotor cortex (often called PMA, for “Pre-motor Area,” not labeled in Figure 15–5; see the darker green gyrus immediately anterior to the lighter green gyrus in Figure 15–5, bottom right). The ventral part of premotor cortex is Broca’s area, colored dark green in Figure 15–5, lower right image. In the lateral view of the left hemisphere, the dorsal limit of premotor cortex is the supplementary motor cortex (SMA) (Figure 15–5, lower right). The SMA also extends down the medial wall of the frontal lobe (Figure 15–5, upper right). PMA and SMA are thought be cortical areas in which action (motor acts) is planned; primary motor cortex is the cortical area from which commands are issued to perform movements. This distinction is important to the speech-language pathologist, who is often asked to make a diagnostic judgment of whether a speech motor control disorder is one of execution or planning (or both). Execution disorders of speech motor control are called dysarthrias; planning disorders are called apraxias or dyspraxias.

Prefrontal Cortex. The large mass of frontal lobe tissue anterior to the primary motor cortex and PMA/SMA is called prefrontal cortex. Many scientists (see, for example, Ridderinkhof, Ullsperger, Crone, & Nieuwenhuis, 2004) believe that this part of the frontal lobe, plus its ventral surface, performs executive function in the brain. Executive function is the guidance of all cognitive (and perhaps lower level) brain functions. Executive function is the brain’s monitoring, selecting, and “tuning” of higher-level actions and behavioral goals. Executive function guides decisions such as when it is “okay” to use certain words or drink certain beverages, why it is not okay to employ violence as a reaction to situations, whether or not a certain behavior may have a profound effect on your own or someone else’s life in the future, and how the tuning of your sensory systems (like hearing and vision) changes from situation to situation (among other things). This short list makes it is easy to understand why scientists have connected prefrontal cortex with aspects of personality. Lesions of the prefrontal cortex may have direct relevance to speech-language pathologists who work with patients with frontal lobe lesions and impaired executive function. Patients with traumatic brain injury, or with dementia, may show executive function problems due (in part) to frontal lobe lesions.

Parietal Lobe

The parietal lobe (shown in light brown in Figure 15–5 except for its most anterior gyrus, which is colored orange) is bounded at the front by the central fissure (or sulcus) or fissure of Rolando, below by the posterior part of the sylvian fissure (or lateral sulcus), and toward the back of the brain by the parieto-occipital fissure. The parieto-occipital fissure, which is the boundary between the parietal and occipital lobes, is easy to see on the medial wall of the cerebral hemisphere (see Figure 15–5, top right) but only partially visible when viewing the external surface of the hemisphere. The boundary shown between the parietal and occipital lobes in the bottom right view of Figure 15–5 is therefore approximate. Similarly, the boundary between the lower part of the parietal lobe and the back of the temporal lobe is not clearly marked in the bottom right view of Figure 15–5.

Primary Somatosensory Cortex. The gyrus immediately in back of the central (Rolandic) fissure—the most anterior gyrus of the parietal lobe—is the primary somatosensory cortex. The primary somatosensory cortex, colored orange in Figure 15–5, runs parallel to the primary motor cortex. Like the primary motor cortex, the primary somatosensory cortex is organized somatotopically, although not in precisely the same way as the motor cortex (see sidetrack on “Somatotopic Representation Is Not Always ‘Clean,’” and Figure 15–6).

Stated broadly, cells in the primary somatosensory cortex respond to touch and pain stimuli from all body locations. The primary somatosensory cortex is, in fact, a good deal more complicated than this broad view. There are extensive interconnections among cell types within the somatosensory cortex and subcortical nuclei, as well as brainstem nuclei and the cerebellum. Some cortical cells receive basic touch information from subcortical and brainstem nuclei, some use this basic information to encode the texture or shape of touched objects, and some may respond to the magnitude and direction of a tactile stimulus (Bear, Connors, & Paradiso, 2007). Cells in the primary somatotopic cortex receive information about pain and temperature, and some cells are interconnected with cells in primary motor cortex.

Posterior Parietal Cortex. The parietal cortex posterior to the primary somatosensory cortex and anterior to the occipital lobe as well as the portion sharing a boundary with the temporal lobe is called the posterior parietal cortex (PPC). The PPC contains cell groups that integrate and process different sensory stimuli to create complex sensory experiences. These cells are also involved in the planning of complex motor acts such as reaching, grasping, and tool use (Culham & Valyear, 2006). Recall that primary somatosensory cortex receives information on touch and pain; what of other sensations and combinations of sensations we experience? For example, the experience of taking care to cross a street when hearing an ambulance siren and then seeing the rapidly approaching vehicle involves an integration of (at least) auditory and visual sensations with the action plan of stepping backward to the curb or sprinting across the street. Discussion of auditory and visual cortex follows below, but here it can be stated that the “primary” information on auditory and visual stimuli arrives in the temporal and occipital lobes, respectively. This information is sent to the PPC where it is analyzed and integrated into increasingly more complex perceptual and action forms. In this sense, the PPC functions as association cortex, literally associating different types of sensory stimuli and directing action plans based on this integration.

Two examples of the integrative function of PPC are noteworthy. First, object recognition by the hand requires the ability to identify size, shape, texture, hardness/softness, and other characteristics. Activity in primary somatosensory cortex related to these “simple” characteristics of an object is sent to PPC for association and integration, and ultimately recognition of the object. Recognition of an object requires an attachment of meaning to the object’s properties. Individuals with damage to PPC may experience agnosia, which is the inability to recognize objects even though basic sensory skills (as revealed by a simple test of tactition) appear to be normal. Agnosias may also occur in the visual and auditory modalities. Although basic tests of visual and auditory sensitivity reveal “normal” abilities, the person with damage to PPC may not be able to connect meaning to visual or auditory input.

The second example illustrates the complexity of PPC function in representing the sensations around us. The PPC is a major player in the creation of proper spatial relationships between our bodies and the world. The absence of such relationships can play havoc with our concept of body image and the ability to produce coordinated movements to negotiate or influence an environment. People with damage to PPC in one hemisphere experience a neurological symptom called hemineglect, where one side of the body or one-half of the environment is regarded as if it doesn’t exist. People with hemineglect may dress themselves on only one side of the body, or even reject one of their limbs as belonging to their body. Clinical observations such as these, when a person is known to have brain damage in PPC, illustrate the complex role of the parietal lobe in the integration and even construction of perception and action.

Supramarginal Gyrus and Angular Gyrus. Two other landmarks on the parietal lobe are shown in Figure 15–5. These are specific regions of parietal association cortex involved in high-level language function. The angular gyrus is shown as the dark turquoise region immediately behind and slightly above the posterior end of the sylvian fissure (see Figure 15–5, lower right). Note the location of the angular gyrus at the boundaries of the parietal, occipital, and temporal lobes. Clinically, lesions to the angular gyrus result in complex language deficits, such as the understanding of metaphor, and difficulty with mathematical concepts and performance. An older theory of language and brain functioning imagined the angular gyrus as the site where written language was transformed into an auditory code for spoken language (Geschwind, 1965). Immediately above and slightly in front of the angular gyrus is the supramarginal gyrus, shown in Figure 15–5 (lower right) as a yellow region of PPC. The supramarginal gyrus is thought to be involved in word meaning, the relation of individual speech sounds to the formation of words, and the ability to connect word meanings with action patterns (i.e., to enable the performance of action on verbal command, such as, “Show me how you whistle”).

Temporal Lobe

The temporal lobe, shown in blue in Figure 15–5, is located on the lower side of each cerebral hemisphere. Toward the front of each cerebral hemisphere, the sylvian fissure is the boundary between the temporal lobe and the frontal lobe above; toward the back the same fissure separates the temporal lobe from the parietal lobe. The upper part of the temporal lobe has a back boundary with the lower parietal lobe, and the lower parts of the temporal lobe have a posterior boundary with the occipital lobe.

Major Gyri. The surface of the temporal lobe visible from the side has three major gyri—superior, medial, and inferior. Immediately below the sylvian fissure is the superior temporal gyrus. As shown in the light blue area of Figure 15–5 (lower right), the upper lip and some surrounding tissue of the superior temporal gyrus is called primary auditory cortex or Heschl’s gyrus. The primary auditory cortex is the first cortical location for processing of auditory signals. More complex processing follows when this initial analysis is forwarded to other locations within the temporal lobe.

Primary Auditory Cortex and Planum Temporale. The anatomy of primary auditory cortex—Heschl’s gyrus—requires additional description with the assistance of a different view of the brain. Imagine drawing an oblique line parallel to and slightly above the sylvian fissure, as shown by the cut line in the upper image of Figure 15–7. Think of this line as one edge of a plane cutting through the cerebral hemispheres, dividing them into upper and lower halves. This horizontal (axial) section is angled to follow the “pitch” of the sylvian fissure. Looking down on the cerebral hemispheres from above after this cut is made and the top half of the hemispheres removed, the top (dorsal) surfaces of the temporal lobes are visible. This view is shown in the bottom image of Figure 15–7. The superior temporal gyrus extends medially from its upper lip, like a shelf of cortex. This shelf, previously hidden inside the sylvian fissure, is an important part of the auditory cortex; it is a continuation of the superior temporal gyrus toward the center of the brain. Part of this surface is called the planum temporale. The lower image in Figure 15–7 shows, for both hemispheres, the primary auditory cortex (colored salmon) as well as the planum temporale (colored blue). In both hemispheres the primary auditory cortex is anterior to the planum temporale. Note the larger surface area of the planum temporale in the left as compared with the right hemisphere.

The primary auditory cortex and planum temporale have interesting features. First, the cells in primary auditory cortex are arranged systematically with respect to auditory signal frequency; this does not seem to be the case for the planum temporale (Langers, Backe, & van Dijk, 2007). A fundamental aspect of peripheral auditory anatomy is the systematic frequency map of sensory receptors embedded within the organ of Corti along the extent of the basilar membrane. The basilar membrane is a ribbonlike membrane housed inside the cochlea, the snail shell–like structure in the inner ear. The sensory receptors along the basilar membrane, and specifically within the organ of Corti, are called hair cells. The cochlea spirals from its base to its tip, and the basilar membrane follows this spiral. Along this systematic map, the highest frequencies (~20,000 cycles per second) heard by humans are sensed by hair cells at the base of the basilar membrane. Moving from the base toward the tip of the membrane the hair cells are sensitive to increasingly lower frequencies. At the tip of the basilar membrane the hair cells are sensitive to the lowest frequency (~20 cycles per second) heard by humans. This systematic frequency representation along the basilar membrane is referred to as tonotopic representation. More detailed information on cochlear anatomy is presented in Chapter 13.

The tonotopic representation of the basilar membrane is projected onto the primary auditory cortex. Research has shown that, moving in a roughly straight line across cell groups within the primary auditory cortex, cells change their responsiveness from very high frequencies to very low frequencies with the same orderly progression of base to tip of the basilar membrane. Note the conceptual similarity between tonotopic organization and somatotopic representation in primary motor and sensory cortex.

Kandel, Schwartz, and Jessel (2000) described the primary auditory cortex as a core of cells surrounded by cortical tissue devoted to increasingly higher-level processing of auditory information. This surrounding tissue is called secondary auditory cortex, and the planum temporale falls into this category. The basic characteristics of acoustic signals, such as frequency, intensity, and duration, are analyzed in primary auditory cortex. This analysis is sent to surrounding temporal lobe tissue (secondary auditory cortex) for higher-level analyses. A form of higher-level auditory analysis is that required for speech and language perception and understanding. The planum temporale is important to perceptual analysis of speech and language, and the evidence of anatomical asymmetries for this part of the temporal lobe has encouraged the view of this cortical region as important to the aptitude among humans to develop and use speech and language.

Wernicke’s Area. Just posterior to the primary auditory cortex, along the back portion of the superior temporal gyrus, is a region of the temporal lobe (and perhaps the lateroventral portion of the parietal lobe) called Wernicke’s area (see lower right image in Figure 15–5, dark blue area toward back of the sylvian fissure). Wernicke’s area was originally defined as the brain region associated with speech and language comprehension. Postmortem examination of the brains of several patients revealed a lesion in this area for those who were known in life to have comprehension impairment for spoken language, with little or no impairment of speech production. These patients had lesions at the very back of the sylvian fissure, in the superior temporal gyrus. Dr. Carl Wernicke, a Prussian physician, examined one famous patient when he was alive and on postmortem examination of the brain located the region known as Wernicke’s area.

Planum Temporale

The planum temporale has a lofty sounding name (the temporal plane) and a scientific history as murky as lofty ideas tend to be. This wedge of brain tissue (see Figure 15–7) tends to be larger in the left than the right hemisphere, including in preverbal infants (Tervaniemi & Hugdahl, 2003). Unfortunately, at least for scientists who enjoy equating size differences with functional differences, chimps also have a larger left than right planum temporale. If only chimps communicated like humans, this would not be a theoretical problem. “Oh bother,” as Winnie the Pooh might say, that is, if bears could talk. In fMRI studies the planum temporale tends to light up on the left side for speech sounds, and on the right side for tones. Some scientists argue that lateralization to the left hemisphere for detection of speech sounds can be found in the planum temporale, but that the broader needs of speech processing (extracting meaning from sound sequences) are accomplished bilaterally (Hickok, 2009). Maybe more chimp research can resolve the true role of the planum temporale in human communication: according to Winnie the Pooh, “Some people talk to animals. Not many listen though. That’s the problem.”

Recent imaging of the perisylvian (surrounding the sylvian fissure) language areas, and the temporal lobe specifically, during speech perception and language comprehension tasks supports the general ideas outlined above. The primary auditory cortex, roughly in the middle of the upper lip of the superior temporal gyrus, is active when a person is required to make decisions concerning individual speech sounds, and especially when these decisions do not require word, sentence, or discourse meaning. The task of detecting and identifying speech sounds activates the part of the auditory cortex devoted to analysis of basic signal characteristics. When a person is required to make language input decisions involving meaning, more widespread regions of the temporal lobe are activated. Single word meaning, meaning tied to different levels of grammatical complexity, and abstract meaning (as in metaphor) engage many different regions of the temporal lobe, as well as regions of the parietal, frontal, and occipital lobes (see Price, 2010 for an excellent review of brain imaging and speech and language perception/comprehension).

The anterior part of the temporal lobe and the middle and inferior temporal gyri also seem to play a role in naming of objects or actions. Almost certainly, the idea of Wernicke’s area as the important location in the temporal lobe for speech and language understanding is an oversimplification.

Occipital Lobe

The occipital lobes (colored purple in Figure 15–5) comprise the posterior parts of the cerebral hemispheres; they are the smallest among the four lobes of the brain. The occipital lobes contain primary visual cortex, whose tissue processes information entering the brain through the eyes. Like the auditory cortex, the occipital lobes contain cells that perform basic analysis of visual signals, as well as cells that perform more abstract, elaborate visual processing. The top right image of Figure 15–5 shows the medial wall of one hemisphere, where the calcarine fissure divides the occipital lobe into an upper (cuneus) and lower (lingual gyrus) portion. Primary visual processing is performed by cortical tissue deep within the calcarine fissure. Extensive connections between cells in the calcarine fissure and other cells within the occipital cortex power more elaborate visual processing.


In Figure 15–8 the left cerebral hemisphere is shown with the lower “lips” of the frontal and parietal lobes and the upper lip of the temporal lobe pulled away to reveal underlying gyri and sulci. The retractable lips of the frontal, parietal, and temporal lobes are referred to as opercula (singular = operculum, from the Latin word for “lid”). The cortex revealed when the opercula are retracted is called the insula or insular cortex. The insula is sometimes described as part of a fifth hemispheric lobe, the limbic lobe (see below). In this chapter the terms insula and insular cortex are used to denote this cortical region without commitment to the notion of the insula as part of a fifth lobe.

Among other functions, the insula appears to be a critical part of cortical tissue engaged in speech and language functions (Ackermann & Riecker, 2010). Clinical cases (where patients with known lesions in and around the insula exhibit certain speech and language difficulties), surgical cases (brain tumors requiring resection of insular tissue), electrical stimulation of exposed brain (in patients undergoing brain resections for severe epileptic seizures), and fMRI studies of healthy individuals point to a role for the front part of the insula in speech motor control and possibly in speech perception. These speech functions of the insula appear to be lateralized to the left hemisphere, in much the same way as speech and language functions are lateralized to Broca’s and Wernicke’s regions in a majority of individuals. Note the proximity of the anterior insula to Broca’s area in the frontal lobe.

The insula also seems to play a role in swallowing, self-awareness, control of heart rate, blood pressure, perception of pain and temperature, dyspnea (breathing discomfort), as well as other functions related to emotions and general body awareness. As pointed out by Ackermann and Riecker (2010), clinical cases involving isolated lesions of the insula are rare. Generally, when a stroke or surgical removal of brain tissue involves insular tissue, other nearby areas of the brain (such as regions in and around Broca’s area, or parts of the primary auditory cortex, or even fiber tracts below the cortical tissue) are also affected.

Limbic System (Limbic Lobe)

Heimer and Van Hoesen (2006) recommend the term limbic lobe to designate the collection of structures within the cerebral hemispheres involved in emotions, motivation, memory, and adaptive functions. Not all authors assign limbic structures the status of a lobe of the cerebral hemispheres, at least in the same sense as the frontal, parietal, temporal, and occipital lobes. Some authors refer to a limbic system to reflect a collection of structures within the brain, all of which play an important role, broadly speaking, in emotional and motivational aspects of behavior. Even authors who have argued persuasively for the existence of a limbic lobe, based on similarities in anatomical characteristics and physiological functions of its component structures, say the “system” versus “lobe” debate is not likely to be settled soon (Heimer & Van Hoesen, 2006). For the following brief discussion, the term limbic system is used.

The most easily visualized structures of the limbic system are seen on the medial surface of a hemisphere (see Figure 15–5, top right). The cingulate gyrus forms an incomplete ring above and around the corpus callosum. The ring is partially completed on the lower side of the hemisphere by the upper gyrus of the medial temporal lobe, called the parahippocampal gyrus (unlabeled in Figure 15–5, top right: locate the most superior blue gyrus to identify the parahippocampal gyrus). The cingulate and parahippocampal gyri are part of the cortex, but their cell structure is different from cells in the primary motor, primary somatosensory, or association cortex (Heimer & Van Hoesen, 2006). The cell structure of limbic cortical areas in humans have been described as more primitive than the cell structure in other cortical areas.

Deep within the parahippocampal gyrus of the temporal lobe is a cell group called the hippocampus, and another cell cluster called the amygdala. These structures are part of the limbic system. In addition, parts of the insula and basal ganglia, as well as cortical structures related to olfaction (sense of smell), and the ventral gyri of the frontal lobe (shown in Figure 15–5, bottom left image) are considered components of the limbic system. These structures are interconnected and make connections with the more sophisticated parts of the cortex as well as with subcortical and brainstem structures.

The sketch presented here of structures and connections of and within the limbic system is complicated, and may seem far removed from the concerns of the speech-language pathologist. Nevertheless, its relevance becomes clear when considering disorders such as dementia, a behavioral syndrome characterized by memory loss, behavioral change, and communication impairment. Dementia has a neurobiological basis that largely originates in structures of the limbic system. In addition, limbic structures are often compromised by brain damage sustained in traumatic brain injury (TBI). Communication problems in persons with TBI often include difficulties with social communication (pragmatics) that can be traced at least partially if not largely to limbic system damage.


The surface of the cerebral hemispheres is made up of many gyri and sulci, some of which have been identified above. This surface topography is composed of gray matter, formed by densely packed clusters of neuronal cell bodies. Cut into the cerebral hemispheres, either by dividing them into left and right halves (a sagittal cut), front and back parts (a coronal cut), or top and bottom parts (a horizontal or transverse cut) and a tremendous volume of white matter is revealed. White matter consists of bundles of myelinated axons running from one group of cell bodies to another group of cell bodies. White matter connects nearby and distant cell groups within the CNS.

As described above, a coronal section of the cerebral hemispheres (see Figure 15–3) shows extensive white matter. At any location within the white matter, fibers run in many different directions, to and from different cell groups. Even though fiber tracts are typically “bundled” together, with a given bundle running from a specific group of cell bodies to another specific group of cell bodies, a selected volume of white matter contains an intermixing of several bundles. A relatively new brain imaging technique called diffusion tensor imaging (DTI) allows scientists to establish the origin, course, and termination of major fiber tracts in the human brain (see sidetrack on DTI). DTI research has established a detailed account of fiber bundles within the brain. Much of the following information, including an organizational scheme for classifying fiber bundles within the cerebral hemispheres, is adapted from a review article by Schmahmann, Smith, Eichler, and Filley (2008). Table 15–1 outlines this classification system. Smits, Jiskoot, and Papma (2014) provide a revew of DTI findings specific to speech and language behaviors.

Association Tracts

Association tracts connect one part of the cortex to another, within the same hemisphere (intrahemispheric). These ipsilateral connections may consist of small groups of fibers running between adjacent gyri, or between more distantly separated gyri within the same lobe. Of greater importance for the current discussion are several tightly organized association tracts that connect cortical areas in one lobe to cortical areas in a different lobe. Table 15–2 lists some of these major ipsilateral, interlobe tracts; these tracts are found in both hemispheres, but may have slightly different forms depending on which hemisphere is examined (see below). The listing of these tracts highlights the extensive interconnectedness of lobes within a single hemisphere. The capability exists for a great deal of information to be shuttled between different cortical regions within a single hemisphere.

As discussed toward the end of the chapter, increasing knowledge of the interconnectedness of intrahemispheric lobes (as well as interhemispheric connections) is guiding scientists away from a “center” oriented view of human behavior (e.g., a focus on Broca’s and Wernicke’s areas in speech and language performance) to a “network” view, wherein multiple brain locations and pathways are organized as a system to generate complex human behaviors.

Arcuate Fasciculus and Speech and Language Functions. Further consideration of one association (intrahemispheric) tract is warranted because of its historical and contemporary importance in speech and language functions of the brain. Table 15–2 lists a tract called the superior longitudinal fasciculus, the main part of which is the arcuate fasciculus (AF) (see Bernal & Ardila, 2009). Figure 15–9 is a DTI reconstruction of the AF, as well as of the inferior longitudinal fasciculus and uncinate fasciculus (compare the course of these latter two tracts to the information provided in Table 15–2). The AF is the arched pathway (hence “arcuate”) with one leg of the “bottom” of the arch in the temporal lobe, from which fibers run slightly back and up into the parietal lobe before turning forward to end as the other leg of the arch in the back part of the frontal lobe. Table 15–2 lists four cortical areas connected by the AF, including Wernicke’s area (temporal lobe), the angular and supramarginal gyri (parietal lobe), and Broca’s area (frontal lobe). This is a standard way to describe the AF, as a fiber tract connecting the receptive language areas (Wernicke’s area and possibly parts of the angular and supramarginal gyri) to the expressive area (Broca’s area). In a structural and functional study of the AF, Takaya, Kuperberg, Liu, Greve, Makris, and Stufflebeam (2015) reported that in semantic tasks, activity in the AF connections in the left hemisphere (the cortical gray matter) were correlated during a semantic task; the connected cortical areas “lit up” together. The right hemisphere regions connected by the AF were not correlated during the semantic task. These functional differences are accompanied by structural differences when the left and right hemispheres are compared. Dick, Bernal, and Tremblay (2014) and Takaya et al. argue that the AF is lateralized to the left hemisphere for speech and language function.

Dorsal and Ventral Streams

An association tract, the arcuate fasciculus (AF), has been known for many years to play a critical role in speech and language functions. The AF connects the posterior part of the superior temporal gyrus (Wernicke’s area) and parts of the parietal cortex to the inferior frontal gyrus (Broca’s area) as well as frontal lobe areas just anterior to Broca’s area. The AF has an important role in phonological processing and complex syntactic structure, for both speech perception and production. Another tract, the uncinate fasciculus (UF), connects anterior portions of the temporal lobe to the inferior frontal gyrus and more anterior regions of the frontal lobe; the UF also has extensive connections with parts of the limbic system. The UF is thought to have a role in semantic processing—the extraction of meaning in language production and comprehension. These two pathways have been combined for a “dorsal stream” (the AF) and “ventral stream” (the UF) model of speech and language processing. The “dual stream” model of the brain basis of speech and language function is controversial but has generated interesting hypotheses for imaging studies and behavioral studies of patients with lesions to the AF and UF. Dick, Bernal, and Tremblay (2014) provide an excellent review of the presumed anatomy and function of the dual-stream hypothesis.


Because fiber tracts within the cerebral hemispheres are intermixed and so densely packed, it is difficult to establish the origins, pathways, and destinations of connections between cell groups. Techniques used in animal research, such as introducing certain chemicals into the brain which “label” specific fiber tracts are mostly not usable in human research. Fortunately, a technique called diffusion tensor imaging (DTI) as well as other related techniques make it possible to monitor selected pathways without posing danger to humans. Water molecules move along specific pathways (fiber tracts) in ways that can be identified by proper computer settings of a brain scanner. In region-of-interest techniques, the brain-scanning instrument is directed at the presumed location of specific pathways, and computer reconstructions of the pathways show their extent, volume, and orientation. Conturo et al. (2008) provide an explanation of the DTI technique, and Saur et al. (2008) and Smits, Jiskoot, and Papma (2014) show how it can be used to understand speech and language connectivity of the brain.

The AF is emphasized because of its prominent role in theories of speech and language functions in healthy and diseased brains. The most prominent and influential of these theories has been referred to as the Wernicke-Geschwind model (Geschwind, 1965), in which the comprehension area of the brain (Wernicke’s area) is connected to the expressive region (Broca’s area) by means of the AF. In this model, acoustic properties of spoken words are first analyzed by the listener in the primary auditory cortex, then sent to Wernicke’s area to convert this “raw” auditory analysis into meaning. The meaningful phonetic sequences thus identified ​—the words—can be transferred to Broca’s area for production via the AF. In the Wernicke-Geschwind model, the processing centers (cortical cell bodies) and pathways (fiber tracts) are fully engaged when a person is asked to repeat a word or series of words. The neurologically intact individual has no problem with this task, because she can comprehend meaning (has a healthy primary auditory cortex and Wernicke’s area) and transfer the comprehended phonetic information via the AF to the brain region specialized for speech production. Within the context of the repetition task, the model predicts that damage to Wernicke’s area impairs repetition due to failure to comprehend. The patient cannot generate a proper, phonetically based “word image” to repeat, even though her brain center for production is intact. This is so even if the primary auditory cortex performs an accurate analysis of the acoustic properties of the incoming speech. A patient with damage to Wernicke’s area who is asked to repeat a simple, short sentence may have normal-sounding articulation (consistent with a “healthy” Broca’s area) but may exchange sounds (“take” instead of “cake” or “burzday” for “birthday”) and even make sentences more complex by adding words and/or additional phrases not included in the target sentence. These errors and complications are recognized by the patient. On the other hand, the limitation on repetition ability in patients with damage to Broca’s area is explained strictly on the basis of impaired production skills. Asked to repeat a short sentence including words such as “cake” and “birthday,” the patient may struggle to produce the words with hesitations, labored dysfluencies, and an unusually slow speaking rate, as evidenced by abnormally long speech sounds. Despite this poor production in the repetition task, the patient demonstrates through comprehension tasks that she knows the words she is supposed to produce.

What are the repetition problems in a patient with a damaged AF but undamaged Wernicke’s and Broca’s areas? This patient can, according to the Wernicke–Geschwind model, comprehend and produce speech in a nearly normal way, but cannot transfer the comprehended message between these two cortical centers. The patient can be shown to have normal comprehension, using nonverbal comprehension tasks such as, “Point to the picture of a dog.” This simple task is challenging for the patient with damage to Wernicke’s area. The patient with damage isolated to the AF has fluent speech, but within this fluent stream may have numerous sound exchange errors (“take” for “cake”) recognized by the patient as mistakes, as revealed by successive attempts to repeat the target utterance to “get it right.” What is unique about the patient’s repetition performance is her inability to repeat, on command, words and sentences when intact comprehension skills are demonstrated. The patient’s spontaneous, conversational speech is likely to be better than her repetition performance.

The performance problems in conduction aphasia are said to be the result of a disconnection syndrome, which in this case is the disconnection of Wernicke’s area from Broca’s area due to an AF lesion. Some have argued that the AF is responsible for transmitting the order of sounds in a word from the comprehension to production areas of the brain (Papagno et. al, 2017).

Other disconnection syndromes have been discussed in the literature for their potential to disrupt speech and spoken or written language performance. For example, disconnection of the occipital from temporal and parietal cortex, resulting from damage to the inferior longitudinal fasciculus (see Table 15–2, Figure 15–9), may impair the ability to read words even though the cortical tissue is healthy (Epelbaum et al., 2008). More generally, a wide range of white matter diseases, in which fiber tracts are damaged but cortical regions are spared, appears to play a significant role in dementia (Schmahmann et al., 2008). Dementia, a disorder of cognition and more specifically of memory and its use in complex tasks such as speech and language, has a high prevalence within the aging population. With respect to speech and language function, white matter clearly matters.

Striatal Tracts

Deep within the cerebral hemispheres there are several clusters of cell bodies, collectively referred to as sub-cortical nuclei. One group of these nuclei comprises components of the basal ganglia (sometimes called basal nuclei). The thalamus, itself a collection of many nuclei, is another major subcortical nucleus. Beneath the cortical rind of gray matter, these nuclei appear as collections of gray matter within the extensive white matter of the cerebral hemispheres. Striatal tracts are fiber tracts connecting the cortical gray matter and these subcortical nuclei. Many of these connections form a loop between cortical and subcortical gray matter structures. This loop plays an important role in motor control, including speech motor control and possibly language production. There are also fiber tracts that connect individual nuclei of the basal ganglia, as well as components of the basal ganglia and the thalamus. These connections also fall under the general category of striatal tracts. Additional detail on the cortical-basal ganglia-thalamus-cortical loop is provided below.

Commissural Tracts

Commissural tracts typically connect a specific region of one hemisphere with its similar topographical region in the other hemisphere. The wording of this description is purposely careful, because of the notion of lateralization of function. Brain regions having the same locations in the two hemispheres most likely do not have the same function. For example, Broca’s area has a sister region in the right hemisphere, but it is not called Broca’s area. The lateralization of speech and language function to the left hemisphere in most people suggests that the same topographical regions in the two hemispheres do not share the same functions. Nevertheless, the third frontal convolutions are connected across the hemispheres by fibers running in the corpus callosum. The same can be said for the other cortical regions described above—they are connected across the hemispheres by the corpus callosum, but the connection does not imply connection for identical function.

Corpus Callosum. The corpus callosum is a massive and complex bundle of fibers. A classic view of the corpus callosum is viewed in the midsagittal plane (see Figure 15–5, top right), where the front-to-back extent of the tract appears as a thick length of arched white matter, shaped somewhat like a flattened letter “C” turned on its right side. The frontmost and backmost parts of the corpus callosum are the genu and splenium, respectively. Between the genu and splenium is the central, main bulk of the corpus callosum, called the body. At the genu, the corpus callosum has a curl of fibers (one end of the “C”) pointing slightly downward and toward the back of the cerebral hemispheres; this backward-directed curl is called the rostrum.

The most anterior and most posterior extensions of the corpus callosum do not extend to the front and back “poles” (end points) of the hemispheres (see Figure 15–5, upper right). Nevertheless, fiber tracts extend from the corpus callosum forward and backward into the most anterior regions of the frontal lobe and most posterior regions of the occipital lobes, connecting these regions across the hemispheres. Finally, although in the midsagittal plane the body of the corpus callosum is beneath cortical tissue, the connecting fibers project upward to reach cortical layers at the top of the hemispheres. The extension of corpus callosum fibers into the front and back parts of the hemispheres, as well as to the top of the cortex, is shown in the sagittal plane (DTI image in Figure 15–10). The “flat” part of the tract, corresponding to the view in Figure 15–5 of the medial part of the corpus callosum, is seen in the middle of the tract in Figure 15–10, and the upcurled fibers reaching to cortical layers are seen all along the length of the tract.

The many millions of fibers (about 200,000,000) in the corpus callosum have a topographical arrangement. The term topographical in this context implies both somatotopicity and systematic fiber arrangement for external signal properties (as in audition and vision).

As shown in Figure 15–5, top right, the rostrum of the corpus callosum terminates its backward path immediately in front of a structure identified as the anterior commissure. The anterior commissure is an interhemispheric (commissural) pathway that connects the orbital cortex (frontal lobe) and parts of the temporal lobe cortex across the two hemispheres. If a pencil point is placed on the anterior commissure and moved toward the back of the brain along a straight line angled slightly downward, the pencil line intersects the posterior commissure (not shown in Figure 15–5). The posterior commissure is an interhemispheric (commissural) pathway connecting parts of the brain involved in the reflex response of the eye’s pupil to light. The line connecting the anterior and posterior commissures is often used to define a surgical reference plane, especially for the therapeutic placement of intracranial electrodes.

The corpus callosum plays a storied role in the history of disconnection syndromes. As reviewed by Gazzaniga (2000) and Doron and Gazzaniga (2008), various parts of the corpus callosum, and in many cases the entire corpus callosum, have been surgically cut to relieve chronic epileptic seizures that cannot be controlled by drugs. Many of these “split-brain” patients, when tested under controlled laboratory conditions, have provided evidence of the different functions of the two hemispheres and the consequences of not having communication between the hemispheres. In some cases, one side of the brain literally does not know what is going on in the other side of the brain.

Descending Projection Tracts

Descending projection tracts include the corticobulbar and corticospinal tracts, as well as tracts running from many cortical regions to the thalamus (corticothalamic tracts, see Table 15–1). Figure 15–11 shows in a schematic coronal view the corticobulbar and corticospinal fiber tracts. The corticobulbar tract (“bulbar” is a term used to indicate the brainstem, and more specifically the pons and medulla), represented in Figure 15–11 by the solid pinkish-red and orange lines, includes fibers originating in cortical cell bodies that make a first synapse in one of the several brainstem motor nuclei. The corticospinal tract, represented in Figure 15–11 by the dashed blue lines, includes fibers originating in the cortex and making a first synapse in motor cells of the ventral spinal cord. Motor nuclei in the brainstem and spinal cord axons leave the CNS to innervate muscles of head and neck structures and the limbs and torso.

As the corticobulbar and corticospinal tracts descend from the cortex to lower regions of the CNS, their location within the brain is designated by different terms. For example, fibers of the two tracts are issued from cell bodies all over the cortex and form a fanlike pattern called the corona radiata. The fibers of the corona radiata contribute to a good portion of the white matter immediately below the cortex. The corona radiata are represented schematically in Figure 15–11, and in the more anatomically correct image of Figure 15–12. As fibers in the corona radiata descend, they gather into a relatively tight bundle that passes between subcortical nuclei to reach the more inferior brainstem. The sagittal view in Figure 15–12 shows the corona radiata merging into this tight bundle. This part of the descending tracts, where the corticobulbar and corticospinal tracts are lateral to the medial thalamus and medial to the caudate nucleus and lateral lentiform (globus pallidus and putamen) nuclei is called the internal capsule (see Figures 15–11 and 15–12).

Internal Capsule. The coronal slices in previous figures show the internal capsule at a single location along the front-to-back extent of the cerebral hemispheres (e.g., see top right of Figures 15–2 and 15–3, slice location not labeled in the figures). A greater appreciation for the distribution of these fiber tracts is gained from careful examination of Figure 15–12 (upper left), where the front of the head is toward the left of the image. Here the cortical tissue has been stripped away to reveal the fibers of the corona radiata and internal capsule. Even though the internal capsule is the tightly gathered merger of the many fibers of the corona radiata, the internal capsule has an anterior, middle, and posterior part (IC = internal capsule in Figure 15–12, upper image). The precise location of a coronal slice therefore determines which part of the internal capsule is displayed. Like so many other parts of the brain, the internal capsule is not a random jumble of fibers, but is arranged systematically based on the cortical origin of the fibers. In a horizontal (axial) slice (inset, lower right of Figure 15–12; the anterior part of the brain is toward the top of the image) the internal capsule in each hemisphere has a boomerang shape with the “angle” of the boomerang most medial and the two arms extending away from this angle anterolaterally and posterolaterally. To provide a rough idea of the systematic arrangement of fibers within the internal capsule, most corticobulbar fibers associated with control of facial, jaw, tongue, velopharyngeal, and laryngeal muscles run through a compact bundle close to or within the angle (called the genu) of the internal capsule. Fibers descending to motor neurons in the spinal cord are mostly located in the posterior arm (called the “posterior limb”) of the internal capsule, and within that limb the fibers for the legs are most posterior, and those for the arms are closer to the angle. These are illustrations of the systematic arrangement of fibers within the internal capsule, and are not meant to be exhaustive. For example, fibers running to the cortex from the thalamus also form parts of the internal capsule (see section below on Ascending Projection Tracts).

Descending fibers leave the internal capsule and continue their downward path in the cerebral peduncles, a tract in the central part of the midbrain. The largest portion of these fibers runs in the crus cerebri, an anterior part of the cerebral peduncles (the terms “cerebral peduncles” and “crus cerebri” are occasionally used interchangeably; see Figure 15–12). The fibers continue through the pons in small bundles, or fascicles, and are gathered back together in the medulla as the pyramids. Some descending fibers in the cerebral peduncles, pontine fascicles, and medulla leave the descending tract to make synapses with motor nuclei in the midbrain, pons, and medulla. These fibers belong to the corticobulbar tract, and the synapses they make within the brainstem define the termination of the tract for those fibers. The fibers continuing into the spinal cord belong to the corticospinal tract; these make synapses in the ventral (anterior) gray matter of the spinal cord, where spinal motor neurons are found.

The general routes of the corticobulbar and corticospinal tracts are summarized in the right column of Figure 15–11. A slightly more detailed representation of the corticobulbar tract is provided in Figure 15–13. For the sake of simplicity, Figure 15–13 shows connections originating from only the right hemisphere. The left corticobulbar and corticospinal tracts are a mirror image of the tracts issued from the right hemisphere. The view is as if you were looking at the ventral surface of the brainstem and spinal cord. The lines are shown terminating at each of the three levels of the brainstem (midbrain, pons, medulla), indicating the presence of motor nuclei at each level (see section below on Cranial Nerves and Associated Brainstem Nuclei).

Innervation of Brainstem and Spinal Motor Neurons. In Figure 15–13, the solid pink lines represent bilateral innervation of cell bodies in the brainstem by cortical cell bodies. In other words, cells in the cortex of one hemisphere—say, those controlling contraction of the palatal levator muscle, the muscle that lifts the soft palate and pulls it back toward the posterior pharyngeal wall—are connected by corticobulbar fibers to the nucleus containing palatal levator motor neurons on both sides of the brainstem. Bilateral innervation means that there is an ipsilateral (same side) and contralateral (opposite side) connection. This is shown in Figure 15–13 by solid lines extending from the right hemisphere to the right (ipsilateral) and left (contralateral) sides of the brainstem, at all three levels.

The overall innervation pattern of brainstem motor nuclei by corticobulbar fibers is mostly, but not exclusively, bilateral. Figure 15–13 shows by dashed lines exclusively contralateral connections between cortical cells in the right hemisphere and motor nuclei in the pons and medulla levels of the left brainstem. Certain brainstem nuclei, or parts of nuclei, are innervated only by fibers arising in the cortex of the opposite hemisphere. These facts concerning the connection patterns in the corticobulbar tract are considered in greater detail below in the section on cranial nerves. As explained in that section, knowledge of the bilateral and contralateral connection patterns in the corticobulbar tract has substantial value to the practicing speech-language pathologist.

The path of the corticospinal tract is shown in Figure 15–14 for one side of the brain. Fibers from each hemisphere run down their respective sides until the majority of fibers from one side (about 80%) cross to the other side at the decussation of the pyramids, a landmark on the ventral surface of the medulla created by the crossing fibers (see below, Figure 15–20). The fact that so many fibers from one cerebral hemisphere eventually travel in the spinal cord on the side opposite to their cortical origin accounts for the well-known fact that the left hemisphere controls limbs on the right side, and the right hemisphere controls limbs on the left side. In Figure 15–14, the descent of the corticospinal tract through the internal capsule and to its crossover point within the inferior medulla is summarized by the pathway of the green line.

Ascending Projection Tracts

Ascending fiber tracts are typically associated with sensory pathways, which are projection tracts from points below to points above. Sensory events begin in an end organ of the body, which may include touch, pressure, limb position and velocity, vibration, pain, temperature, taste, odor, light, and sound receptors. When these receptors are stimulated, an impulse is, in most cases, sent from them to a ganglion. Ganglia contain first synapses along a sensory pathway located outside the CNS but close to the entry point near the spinal cord or brainstem.

Somatosensory Pathways. Somatosensory pathways constitute a major portion of the ascending projection tracts. These tracts run in the opposite direction from the descending projection tracts. The “points below” are the end organs, where stimuli are sensed, and the “points above” include several synapses along the ascending pathway with a final destination in the cortex.

Posterior Columns. There are two major somatosensory pathways for stimuli sensed below the neck (that is, on the torso or limbs). One of these, the posterior column-medial lemniscal tract (Blumenfeld, 2010), carries sensory information from one side of the body. This sensory information enters the spinal cord after making a first synapse in a dorsal root ganglion. The fibers entering the spinal cord run up the same side of the body until reaching the dorsal part of the medulla (the lowest part of the brainstem, at the top of the spinal cord), where the fibers make a synapse and cross to the opposite side to run up through the brainstem and thalamus before terminating in the primary sensory cortex and surrounding areas. This means that sensation from one side of the body is processed in the cortex on the opposite side of the brain. Note the parallel to the corticospinal tract, one of the major descending projection tracts described above. The descending corticospinal tract crosses over on the ventral surface of the medulla, whereas the ascending posterior column-medial lemniscus tract crosses over in the dorsal (posterior) part of the medulla. This ascending tract carries information on fine touch, vibration, and joint position.

Pain, Temperature, Crude Touch. A second ascending pathway for sensory stimuli entering the spinal cord is called the anterolateral tract. This tract carries information on pain, temperature, and “crude” touch (Blumenfeld, 2010). Like the posterior column-medial lemniscus tract, the anterolateral tract conveys information to the cortex on the side opposite to the stimulation. An important difference from the posterior column-medial lemniscus tract is the crossover point—the decussation—for pain/temperature/crude touch fibers entering the spinal cord. The latter fibers cross over to the other side of the spinal cord almost immediately after entering the cord, roughly at the level of entry. The fibers then ascend in the anterolateral tract on the side opposite their entry point. Recall that the posterior column-medial lemniscus fibers ascend in the spinal cord on the same side of entry before crossing over in the posterior medulla. The difference in decussation points for these two major ascending tracts has important clinical implications when a neurologist administers a set of tests to localize a lesion.

Both the posterior column-medial lemniscus and anterolateral tracts send their information to the thalamus, where synapses are made and fibers are sent to the cortex. In addition, visual and auditory ascending fibers, carrying information from the retina (vision) and hair cells (audition), make synapses in the thalamus before projecting to the visual and auditory cortical areas. This mass of thalamocortical fibers, or projections, constitute a significant volume of the white matter of the cerebral hemispheres. The internal capsule and corona radiata include these ascending fibers. Typically, any region of white matter in the cerebral hemispheres includes a mix of descending and ascending pathways as well as fibers running to and from the cortex and striatum, and cortex and cerebellum. The intertwined, dense, multimillion-fiber nature of the white matter requires special techniques to determine where fibers originate and where they end (see sidetrack on “DTI”). This mixing of so many fiber types within any given region of white matter also means that white matter disease, as in certain dementias, is likely to produce multiple symptoms associated with multiple systems within the brain that send and receive axon bundles for transmission of important information.


The subcortical nuclei include the various structures of the basal ganglia (also referred to as the basal nuclei), the thalamus, the hypothalamus, and other structures of the limbic system (such as the amygdala and septal nuclei). The cerebellum is subcortical but is typically discussed separately from subcortical structures. In this section the focus is on the basal ganglia, thalamus, and cerebellum.

Basal Ganglia

The basal ganglia include the caudate and putamen nuclei (which together constitute the striatum), the globus pallidus (which paired with the putamen is referred to as the lenticular or lentiform nucleus), the subthalamic nucleus, and the substantia nigra. Technically, the substantia nigra is not a subcortical nucleus (that is, below the cortex but within the cerebral hemispheres) but rather a brainstem nucleus, located in the ventral midbrain (see below, Figure 15–23). The substantia nigra is included here as a subcortical nucleus because of its close anatomical and functional connections with the striatum and subthalamic nucleus.

The gross anatomy of the basal ganglia is best appreciated in two views, one coronal and the other sagittal. Figure 15–15 shows coronal slices of the cerebral hemispheres and the top of the brainstem, roughly midway between the front and back of the brain. In these artist’s renditions, nuclei are shown as darker areas, tracts as lighter areas. The caudate, putamen, globus pallidus, substantia nigra, and subthalamic nucleus are labeled on the left and right images; the left image is a “zoom” view of the full slice on the right. The thalamus is not a basal ganglia structure but is shown here for orientation purposes and because of its role in the processing of basal ganglia information (see below). Note the location of the putamen, deep to the insula; in this coronal slice the putamen is the most lateral of the basal ganglia structures. Just medial to the putamen is the globus pallidus, and together these two structures form a curved, lens-like mass of cells, explaining why the combined nuclei are called the lentiform or lenticular nucleus. Superior and medial to the lentiform nucleus and just lateral to the lateral ventricle is the caudate nucleus, which appears in this slice as a small, oval mass. Recall that the caudate and putamen are together called the striatum—note how the superior tip of the putamen is “pointing” toward the caudate. The significance of the caudate-putamen proximity is explained in the next paragraph. Inferior and medial to the lentiform nucleus is the aptly named subthalamic nucleus (note its position relative to the massive thalamus). Inferior to the subthalamic nucleus, the relatively long, oblique strip of darkened tissue is the substantia nigra, located ventrally in the superior part of the midbrain.

Also labeled in Figure 15–15 is a pale white strip of tissue—a fiber tract—separating the lentiform nucleus from the more medial caudate, thalamus, subthalamic nucleus, and substantia nigra. Much of this fiber tract is composed of the corticospinal and corticobulbar tracts, as well as striatal tracts. The tract also includes fibers carrying sensory information from the thalamus to the cortex, and from brainstem structures to structures of the basal ganglia. The part of this tract running through the basal ganglia structures is the internal capsule (see Figure 15–12). The internal capsule is an important anatomical landmark and often figures prominently in deficits resulting from stroke.

The specific appearance of basal ganglia structures, and in some cases the presence of a structure in a single coronal slice depends substantially on the location of the slice along the anteroposterior axis of the cerebral hemispheres. An appreciation for this dependency can be gained by studying Figure 15–16 (the front of the brain is to the left), a sagittal-view drawing of the complex configuration of basal ganglia structures. The cerebral cortex and cerebral white matter have been eliminated from the figure, leaving the structures of the basal ganglia “floating” free from their moorings within the cerebral hemispheres. Note the “C”-shaped form of the caudate nucleus, which is massive toward the front of the hemispheres and increasingly narrow as it curls toward the back of the brain and turns around to point forward. The tail of the caudate nucleus points so far forward it terminates ventral to the globus pallidus. The caudate and putamen are joined at the anterior end of the nuclei and split apart as the image is viewed from left to right (that is, from anterior to posterior within the cerebral hemispheres). The channel between the caudate and putamen, created as they separate, is filled by fibers of the internal capsule. The strands of light pink tissue “bridge” the spaces between the caudate and putamen at their most anterior location. This streaked or striated appearance of the internal capsule gives the name “striatum” to the putamen and caudate nuclei. The sagittal view also shows the globus pallidus in relation to the more lateral putamen, and the complex spatial configurations of the other nuclei discussed above.

Stay updated, free articles. Join our Telegram channel

Aug 28, 2021 | Posted by in OTOLARYNGOLOGY | Comments Off on Neural Structures and Mechanisms for Speech, Language, and Hearing

Full access? Get Clinical Tree

Get Clinical Tree app for offline access