The Lobes of the Brain

The four lobes at a glance

The cerebral cortex of each hemisphere is conventionally divided into four lobes. The boundaries follow major grooves in the folded surface and, historically, the bones of the skull, which is where the lobes take their names. The scheme is a practical way to organise a vast sheet of tissue, and it maps loosely onto function: each lobe is home to areas that lead on particular tasks.

Two landmarks separate the lobes. The central sulcus, a deep groove running across the top of the hemisphere, marks the border between the frontal and parietal lobes, and it neatly divides the motor cortex in front from the somatosensory cortex behind. The lateral sulcus, a deep fold on the side, marks off the temporal lobe below.

The frontal lobe

The frontal lobe is the largest of the four and sits at the front of the hemisphere. It is most strongly associated with action and control, both the control of movement and the higher control of behaviour that we call executive function. Damage to the frontal lobe can change movement, judgement, and even personality, which is why it has long fascinated clinicians.

Its internal arrangement is not a jumble. Running from the back of the lobe to the front, the areas form a gradient: the primary motor cortex on the precentral gyrus, immediately in front of the central sulcus, which sends the final commands for voluntary movement to the muscles of the opposite side of the body; then the premotor cortex and supplementary motor area, which prepare and sequence movements before they are executed; then the frontal eye fields and, usually on the left, Broca's area, which contributes to producing speech; and finally, occupying roughly the front third of the lobe, the prefrontal cortex, which holds goals, plans, and rules that may not be acted on for hours. The further forward you go, the more abstract the representation and the longer the time horizon of the behaviour it controls, which is why damage at the back of the lobe paralyses a hand while damage at the front changes a person.

The frontal lobe's reputation owes much to a single accident. On 13 September 1848, near Cavendish, Vermont, a railway construction foreman named Phineas Gage was tamping blasting powder into a hole when the charge went off. The tamping iron, a rod more than a metre long and about three centimetres thick, was driven upward through his left cheek, behind his eye, and out through the top of his skull, destroying a large part of the left frontal lobe, including the orbitofrontal cortex on the underside. He did not lose consciousness for long. He spoke within minutes. His movement, speech, and memory were spared, and he lived another twelve years.

What is reliably documented is the injury and the survival. What made the case famous is the claim that his character changed: his physician, John Harlow, reported that a capable and well-liked foreman became profane, impatient, and unable to hold to a plan, and summarised the transformation in the phrase that his friends said he was "no longer Gage".

Modern reconstructions from Gage's preserved skull place the damage predominantly in the left prefrontal and orbitofrontal cortex, and the syndrome that later work has associated with orbitofrontal damage, impaired social judgement and poor decision making with intellect and memory intact, fits the reliable parts of the account rather well. For what that region does and how it does it, see the prefrontal cortex.

The parietal lobe

Behind the central sulcus lies the parietal lobe, whose leading role is to make sense of the body and its place in space. Immediately behind the central sulcus is the primary somatosensory cortex, the strip that receives the sense of touch, pressure, temperature, pain, and the position of the limbs from the opposite side of the body. Like the motor cortex opposite it, this strip is arranged as an orderly map of the body, and the map is grossly out of proportion: the hands and lips are enormous, the trunk and legs small.

The distortion is not a quirk. Cortical territory is allocated by how much fine sensation or control a body part requires, not by how big that part is. A fingertip is packed with touch receptors and must resolve detail at a scale of millimetres; the skin of the back is sparsely supplied and resolves almost nothing. More receptors send more fibres, more fibres claim more cortex, and the resulting figure, the homunculus, is the body redrawn in the proportions the brain actually assigns it. The cerebral cortex page sets this out in full.

Beyond raw touch, the parietal lobe integrates sensation into a coherent sense of space. It helps us know where our body is, track where objects are around us, and guide reaching and grasping. The right parietal lobe in particular is important for spatial attention, and damage there produces one of the strangest syndromes in neurology, which is worth deriving rather than merely describing.

The parietal lobe does not stop at the sensory strip. Behind it, a deep groove, the intraparietal sulcus, divides the superior from the inferior parietal lobule and is itself the engine room of the lobe's more interesting work. The cortex lining it holds maps of space in the coordinate frames the body actually needs, centred on the hand, the eye, and the head, and it is the destination of the dorsal visual stream described below. It is also, and this is not a coincidence, one of the regions most reliably engaged when people process quantity and number. A structure that already computes how far and how much in space is well placed to compute how many, and the fact that arithmetic recruits a spatial structure is one of the more suggestive findings in cognitive neuroscience.

Below the intraparietal sulcus, in the inferior parietal lobule, sit two gyri that are easy to skip over and should not be: the supramarginal gyrus, curling around the upturned end of the lateral sulcus, and the angular gyrus just behind it. They lie at a junction, the meeting point of the parietal, temporal, and occipital lobes, which means they are ideally positioned to combine touch, sound, vision, and space. Cultural skills that are late in evolution and late in a child's life, reading, writing, and calculation, all depend on exactly that kind of cross-modal combination, and all of them are disturbed by damage here.

Look at that list of four. Nothing obviously connects them: naming your fingers, telling left from right, writing, and arithmetic have no shared subject matter, and no textbook would have grouped them in advance. Yet a lesion in one small patch of left inferior parietal cortex can take out all four together, and that fact is instructive whatever its ultimate explanation turns out to be. Reading, writing, and arithmetic are not faculties in the way that vision is a faculty. They are recent assemblies, built by education out of older parts, in this case out of spatial and body-related machinery that evolved for something else, and an assembly can come apart. Fingers are the first counters a child uses; number is laid out in space; letters are shapes that must be mapped to sounds and to hand movements. Damage the region where body, space, and symbol meet, and all of it goes at once.

Honesty requires a caveat, and it is a real one. Whether Gerstmann syndrome is a genuine unitary syndrome, a single underlying deficit with four faces, or merely a coincidence of anatomy, four separable functions that happen to be supported by adjacent tissue and so tend to be lost together, has been debated for decades and is not settled. The four elements are often reported without each other, and the pure tetrad is rare. The safe conclusion is the one that matters here anyway: whatever binds them, these abilities are assembled rather than given, and that is why a single well-placed lesion can dismantle them.

Why neglect is a right-hemisphere syndrome

A patient with damage to the right parietal lobe may eat only the food on the right half of the plate, shave only the right half of the face, and, asked to draw a clock, crowd all twelve numerals into the right-hand side of the dial. Asked whether anything is wrong, they will say no. This is hemispatial neglect: not blindness on the left, since the eyes and the visual pathways are intact, but a failure to attend to the left at all, so complete that its absence is not noticed. The left half of the world has not gone dark. It has stopped being a place where things can be.

Now the question the textbooks skip. Neglect after right parietal damage is common and can be severe and lasting. Neglect of the right side of space after left parietal damage is rare, mild, and usually transient. Why should the brain be asymmetric here at all? Everything else about the cortex is mirrored: the left hemisphere feels and moves the right side of the body, the right hemisphere the left, cleanly and symmetrically. Why not attention?

The answer is that attention to space is not mirrored, and the anatomy of its distribution is the whole explanation.

Follow the consequences. Destroy the left parietal cortex and you lose a system that was attending to the right side of space. But the right hemisphere was also attending to the right side of space, and it survives. It covers for the loss, and the patient shows little or no lasting neglect. Now destroy the right parietal cortex. You lose the only system that was attending to the left side of space, because the left hemisphere never attended there. Nothing is left minding the left half of the world, and it disappears from the patient's awareness entirely.

The asymmetry of the deficit is therefore a consequence of an asymmetry in the design. There is no mirror-image syndrome because there is no mirror-image anatomy. This also explains a second observation that is otherwise puzzling: neglect is a disorder of attention rather than of sensation, which is why it can be temporarily relieved by manoeuvres that force attention leftward, and why a neglect patient who is prompted can often report the very object they had failed to notice a moment before. The information was arriving. Nothing was asking for it. See attention and the brain for the wider account of how attention is directed.

The temporal lobe

The temporal lobe lies low on the side of the hemisphere, roughly behind the temples, below the lateral sulcus. It has three closely related headline roles: hearing, understanding language, and memory. On its upper surface sits the primary auditory cortex, the first cortical stage in processing sound, which analyses pitch and the pattern of what we hear.

Nearby, usually in the left hemisphere, is Wernicke's area, central to the comprehension of language. Where Broca's area in the frontal lobe helps produce speech, Wernicke's area helps make sense of it. The two are linked by a bundle of fibres, and damage to Wernicke's area produces a different pattern from Broca's: speech remains fluent but its meaning is disturbed, and understanding is impaired.

That two-area picture is worth stating, because it is the one every textbook diagram shows and the one most people have met. It is also, in its strong form, out of date. Inherited from nineteenth-century neurology, it has been substantially revised: language depends on distributed dorsal and ventral streams running across the left frontal, temporal, and parietal cortex and on the white matter linking them, and the classic areas are neither as sharply bounded nor as functionally specific as the diagram implies. Damage confined to Broca's area, for instance, rarely produces a lasting Broca's aphasia on its own. The language and the brain page sets out what has replaced the classic model and why.

The temporal lobe is also bound up with memory. Buried on its inner surface is the hippocampus, a structure that is essential for forming new long-term memories of facts and events. Although the hippocampus is part of the limbic system rather than the cortical sheet itself, it lies within the temporal lobe, and damage there can leave a person unable to lay down new memories while older ones remain. This is why the temporal lobe is so often described as the seat of memory as well as hearing and language.

Immediately in front of the hippocampus, still within the medial temporal lobe, lies the amygdala, a small almond-shaped group of nuclei. It attaches emotional significance to what the senses deliver, and it is central to the learning and expression of fear: it is the structure that decides that this face, this sound, or this place is one to be wary of, and it does so fast, before the cortex has finished identifying what it is looking at. Its position matters. Sitting at the front of the temporal lobe, it receives the output of the ventral visual and auditory streams just as they finish computing what a thing is, which is precisely where a valuation stage belongs. It is treated in full on the limbic system page.

The temporal lobe's most revealing region, however, is on its underside, and it is the one that settles an argument the rest of this page has to make.

The fusiform gyrus: what prosopagnosia proves

Turn a hemisphere over and look at the underside of the temporal lobe. Running front to back along it is the fusiform gyrus, and within it, usually with a marked bias to the right hemisphere, is a patch of cortex that responds far more strongly to faces than to any other class of object. Nancy Kanwisher, Josh McDermott, and Marvin Chun described it with functional imaging in 1997 and named it the fusiform face area.

The reason to care is not that the brain has a face detector, interesting though that is. It is what happens when the region is destroyed.

Take the deficit apart carefully, because every clause of it is doing work.

Now assemble the conclusion. Vision works. Identity works. Voice reaches identity. Only the visual route to identity is severed. Therefore the visual route to identity must be a separate thing, made of separate tissue, capable of being destroyed on its own while everything on either side of it survives. Recognition is not a by-product of seeing clearly; it is a computation performed by particular cortex, and you can take that cortex away and leave the eyes, the visual field, the memory, and the person perfectly intact.

Two further points keep the account honest. First, it is disputed whether the fusiform face area is a module built specifically for faces or a region of expertise for any category whose members must be told apart at the level of the individual rather than the class; bird experts and car experts show heightened responses there too, and the argument is not closed. Second, ventral-stream damage does not confine itself to faces. Related lesions produce visual agnosia more broadly, an inability to recognise objects that are plainly seen, which is the same lesson written larger. Neither caveat touches the inference above. Whatever the fusiform gyrus is for, its destruction separates seeing from knowing, and that separation is the thing that had to be shown.

The occipital lobe

At the very back of the hemisphere sits the occipital lobe, the smallest of the four and the most single-minded: it is devoted almost entirely to vision. Signals from the eyes travel through the thalamus to the primary visual cortex, known as V1 or Brodmann area 17, which is the first cortical area to receive and analyse what we see. It occupies the rear tip of the lobe and, mostly, the walls of a deep groove on the medial surface called the calcarine sulcus, so that the great majority of it is hidden between the hemispheres rather than exposed on the outside.

V1 breaks the visual scene into its basic elements, detecting edges, orientation, movement, and contrast, before passing the results forward to surrounding visual areas that build up colour, shape, depth, and, eventually, recognition. Because vision is processed here, damage to the occipital lobe causes blindness in part of the visual field even though the eyes themselves are intact, a reminder that we see with the brain as much as with the eyes.

The reason such damage produces a hole in a particular part of the visual field, rather than a general dimming, is that V1 is a map. The visual field is laid out across the cortical sheet in an orderly fashion, so that neighbouring points in the world are handled by neighbouring points of cortex. This property is called retinotopy, and it is why a small stroke in the occipital lobe blinds a small, predictable, precisely located region of the visual field on the opposite side.

V1 also has a signature you can see with your unaided eye. Running through it, in cortical layer IV, is a dense band of myelinated fibres carrying visual input from the thalamus, and it is thick enough to appear in an unstained slice as a pale stripe across the grey matter. It is called the stria of Gennari, after Francesco Gennari, who described it in the eighteenth century, and it is the reason V1 is also called the striate cortex and the visual areas around it the extrastriate cortex. The detail is worth pausing on, because it is a rare case of function being legible in raw tissue: cut a brain across the occipital pole and you can literally see where the visual cortex begins and ends, marked out by the sheer density of the input arriving there. Notice, too, that it is layer IV that is enlarged, the input layer, which is exactly the opposite of what happens in the primary motor cortex, where layer IV nearly vanishes and the output layer V swells. Architecture follows job.

The visual system page treats retinotopy, the receptive fields of V1 neurons, and the hierarchy of extrastriate areas in full.

Vision, however, is not finished in the occipital lobe. From here it flows forward into the parietal and temporal lobes along two separate routes, and that split is worth explaining rather than merely naming.

Why vision splits in two

Leaving the occipital lobe, visual information travels forward along two anatomically distinct routes. The dorsal stream runs upward into the parietal lobe. The ventral stream runs downward and forward into the temporal lobe. They are often labelled the "where" and "what" pathways and left at that, which explains nothing. The real question is why a brain would go to the trouble of building two visual systems when it has already gone to the trouble of building one.

The answer is that acting on an object and identifying an object are not the same computation, and the information each needs is close to incompatible.

Set those two requirement lists side by side and the split stops looking like an oddity. One system must be exquisitely sensitive to size, position, and moment-to-moment change; the other must be blind to size, position, and moment-to-moment change, because that is what invariance means. One works in coordinates centred on the body; the other in coordinates that ignore the body entirely. A single system cannot be both. So the brain builds two, feeds them from the same occipital source, and sends them to the two lobes best placed to use them: the parietal lobe, which already holds a map of the body in space, and the temporal lobe, which already holds memory.

The evidence that these really are two systems, and not one system described two ways, is a double dissociation, the most convincing pattern a lesion study can produce. Take a slot cut in a disc and hand a patient a card.

1. Ventral-stream damage

   The patient cannot say what orientation the slot is in. Asked to match the card's angle to it, or to describe it, or to draw it, they fail: the shape simply is not available to them. Then ask them to post the card through the slot, and they do it in one smooth movement, rotating the wrist correctly on the way. They cannot see the orientation. They can act on it.

2. Dorsal-stream damage

   The mirror image. The patient describes the slot's orientation without difficulty, matches it, draws it. Then ask them to post the card and the hand goes to the disc at the wrong angle and jams. They can see the orientation. They cannot act on it.

Two patients, two lesions, two failures that are precise opposites of each other. Neither is "blind" and neither is "clumsy" in any general sense; each has lost one specific use of vision while keeping the other intact. That is what forces the conclusion. Seeing in order to know and seeing in order to do are carried by different tissue, and you can take away one and leave the other standing.

And the ventral half of that argument has already been made, in the strongest form available, a few sections above. Prosopagnosia is ventral-stream damage in isolation: temporal-lobe lesion, "what" lost, "where" untouched, the patient reaching accurately for a face they cannot identify. The fusiform gyrus is where the ventral stream's final answer is computed, and destroying it removes the answer while leaving the seeing and the knowing on either side of it intact. The stream is not a diagram drawn over a scan. It is tissue, and it can be cut.

The insula and limbic lobe

The four named lobes do not quite account for the whole cortex. Folded deep within the lateral sulcus, hidden beneath the frontal, parietal, and temporal lobes, lies the insula, sometimes counted as a fifth lobe. It is involved in the sense of the body's internal state, in taste, and in the felt quality of emotions such as disgust.

On the inner border of each hemisphere, wrapping around the corpus callosum, is the limbic lobe, a ring of older cortex that includes the cingulate gyrus and connects closely with the deeper structures of the limbic system. This border region is bound up with emotion, motivation, and memory, linking the thinking cortex to the parts of the brain that give experience its emotional colour.

Why the lobes work together

It is tempting to read the map of lobes as a set of separate machines, one for movement, one for touch, one for hearing, one for vision. That picture is misleading. The lobes are densely interconnected, and almost everything the brain does draws on several at once, coordinated through the white-matter tracts running between them.

Consider recognising a friend across a room. The occipital lobe processes the visual image; areas at the border of the temporal and occipital lobes identify it as a face; the temporal lobe links that face to memories and a name; and the frontal lobe directs attention, decides how to respond, and may prepare a greeting. No single lobe does this alone. The division into lobes tells us where particular kinds of processing are concentrated, but the mind emerges from the whole cortex working as one.

Related topics