Epilepsy, psychiatry, and neurology


Address correspondence to Dr E. H. Reynolds, Institute of Epileptology, Weston Education Centre, King’s College, Denmark Hill Campus, Cutcombe Road, London SE5 6PJ. E-mail: reynolds@buckles.u-net.com


This article reviews the relationship between the psychiatry and neurology of epilepsy, especially in the last 100 years. Throughout most of its recorded history of 3 to 4 millennia epilepsy has been viewed as a supernatural or mental disorder. Although first suggested by Hippocrates in the 5th century B.C., the concept of epilepsy as a brain disorder only began to take root in the 17th and 18th centuries. The discipline of neurology emerged from “nervous disorders” or neuropsychiatry in the late 19th century, when vascular theories of epilepsy predominated. By the turn of the 19th century psychiatry and neurology were diverging and epilepsy remained to some extent in both disciplines. It was only in the middle of the 20th century with the development of electromagnetic theories of epilepsy that the concept of epilepsy per se as a neurological disorder was finally adopted in international classifications of disease. This was associated with a refined definition of the ictal, pre-, post-, and interictal psychological disorders of epilepsy, which have contributed to a renaissance of neuropsychiatry. At the beginning of the 21st century and the centenary of the ILAE psychiatry and neurology have been converging again, led in some respects by epilepsy, which has provided several useful models of mental illness and a bridge between the two disciplines.

One of the most remarkable developments in the history of epilepsy occurred in the middle of the last hundred years since the birth of the International League Against Epilepsy (ILAE) in 1909. In 1960 the World Health Organization (WHO) made the distinction between epilepsy “per se” and epilepsy with deterioration or psychosis in their “International Classification of Diseases.” This simple but momentous decision overturned the millennia-old traditional view that epilepsy was a mental disorder, although it took many more years for all national and psychiatric classifications to fall into line with the WHO (Hill, 1981).

This review explores neurologic and psychiatric developments in the late 19th century and early 20th century, which culminated in this fundamental change in perception, and how this has influenced recent trends in the psychiatry of epilepsy.

Epilepsy as a Mental Disorder

The earliest detailed descriptions of the various seizure types we recognize today can be traced to the Babylonians in the middle of the second millennium B.C. (Kinnier Wilson & Reynolds, 1990). Epilepsy is such an unusually dramatic and frightening disorder that it is hardly surprising that our ancestors viewed it as supernatural, caused by the invasion of the body by evil spirits, the forerunner of the Greek concept of “the sacred disease,” which included possession by gods. In the Greco-Roman world the word “lunatic” was restricted to epilepsy and was distinguished from “maniacs” or mad people who were possessed by demons, not gods (Hill, 1981). Although Hippocrates and later Galen taught that epilepsy was not sacred but arose from natural causes in the brain, the prevailing view until the 17th and 18th centuries remained that of demonic possession (Temkin, 1971).

Even the Babylonians recognized that occasionally people with epileptic seizures experienced what today we would call interictal psychoses, also the result of possession (Reynolds & Kinnier Wilson, 2008). Since then and throughout its history, epilepsy has frequently been associated with behavioral and personality disorders. It was, however, in the 19th century, with the growth of institutional and hospital systems in Europe, with their high concentration of patients with chronic epilepsy within them, that perceived associations between epilepsy and mental illness were more widely studied and documented. Prominent at that time were contributions from France (e.g., Morel, 1857) and Germany (e.g., Neumann, 1859), where hereditary concepts of epilepsy and insanity and of progressive intellectual and moral degeneration associated with epilepsy were common. These ideas were also taken up in the English literature for example by Maudsley (1873) and Turner (1907).

Morel’s concept of hereditary degeneration implied progressive cognitive decline within a lifetime, associated with increasing stigmata of degeneration as the disorder became progressively more severe through successive generations. Concepts of moral degeneration, with a Darwinian emphasis, permeated French psychiatry for several decades, and influenced the concept of the epileptic personality in the English literature well into the 20th century (Guerrant et al., 1962).

In addition to persistent disorders of personality, cognition, mood, and behavior associated with institutionalized epilepsy, the terms “epilepsy equivalent” or “masked epilepsy” were commonly used in the late 19th century and early 20th century literature to imply that paroxysmal abnormal behaviors were epileptic seizures, even in the absence of more classical seizures (e.g., Turner, 1907).

The concept of epilepsy as a brain disorder began to take root in the 17th century as Willis (1622–1675) and Sydenham (1624–1689) developed the concept of “nervous diseases” in place of more ancient concepts of the wandering uterus or vapors emanating from the spleen, liver, and stomach, for example to explain hysteria and hypochondria. Nervous diseases were consolidated in the 18th century by Cullen (1710–1790) and Pinel (1745–1826), who translated Cullen’s (1777) textbook and who both developed the concept of “neuroses” as alterations of sensitivity and motion that were not accompanied by fever, inflammation, or structural lesions (Pinel, 1798). The classifications of the Neuroses by both Cullen and Pinel contain a mixture of modern neurology and psychiatry, including convulsions, chorea, tetanus, hysteria, apoplexy, paralysis, and all forms of mental illness (Reynolds, 1990).

In the second half of the 19th century the discipline of neurology gradually emerged from these early concepts of neuropsychiatry, as more and more neuroses, for example, strokes, Parkinson’s disease, and neurosyphilis, were found to be associated with brain lesions. Because convulsions were not commonly associated with brain lesions but seemed to “psychiatrists” or alienists, who were treating them in mental institutions, to be more frequently associated with mental or behavioral disorders, epilepsy remained predominantly within the province of psychiatry or neuropsychiatry until well into the 20th century. At the turn of that century it must be recalled that prior to the evolution of psychodynamic psychiatry from neuropsychiatry, most forms of mental illness were viewed as brain disorders for which as yet no neuropathology had been discovered (Reynolds, 1990).

Epilepsy as a Neurologic Disorder

As the discipline of neurology evolved out of 19th century neuropsychiatry (Hill, 1964; Reynolds & Trimble, 1989), so the neurologic foundations of epilepsy became more firmly established, especially as 19th century vascular theories gradually gave way to 20th century electromagnetic concepts.

Vascular theories

Throughout the 19th century the dominant cerebral view of epilepsy was that it was vascular in origin, especially perhaps due to some temporary insufficiency of blood supply. Even the Babylonians understood that a sacrificed sheep displayed terminal convulsions associated with severe blood loss. The experimental convulsions induced by Sir Astley Cooper (see Todd, 1849) by ligatures to the carotid and especially the vertebral arteries reinforced this approach, which was also emphasized by van der Kolk (1859), who proposed a vascular medullary theory of epilepsy. The experimental studies of Brown-Séquard, who greatly influenced Hughlings Jackson (York & Steinberg, 2006), suggested a reflex vasomotor contraction of cerebral arteries underlying seizures. On the other hand, Marshall Hall (1847) developed a complex reflex theory leading to vascular congestion resulting in seizures.

In the first decade of his writings on epilepsy from 1861 onward, Hughlings Jackson favored wholly vascular theories of epilepsy (Temkin, 1971; York & Steinberg, 2006). Later, as he developed his functional hypotheses of nervous instability and chemical discharge, he continued to invoke a vascular dimension. Thus in his famous 1870 “A study of convulsions” he concluded that the liability to the unilateral convulsions he was describing was “due to persistent changes in the region of the middle cerebral artery and that the paroxysm itself is owing to local vascular contraction.” Later he was less certain whether plugging of vessels leads to anemia or hyperemia, but either way it resulted in altered nutrition underlying nervous instability of gray matter (Jackson, 1873).

Gowers (1881) discussed at length and cast doubt on current vasomotor theories of epilepsy, including the relatively new theory of Hughlings Jackson that local discharges in the brain excited local arterial contraction and that this determined the spread of the discharge.

The continuing prominence of vascular theories is highlighted by Turner (1907), 2 years before the birth of the ILAE. Although the author acknowledges the great interest of Hughlings Jackson in the preparation of the book, there is no mention of Todd or Jackson’s discharge theories but instead a confident statement about a vascular theory of epilepsy based on the latest neuropathologic studies. He believed that there was a special tendency to intravascular clotting and that prior to seizures intravascular coagula became lodged in some of the cortical blood vessels. He acknowledged that this view of epileptic fits caused by plugging of small cerebral arteries was not new as it was advocated by Hughlings Jackson in 1864. However, the essential difference from Jackson was that he now proposed that this only occurred in persons with a congenital hereditary defective development of the nervous system, as indicated by certain neuropathologic peculiarities and clinical stigmata of the disease.

Even Kinnier Wilson, writing in the 1920s before the advent of the electroencephalograph (EEG), while emphasizing the neural discharge theories of Jackson, continues to ponder the possible preceding vascular mechanisms as envisaged by Jackson himself. He came to the view that “the truth is that the gulf between the vascular and the neural is almost as impassable as that between the physical and the psychical. If a vascular factor is involved, I have tried to show the possibility of it being of neural origin” (Kinnier Wilson, 1929).

It was at the 1935 2nd International Congress of Neurology in London, the centenary of Jackson’s birth, that vascular theories were finally laid to rest by Lennox, who could not demonstrate any changes in cerebral blood flow prior to convulsions in 10 patients but who could show abnormal electrical signals prior to seizures in the newly discovered electroencephalography (EEG) of Berger (Reynolds, 2005a).

The concept of neuronal discharges

Our modern electromagnetic theories of epilepsy became widely accepted only from the mid-1930s onward as a result of Berger’s (1929) discovery of the human EEG in the 1920s. However, the concept of neuronal discharges can be traced back over 80 years to the original studies of Robert Bentley Todd (1809–1860). Todd, who is best known for his first description of “Todd’s paralysis,” was Professor of Physiology and Morbid Anatomy at King’s College London and a physician with a special interest in disorders of the nervous system. In his Lumleian Lectures to the Royal College of Physicians of London “on the pathology and treatment of convulsive disorders” (Todd, 1849), he gives a masterly critique of vascular and other currently fashionable theories of epilepsy, all of which he found unsatisfactory. He then develops a radically new electrical theory of epilepsy, influenced by his contemporary in London, Michael Faraday (1791–1867), who was then laying the foundations of our modern understanding of electromagnetism. Todd conceived of nervous polarity (force) generated in nervous centers (gray matter) and compared this with the polar force of voltaic electricity developed in the galvanic battery. As a pioneering neurohistologist, he brilliantly foresaw each nerve vesicle (cell) and its related fibers, that is, neuron of later terminology, as a distinct apparatus for the development and transmission of nervous polarity. He applied these concepts to epilepsy in which he envisaged a periodic rise in electrical tension in gray matter, which could, at a certain threshold, result in a sudden change in polar state, leading to a seizure discharge, comparable to the spark from a battery or lightening. He emphasizes: “These periodical evolutions of the nervous force may be compared to the electrical phenomena described by Faraday under the name of ‘disruptive discharge.’” The rise in electrical tension Todd attributes to abnormal nutrition of the brain from many different causes. Finally, he supports his views by electrical experiments in the rabbit using Faraday’s newly discovered magnetoelectric machine (Reynolds, 2004, 2005b).

Todd’s brilliant electrical observations and insights fell on deaf or uncomprehending brains, perhaps in part because he died in 1860 just before the establishment of the National Hospital for the Paralysed and Epileptic, where vascular theories of epilepsy influenced by Brown-Séquard and Hughlings Jackson continued to dominate thinking, as described previously. Later, however, from 1870 onward Jackson began to develop his own now famous theories of sudden excessive discharges of gray matter in epilepsy. However, for Jackson, a brilliant philosopher physician greatly influenced by the scientifically untrained evolutionary philosopher Herbert Spencer (1820–1903), the word “discharge” had a very different chemical, but not electrical, meaning. For Spencer and Jackson to “discharge” by releasing chemical energy to greater or lesser degrees was the normal function of all nerve cells. In epilepsy there was apparently excessive discharge in gray matter on a background of instability, perhaps due to some imbalance between normal anabolism and catabolism (Reynolds, 2007). In his extensive writings on epilepsy and other neurologic subjects, Jackson never strays into physics or electromagnetism, perhaps because it was beyond his range. Indeed he states in one footnote: “I have no theory as to “nerve force” or force of any kind.” Why Jackson never referred to Todd’s electrical theories or his 1849 Lumleian Lectures in Jackson’s (1890) own Lumleian Lectures on the same subject 41 years later is unclear, although he does acknowledge Todd’s descriptions of postepileptic paralysis.

Cortical localization and the discovery of the electroencephalograph (EEG)

In 1870, Fritsch and Hitzig discovered the motor cortex, soon confirmed by David Ferrier (1876) who developed the concept of cortical localization inspired by the clinical observations of his colleague, Hughlings Jackson, who relied on Ferrier in scientific matters. In the same decade Caton (1875) first reported electrical potentials from the cortex of animal species, mainly rabbits, but also cats and monkeys. Relatively little notice was taken of this important observation until Berger (1929) first described the human EEG 54 years later. Meanwhile, in 1906, Cajal and Golgi received the Nobel Prize for their histologic development of the “neuron doctrine,” already glimpsed more than 50 years earlier by Todd.

Hans Berger, who was born in 1873, when Caton was studying animal electrical potentials, succeeded Binswanger as head of the University Psychiatric Hospital in Jena, where he spent 27 years studying the electrical activity of cats and dogs, and in the last 5 years, human volunteers (mainly himself and his son) and patients, before he first published his discovery of the human EEG in 1929. By 1933 he had already recorded in patients with epilepsy: (1) the interictal resting EEG, (2) the EEG response to hyperventilation, (3) the postictal EEG immediately following a tonic–clonic seizure, (4) the ictal-EEG of focal motor seizures, and (5) the ictal-waves (but not the spikes) of an absence attack (Gloor, 1974; Karbowski, 1990). From 1935 onward the clinical neurophysiology of epilepsy progressed rapidly led by Lennox and the Gibbses in Boston, who in that year published and also presented at the London International Neurology Congress EEG recordings of a tonic–clonic (grand mal) seizure and the three cycles per second spike and wave activity of “petit mal” absences (Gibbs & Gibbs, 1952; Reynolds, 2005a).

It was only in retrospect that the various electrical “discharges” recorded in the human EEG during seizures were equated incorrectly with Jackson’s view of excessive “discharges” of gray matter by those who seemed unaware that Jackson was referring to an increase of the normal chemical activity of neurons based on an imbalance of anabolism and catabolism, and were also ignorant of the fact that a generation before Jackson, Todd had described a truly electrical concept of seizure discharges in epilepsy based on Faraday’s concepts of the polar forces of electromagnetism (Reynolds, 2007).

Epilepsy and Psychiatry in the 20th century

With the development of clinical EEG in the 1940s and 1950s (Gibbs & Gibbs, 1952; Karbowski, 1990), including the cortical recordings and stimulation studies of Penfield and Jasper (1954), the evidence grew that various “epileptiform discharges” on the EEG, for example, spike and wave, multiple spikes, focal spikes, and sharp waves, none of which even now are entirely understood, provided a reasonably sound basis for distinguishing epileptic seizures (epilepsy “per se”) from nonepileptic attacks, especially with the later advent of video telemetry in the 1970s and 1980s.

At the same time, some of the confusion surrounding the nature of some paroxysmal behavioral and psychological, including psychotic, disorders began to be dispelled by increasing precision in electroencephalographic cortical localization and the development of the concepts of temporal lobe and, later, frontal lobe epilepsy (e.g., Gibbs, 1951). Furthermore, EEG studies in psychotic epileptic patients revealed that in at least a subgroup of them epileptiform discharges disappeared as the patient became psychotic only to reappear as the psychosis improved and seizures became more prominent, that is, the so-called “forced normalisation” of Landolt (1958). As Landolt noted, these electrical phenomena were often associated with changes in antiepileptic medication but could be spontaneous.

By the 1960s, therefore, the stage had been set for the increasingly precise distinction between preictal, ictal, postictal, and interictal psychiatric disorders, which led to the radically new WHO classification of epilepsy “per se” as a neurologic (not mental) disorder, and epilepsy with additional mental syndromes, the province of psychiatry. At that time also, epidemiologic studies were beginning to reveal a clearer picture of the relative distribution of uncomplicated epilepsy and epilepsy complicated by psychological disorders. Pond and Bidwell (1960) surveyed 14 general practices in the South East of England and concluded that 29% of patients with epilepsy had psychological difficulties. In adults with epilepsy in Iceland, Gudmundsson (1966) found that about 50% had psychological or personality problems but emphasized that this was mainly due to underlying brain diseases. Likewise in school children with epilepsy on the Isle of Wight, Graham & Rutter (1970) found a much higher prevalence of psychiatric disorders if the epilepsy was associated with brain lesions or mental handicaps.

Another important contribution was that of Slater et al. (1963), who clarified the interictal schizophrenia-like psychoses of epilepsy. This was a phenotype not linked with a genotype. Slater, being a geneticist, found no familial associations with this psychosis, which resembled, but was not identical to schizophrenia. He noted the associations with temporal lobe epilepsy, and considered the syndrome an organic replica of the psychiatric disorder in the absence of epilepsy. By this time, therefore, clear links were implied between certain forms of epilepsy and more severe psychopathology, in particular schizophrenia-like and paranoid states, and in some cases these could be linked to EEG findings. Taylor (1972) took this further, noting in surgically resected specimens the underlying temporal lobe pathology in people with these psychoses of epilepsy. Flor-Henry (1969) observed associations between left-sided temporal lobe epilepsy and schizophrenia-like presentations, and right-sided pathology with manic-depressive disorder, which reinforced the significance of laterality of pathology in the brain when investigating neuropsychiatric disorders. An important outcome of these studies in epilepsy was to stimulate a search for temporal lobe and lateralized pathology in schizophrenia and then bipolar disorders, which over the ensuing 40 years has provided a wealth of data supporting the biologic basis of such psychopathologies (Trimble, 1991, 1996).

Also during the middle of the 20th century, neuroanatomists such as Papez (1937) and MacLean (1976) were defining the cerebral circuitry, especially the limbic system, which modulated human emotion. Central to this were the hippocampus and amygdala and related anatomic structures. MacLean had been stimulated by observations of patients with temporal lobe epilepsy. Neuropathologic studies throughout much of the 19th and 20th centuries, reinforced more recently by modern neuroimaging techniques, have confirmed the importance of the medial temporal structures and their associated connectivity not only in the psychiatry of epilepsy, but also in some psychological disorders without epilepsy (Trimble, 1996; Heimer et al., 2007). A further important catalyst was the work of Waxman & Geschwind (1975) who defined the interictal behavior syndrome of temporal lobe epilepsy, central features of which included alterations of the personality in the direction of religiosity and spirituality, irritability, some alteration of sexual inclination, and hypergraphia. This was the beginning of the discipline of behavioral neurology, a term introduced by Geschwind.


At the time of the foundation of ILAE a century ago, neurology was rapidly evolving as a new discipline based on the 19th century triumphs of neuropathology and the new clinicoanatomic method, associated with a rather rigid and, in retrospect, over simplistic view of functional localization. At the same time, the emergence and growth of psychodynamic psychiatry led by Freud, Jung, and many others contributed to the progressive separation of neurology and psychiatry with a contracting neuropsychiatry sitting uncertainly in the middle awaiting hopefully evidence of the cerebral basis of many common forms of mental illness. Epilepsy, which has been viewed as a mental disorder for millennia, continued to straddle both neurology and psychiatry, partly because of the absence of neuropathology in many patients and partly because of the high incidence of psychological disorders associated with it.

It took until the middle of the last century, following the development of electromagnetic concepts and electroencephalography, for epilepsy per se to be defined as a neurologic disorder, releasing and refining its associated pre-, post-, and interictal psychological disorders, and contributing more recently to a renaissance of neuropsychiatry. In the latter half of the 20th century, many developments in neuroscience have revealed that the brain is much more complex than our 19th century predecessors anticipated. A single structural lesion may have remote functional effects far from the pathologic site. As Jackson implied, localizing a lesion is not the same thing as localizing a function. At the same time it is now apparent that cerebral function is profoundly influenced by psychological and social factors, and nowhere is this more obvious than in the field of epilepsy (Reynolds, 1990;Trimble, 1996).

A century ago, neurology and psychiatry were diverging and epilepsy remained to some extent in both camps. As we approach the centenary of ILAE, neurology and psychiatry have been converging again, led in many respects by epilepsy, which has provided several useful models of mental illness, a window on brain function, and a bridge between the two disciplines (Reynolds & Trimble, 1989).


Conflict of interest: The authors declare no conflicts of interest.