When Hans Berger first recorded the human electroencephalogram, he proved that electrical activity of the living brain could be measured. The next question was far more clinically important: could particular electrical patterns identify epilepsy?
During the mid-1930s, the answer began to become yes.
A small group of researchers in the United States — including Frederic Andrews Gibbs, Erna Leonhardt Gibbs, William Gordon Lennox, Hallowell Davis, Herbert Jasper and others — transformed EEG from an experimental recording technique into a method for studying seizure disorders. Their work revealed that different clinical seizure types could be associated with different electrical patterns, and that abnormal epileptiform activity could appear even between seizures.
This was the beginning of EEG-based epilepsy classification.
But the history needs one important correction. The first published human epileptiform spikes were reported in 1934 by Fisher and Lowenback. Gibbs and his collaborators did not make every EEG discovery that is sometimes attributed to them. Their distinctive achievement was to build systematic links between electrographic abnormalities and clinical seizure types, particularly through the 3-per-second spike-and-wave pattern of absence seizures and the recognition of focal interictal spikes.
From Berger’s rhythms to epilepsy
Berger’s first EEG work concentrated largely on normal rhythms and changes associated with attention, sleep and altered states.
He did record striking abnormalities in people with epilepsy, including patterns during petit-mal attacks, but he was cautious about publishing some of them because eye movements and facial muscle activity could produce large artefacts.
That caution was understandable. Early EEG had no established library of normal and abnormal patterns, and the distinction between cerebral activity and artefact was still being learned.
Once Berger’s technique had been independently confirmed in 1934, researchers elsewhere began applying EEG directly to epilepsy.
The first published epileptiform spikes
Historical accounts generally credit A. S. Fisher and H. Lowenback with publishing the first human epileptiform spikes in 1934.
This milestone is often omitted in simplified histories that jump directly from Berger to Gibbs and Lennox.
Fisher and Lowenback demonstrated sharply contoured abnormal potentials in people with epilepsy, helping establish that EEG could record electrical events qualitatively different from ordinary background rhythms.
The discovery was important, but an isolated abnormal waveform did not yet create a clinical classification system.
That required the next stage: comparing electrical patterns with the seizure a patient was actually having.
Frederic Andrews Gibbs
Frederic Andrews Gibbs was born in 1903 and trained in medicine in the United States.
Before becoming famous for EEG, he worked with William Lennox on cerebral blood flow during seizures. Their experiments showed that cerebral blood flow increased during epileptic attacks.
By the early 1930s, Gibbs had moved deeply into electroencephalography. At Harvard he worked with Lennox, physiologist Hallowell Davis and EEG technician Erna Leonhardt, who later became his wife.
Gibbs became one of the most influential figures in the early clinical application of EEG to epilepsy.
Erna Leonhardt Gibbs: more than a footnote
Erna Leonhardt Gibbs is frequently reduced in older histories to ‘Gibbs’ wife’ or ‘the technician’.
That description understates her role.
Erna was deeply involved in EEG recording, technical development, pattern recognition and later publications. She worked alongside Frederic Gibbs for decades and became an important electroencephalographer in her own right.
The history of early EEG depended heavily on skilled technical recording. Electrodes had to be placed consistently, amplifiers adjusted, artefacts recognised and long paper traces interpreted. In an era before digital EEG, those practical skills were central to the science itself.
Later work by Frederic and Erna Gibbs would help describe several important EEG patterns, including slow spike-and-wave and the pattern later called hypsarrhythmia.
William Gordon Lennox
William Lennox was already an established epilepsy physician when EEG arrived.
He had a longstanding interest in seizure classification, heredity, treatment and cerebral physiology.
EEG gave Lennox something clinical epileptology had never possessed before: a physiological measurement that could be compared with the outward form of a seizure.
His collaboration with the Gibbs group helped make this electroclinical approach central to epilepsy research.
Hallowell Davis and the Harvard laboratory
Hallowell Davis was a physiologist whose laboratory at Harvard became one of the earliest American centres for human EEG.
After Berger’s findings were confirmed, Davis and colleagues developed equipment and recording methods capable of studying patients systematically.
The collaboration of physiology, neurology and technical engineering was crucial. EEG could not become clinically useful through medical observation alone; it required reliable amplifiers, recording equipment and trained operators.
1935: petit mal becomes visible on EEG
In 1935, Gibbs, Davis and Lennox published one of the foundational observations in epilepsy EEG.
During attacks then called petit mal, they recorded a rhythmic pattern of approximately three spike-and-wave complexes per second.
Today, the classic generalised 3-Hz spike-and-wave pattern is strongly associated with typical absence seizures.
The historical importance of the finding was enormous.
A brief staring spell lasting only a few seconds could now be accompanied by a reproducible electrical event. An attack that might previously have been dismissed as inattention, daydreaming or unusual behaviour could be linked directly with a characteristic change in brain electrical activity.
Why spike-and-wave was so important
The pattern was not simply ‘a spike’.
It consisted of a repeating sequence in which a sharp spike was followed by a slower wave, occurring rhythmically across much of the scalp.
That combination helped demonstrate that epileptic electrical activity could have organised temporal structure.
It also provided one of the earliest clear electroclinical correlations: a particular type of clinical attack and a particular EEG pattern repeatedly appeared together.
This helped move epilepsy classification away from appearance alone.
A seizure could now be classified by two kinds of evidence
Before EEG, physicians relied on what the patient experienced and what witnesses observed.
Did the person collapse? Did they stare? Did one limb jerk? Was consciousness lost? Was there an aura?
EEG added another layer.
Now researchers could ask whether the clinical event was accompanied by a generalised rhythmic discharge, a focal spike, a temporal abnormality or another pattern.
The seizure’s electrical signature became part of its identity.
Interictal abnormalities changed diagnosis
An even more consequential discovery was that abnormal epileptiform activity could occur between clinical seizures.
A patient did not always have to experience an attack while connected to the EEG machine for the recording to provide useful information.
Interictal spikes and sharp waves could reveal a tendency toward epileptic activity during apparently normal periods.
This fundamentally changed the practicality of EEG as a diagnostic tool.
1936: focal interictal spikes
In 1936, Frederic Gibbs and Herbert Jasper described focal interictal epileptiform spikes.
These abnormalities differed from the widespread 3-Hz spike-and-wave activity associated with petit mal.
A focal spike suggested that abnormal electrical activity could arise from a more restricted brain region.
This provided electrophysiological support for the focal-versus-generalised distinction that had been developing clinically since the nineteenth century.
The focal–generalised divide becomes electrical
Hughlings Jackson had argued from clinical seizure spread that some attacks began locally.
EEG now gave neurologists another way to examine that distinction.
Some patients showed focal epileptiform abnormalities. Others showed bilaterally synchronous generalised patterns.
The difference was not absolute or simple, and later decades would reveal many exceptions and network complexities. But in the 1930s, EEG powerfully reinforced the concept that not all epilepsies shared one electrical organisation.
Grand mal, petit mal and psychomotor epilepsy
Gibbs, Erna Gibbs and Lennox compared EEG patterns across the major seizure categories used at the time.
These included grand mal, petit mal and what came to be called psychomotor epilepsy.
The terminology is now obsolete, but historically the work was important because the researchers demonstrated that clinically different seizure types could have different electrographic characteristics.
The EEG was becoming a classification instrument rather than merely a detector of abnormality.
Psychomotor epilepsy and temporal patterns
In 1937, Gibbs and Lennox used the term psychomotor epilepsy for seizures involving mental, emotional, autonomic and motor phenomena associated with characteristic EEG abnormalities.
Many such cases would later be understood within temporal lobe epilepsy.
Some patients with these attacks had previously been diagnosed as having hysterical or psychiatric disorders.
EEG therefore helped move certain episodic behavioural and experiential phenomena back into neurological classification.
EEG did not remove diagnostic uncertainty
The early pioneers were enthusiastic about EEG, sometimes more enthusiastic than later evidence justified.
Not every abnormal slow wave was epileptiform. Not every person with epilepsy showed an abnormal routine EEG. Some patterns were overinterpreted.
Historical review of Gibbs and Lennox’s work notes that excessive significance was sometimes attached to non-specific slowing and to ideas such as ‘cerebral dysrhythmia’.
These mistakes were part of learning how to interpret a completely new physiological test.
The idea of ‘cerebral dysrhythmia’
Early EEG researchers often described abnormal rhythms using the language of dysrhythmia.
The term suggested that epilepsy reflected a disorder of cerebral rhythm.
For a time, ‘cerebral dysrhythmia’ became broader than epilepsy itself and was applied to behavioural, psychiatric and neurological problems in ways that later proved insufficiently specific.
This is an important caution in the history of diagnostic technology: once a new test can produce striking traces, there is a temptation to make the trace explain more than the evidence supports.
Erna Gibbs and the labour behind the trace
Early EEG interpretation required hours of direct visual inspection.
There were no automated spike detectors, no digital montages and no computer storage.
Paper records could extend for metres. Technicians marked clinical events, adjusted equipment and distinguished cerebral patterns from electrical and biological artefacts.
Erna Gibbs’ role belongs within this technical history. Her work illustrates how the emergence of clinical EEG depended on collaborative teams, not only the physicians whose names appeared first in publication titles.
The Atlas of Electroencephalography
Gibbs and Lennox later consolidated early EEG knowledge in major reference works.
In 1941 they published an influential Atlas of Electroencephalography, bringing together normal and abnormal EEG patterns and helping standardise visual interpretation.
An atlas mattered because EEG interpretation depends heavily on pattern recognition.
Clinicians needed shared examples and terminology if a tracing recorded in one hospital was to mean the same thing in another.
EEG begins to shape epilepsy syndromes
The 1935 finding of 3-Hz spike-and-wave provided an early model for syndrome recognition.
Later EEG patterns would become associated with other epilepsy syndromes: slow spike-and-wave, hypsarrhythmia and characteristic focal discharges among them.
In 1938 and 1939, Gibbs and colleagues described slow spike-and-wave activity they called the ‘petit mal variant’, a pattern later associated with the syndrome that came to be known as Lennox–Gastaut syndrome.
In 1952, Frederic and Erna Gibbs described the chaotic high-voltage pattern later known as hypsarrhythmia in infants with spasms.
These developments showed how EEG could help define clinically meaningful groups of patients before the underlying causes were known.
From diagnosis to localisation
Focal EEG abnormalities also had surgical implications.
If repeated interictal or ictal discharges arose from one region, perhaps that region could guide a surgical investigation.
Gibbs later collaborated with neurosurgeons in Chicago and used EEG to help select and localise patients for temporal lobe surgery.
This history belongs more fully to later pages on temporal lobectomy and epilepsy surgery, but it began with the recognition that a spike could contain anatomical information.
Sleep becomes an activation method
Researchers discovered that some epileptiform abnormalities became more visible during sleep.
Gibbs later emphasised sleep EEG in patients with psychomotor seizures, where temporal epileptic abnormalities could emerge more clearly.
This changed EEG practice. Rather than treating sleep as an inconvenience during recording, laboratories learned to use changes in brain state as a way of increasing diagnostic yield.
Classification became electroclinical
The most important conceptual change was that epilepsy classification could no longer be based solely on how a seizure looked.
Clinical signs and EEG patterns had to be interpreted together.
A staring episode plus 3-Hz generalised spike-and-wave meant something different from a behavioural arrest associated with focal temporal spikes.
This principle — electroclinical correlation — became one of the foundations of twentieth-century epileptology.
The limits of the early classifications
The 1930s researchers were working with categories that modern epilepsy medicine has substantially revised.
Petit mal did not map perfectly onto every modern absence syndrome. Grand mal grouped together events that can arise through different mechanisms. Psychomotor epilepsy mixed several focal seizure presentations.
EEG therefore improved classification without completing it.
Later generations would repeatedly revise seizure terminology as video-EEG, intracranial recording, imaging and genetics revealed greater complexity.
Why the epileptiform spike changed medicine
A spike is brief.
Its historical impact was not.
It showed that abnormal epileptic activity could leave a measurable electrical signature.
That signature could appear during a seizure or between seizures. It could be widespread or focal. It could help distinguish syndromes. And in some cases it could point toward a region of abnormal brain function.
EEG therefore converted part of epilepsy from something witnessed only through behaviour into something that could also be recorded physiologically.
Who deserves credit?
The answer is deliberately plural.
Berger established human EEG and had recorded epileptic abnormalities himself.
Fisher and Lowenback published early epileptiform spikes in 1934.
Gibbs, Davis and Lennox produced the landmark 1935 electroclinical description of 3-Hz spike-and-wave in petit mal.
Gibbs and Jasper described focal interictal spikes in 1936.
Erna Gibbs played a major role in the technical and interpretive work that supported the developing field.
Herbert Jasper, Hallowell Davis and many others helped establish EEG laboratories, methods and terminology.
The history is therefore not one discovery by one scientist. It is the rapid construction of a new clinical language.
The central historical lesson
Hans Berger made the electrical activity of the human brain visible.
The epilepsy researchers of the 1930s learned how to read some of what that activity meant.
The key advance was not simply finding a sharp waveform. It was demonstrating reproducible relationships between EEG patterns and clinical seizure types.
That transformed EEG from an experimental curiosity into a diagnostic instrument and laid the foundations for electroclinical epilepsy classification.
Sources and further reading
American Epilepsy Society. A Brief History of EEG. In: Electroencephalography: An Introductory Text and Atlas. 2016.
Stone JL, Hughes JR. Early history of electroencephalography and establishment of the American Clinical Neurophysiology Society. Journal of Clinical Neurophysiology. 2013.
Magiorkinis E et al. Highlights in the History of Epilepsy: The Last 200 Years. Epilepsy Research and Treatment. 2014.
Stone JL, Hughes JR. Frederic Gibbs and his contributions to epilepsy surgery and electroencephalography. Neurosurgery. 2011.
The evolution of the concepts of seizures and epilepsy: What’s in a name? Epilepsia Open. 2020.
Pioneering concepts in epileptology: the cerebral dysrhythmia of Frederic Gibbs and William Lennox. Epilepsy & Behavior. 2011.
Psychomotor seizures, Penfield, Gibbs, Bailey and the development of anterior temporal lobectomy: a historical vignette. 2010.
Hypsarhythmia or Hypsarrhythmia? Pediatric Neurology Briefs. 2015.
Historical evidence checked: September 2026.