Hermann Ebbinghaus

Introduction

Hermann Ebbinghaus (1850-1909) was a German psychologist who carried out the first systematic experimental study of human memory and, in doing so, helped break the assumption that only sensation and perception could be measured in the laboratory. Working almost entirely on himself, he invented nonsense syllables as stimulus material, developed the savings method for quantifying retention, and described the curve of forgetting that still bears his name. His 1885 monograph arrived at a moment when Wilhelm Wundt held that higher mental processes lay beyond experimental reach, and its success is often credited with opening that territory to measurement. Ebbinghaus later turned to colour vision, to the mental testing of schoolchildren, and to writing textbooks, but his reputation rests on the memory work of 1879 to 1885. He is also associated with the size-contrast illusion known as the Ebbinghaus-Titchener circles, though historians disagree about whether he was its originator.

Early Life and Education

Ebbinghaus was born on 24 January 1850 in Barmen, in the Prussian Rhineland, into a Lutheran merchant family; his father Carl was a successful businessman and the household was comfortable enough to support a long university career. He attended the local gymnasium until he was seventeen and then enrolled at the University of Bonn in 1867, with later periods at Berlin and Halle. His early interests lay in history and philology, and he moved gradually toward philosophy, a shift interrupted in 1870 when he volunteered for service in the Franco-Prussian War. He returned to Bonn in 1871 and completed a doctorate in philosophy on 16 August 1873 with a dissertation on Eduard von Hartmann's philosophy of the unconscious.

The conventional story that Ebbinghaus was trained or inspired by Wundt is misleading. He came to psychology largely self-taught, and his decisive intellectual debt was to Gustav Fechner: while living in Paris in the late 1870s he picked up a secondhand copy of Fechner's Elemente der Psychophysik, whose project of measuring the relation between physical stimulus and mental sensation gave him the model he had been looking for. Between his doctorate and his first academic post he spent several years in England and France supporting himself as a tutor, conducting an early and unsuccessful attempt at memory experiments with meaningful material. The failure was instructive: meaningful text carried too many prior associations to allow clean measurement.

That failure is the whole design in miniature. Ebbinghaus solved the problem of prior knowledge by removing meaning from the stimulus rather than by controlling for it statistically, and every strength and weakness of his work follows from that choice. It made the measurement clean and the phenomenon narrow.

Academic and Professional Career

In 1880 Ebbinghaus submitted his memory research to the philosophy faculty at Berlin and was appointed a Privatdozent, an unsalaried lecturer dependent on student fees. Über das Gedächtnis appeared in 1885 and brought swift recognition; he was promoted to a salaried extraordinary professorship in 1886 and established a psychological laboratory at Berlin. In 1890, with the physicist Arthur König, he founded the Zeitschrift für Psychologie und Physiologie der Sinnesorgane, which became the leading German-language outlet for work that Wundt's own journal would not take. In 1894 he accepted a full professorship at Breslau, where he founded a second laboratory and carried out a survey of the mental capacities of schoolchildren. In 1905 he moved to the University of Halle to succeed Alois Riehl, and he died there of pneumonia on 26 February 1909, aged fifty-nine.

Major Contributions to Psychology

Nonsense syllables and controlled measurement

Ebbinghaus's central methodological innovation was the nonsense syllable: a consonant-vowel-consonant string such as ZOF or KEX, stripped of meaning so that prior knowledge could not contaminate the measurement. He generated some 2,300 of them and assembled them into lists that he read aloud at a fixed pace, with a metronome-like regularity, at the same hour each day, with rest periods built in to limit fatigue. The point was not that such syllables resemble anything a person actually wants to learn, but that they made the measurement clean. The same instinct produced the sentence-completion test he devised for the Breslau school survey of 1897, in which children filled in deleted words; it is often described as one of the first successful tests of a higher mental process and an ancestor of later verbal intelligence scales.

The forgetting curve and the savings method

He measured retention indirectly. Having learned a list to the point of one flawless recitation, he waited a fixed interval and relearned it, recording how many repetitions or how much time the second attempt saved. The savings score was the ratio of that saving to the original effort. Applying this across intervals from twenty minutes to thirty-one days, he found that forgetting is steepest immediately after learning and then flattens: savings fell to roughly 58 percent after twenty minutes, about 34 percent after a day, and about 21 percent after a month. The reported bump at twenty-four hours, which he noted himself, is now usually attributed to sleep-dependent consolidation.

Colour vision, textbooks, and the shape of the discipline

After 1885 Ebbinghaus never returned to sustained memory experimentation. He published a theory of colour vision in 1893, gave popular and well-attended lectures, and devoted much of his last decade to the Grundzüge der Psychologie and the shorter Abriss der Psychologie. The Abriss opens with the aphorism usually rendered as "psychology has a long past, but only a short history," and it makes his methodological commitment explicit: a single obscure measurement of the two-point threshold or the just-noticeable difference between weights, he wrote, had advanced psychology more than all the definitions and classifications from Aristotle to Hobbes. Contemporaries also noted that despite his celebrity as a lecturer he trained few students who carried his programme forward; the memory tradition he founded was largely continued by Georg Elias Müller and Müller's assistants rather than by his own pupils.

Spaced versus massed practice, and the learning curve

Ebbinghaus compared cramming a list into one session against distributing the same repetitions across days, and found distributed practice markedly more efficient for a fixed level of mastery. He also showed that relearning is faster than original learning, that the time needed to learn a list grows steeply with its length, and that extra repetitions beyond the first perfect recitation, so-called overlearning, slow subsequent forgetting. His observations on the advantages of items at the beginning and end of a list anticipate what later writers called the serial position effect, although the primacy and recency labels were supplied by others.

Major Publications

Über die Hartmannsche Philosophie des Unbewussten (1873), his doctoral dissertation on Hartmann's philosophy of the unconscious.
Über das Gedächtnis: Untersuchungen zur experimentellen Psychologie (1885), published in English as Memory: A Contribution to Experimental Psychology.
Grundzüge der Psychologie (1897-1902), the general textbook on which much of his contemporary reputation rested.
Abriss der Psychologie (1908), his short and widely read summary, famous for the remark that psychology has a long past but only a short history.

Impact and Legacy

Ebbinghaus demonstrated that memory, learning, and forgetting could be brought into the laboratory, and the methodological template he built, meaningless materials, controlled conditions, quantified dependent variables, shaped a century of verbal-learning research. His editorials and textbooks also gave German psychology an institutional voice independent of Leipzig. Contemporary estimates of his importance were high: on hearing of his death, Edward Titchener remarked that had Ebbinghaus lived, his standing would have rivalled Wundt's.

The criticisms are equally substantial. He was his own only subject, which secures experimental control at the cost of external validity and leaves the results open to the charge that he knew exactly what he was testing and had accumulated years of interfering practice on similar lists. The data were collected in two concentrated bursts, in 1879–80 and 1883–84, rather than continuously, and the published figures are averages over different numbers of lists at different intervals — a degree of aggregation that hides how variable the underlying sessions were.

The materials were deliberately artificial, and the premise behind them did not survive scrutiny. Nonsense syllables are not meaningless: they acquire associations almost immediately, they differ systematically in how associable they are, and later work — notably on association value — showed that the supposedly controlled stimulus varies in exactly the property it was designed to eliminate. Ebbinghaus removed meaning from the list and meaning came back anyway, inside the syllables themselves.

The savings method carries a related assumption. It treats the time or trials saved on relearning as a measure of what was retained, but a person relearning a list is not in the same state as when they first learned it: they have changed strategies, they may recognise fragments without being able to produce them, and recognition and recall are known to dissociate. Savings is an elegant solution to the problem of measuring something invisible, and it is not a pure measure of it.

Historians also argue that the prestige of his method diverted memory research away from semantic organisation, mnemonics, and the qualitative changes memories undergo, and toward rote list learning. In the 1890s he clashed publicly with Wilhelm Dilthey, who argued that an explanatory, experiment-driven psychology was the wrong model for a discipline that should describe and analyse lived mental life; Ebbinghaus replied that quantitative explanation was precisely what made psychology a science rather than a branch of literary reflection.

Two modern findings qualify the textbook picture without overturning it. First, the shape of the curve survives: Jaap Murre and Jaap Dros repeated the original procedure with nonsense syllables and the savings method and recovered a curve closely resembling the 1885 one, an unusual outcome for a nineteenth-century single-subject study. Earlier work by Finkenbinder with fourteen participants had found slower forgetting but the same overall form. Second, the equation is less settled than diagrams suggest. When David Rubin and Amy Wenzel fitted more than a hundred candidate functions to some two hundred published retention datasets, several families, logarithmic, power, exponential, and hyperbolic, described the data about equally well, so the curve's shape is robust while its algebraic form is not. The popular claim that we forget 70 percent of new material within a day is a misreading: a 34 percent savings score means relearning took about two-thirds of the original time, not that two-thirds of the content had vanished. The practical prescriptions drawn from his work, spaced rather than massed review and retrieval practice instead of rereading, have since been supported independently, including by a 2006 meta-analysis of distributed-practice studies, and they no longer depend on the details of his curves.

Two later results extend that verdict in different directions. On the practical side, Latimier, Peyre and Ramus (2021, Educational Psychology Review 33(3):959–987) pooled the studies that spaced retrieval practice rather than spacing exposure alone, drawing on 29 experiments, and found a very large advantage over massed retrieval together with almost no difference between expanding and uniform schedules: the core prescription is robust, but most of the elaborate guidance about how exactly to arrange the intervals has little behind it, and how large the benefit looks depends heavily on which comparisons are pooled. On the mechanistic side, forgetting now has a molecular description Ebbinghaus could not have imagined. Terceros, Chen, Harada, Rajasethupathy and colleagues reported in Nature (2025, 649:1254–1263) that memories in mice are stabilised by a cascade of transcriptional regulators acting on different timescales — CAMTA1 holding a memory over days, then the transcription factor TCF4, then the histone methyltransferase ASH1L holding it over weeks — none of which is required to form the memory in the first place, all of which are required to keep it, with CRISPR knock-outs abolishing the memory at the stage at which each regulator normally acts. What selected which memories entered the cascade was, in their task, how often the experience recurred. On that account forgetting is not the slow erosion of an engraved trace but the failure to be taken up by a maintenance programme: the same curve, drawn for a very different reason.

His influence on practice has been just as durable as his influence on method. The sentence-completion format he designed for the Breslau schoolchildren became a standard component of group verbal tests and fed into twentieth-century intelligence testing, while the forgetting curve and the spacing principle are now standard fixtures of study-skills advice and of the scheduling algorithms built into spaced-repetition software. That applied afterlife comes with a warning its originator would have recognised: the curve was measured on meaningless material learned by rote, so it describes how fast isolated, unconnected items decay rather than how well a student who understands a subject will retain it.

Learn More

Wilhelm Wundt — whose claim that higher processes were beyond experiment Ebbinghaus refuted
Gustav Fechner — whose psychophysics supplied the model he applied to memory
Ulric Neisser — who argued for the ecological study of memory that the verbal-learning tradition neglected
Serial position effect — the list phenomenon his observations anticipated

References

Ebbinghaus, H. (1913). Memory: A Contribution to Experimental Psychology. Teachers College, Columbia University. (Original work published 1885)