Endel Tulving

Introduction

Endel Tulving (26 May 1927 – 11 September 2023) was an Estonian–Canadian experimental psychologist who did more than anyone else to determine how many kinds of memory there are. His 1972 distinction between episodic memory — memory for personally experienced events — and semantic memory — memory for facts and general knowledge — overturned the then-current assumption that long-term memory is a single system, and it reorganised the study of memory in cognition, neuropsychology and cognitive neuroscience.

He also formulated the encoding specificity principle, the idea that a retrieval cue works to the extent that it reinstates what was encoded with the target, and he spent the second half of his career tying memory systems to forms of consciousness: procedural memory is anoetic (non-knowing), semantic memory noetic (knowing), and episodic memory autonoetic (self-knowing). A 2002 survey published in Review of General Psychology ranked him the 36th most cited psychologist of the twentieth century.

Early Life and Education

Tulving was born in Petseri, Estonia, the son of a small-town judge. He was an excellent student who found the curriculum dull — he later said he assumed everything interesting in science had already been discovered — and put his energy into track and field, chasing his personal bests rather than his marks.

The Second World War ended his adolescence. He was nineteen when he left Estonia for Germany, where he lived in a displaced persons camp and worked as a translator for the United States army. It was in that camp that he met Ruth Mikkelsaar, then fourteen, who would become his wife and, much later, an elected member of the Royal Canadian Academy of Arts. In 1949 the couple emigrated to Canada.

Supporting a wife and child, Tulving worked in construction and studied at the University of Toronto, taking his BA in 1953 and his MA in 1954. He then went to Harvard for doctoral work in experimental psychology under Stanley Smith Stevens — the psychophysicist who had founded Harvard's Psycho-Acoustic Laboratory — completing his PhD in the mid-1950s, though sources differ between 1956 and 1957 for the year of conferral.

Academic and Professional Career

Tulving joined the University of Toronto faculty in 1956 and stayed, apart from a period as visiting professor at Yale from 1970 to 1974, for the rest of his career. He chaired the department from 1974 to 1980. He retired in 1992 and moved to the Rotman Research Institute at the Baycrest Centre in Toronto, where he held the Anne and Max Tanenbaum Chair in Cognitive Neuroscience, and he also spent time as a visiting professor at the Center for Neuroscience at the University of California, Davis. He resigned from Rotman to care for Ruth, who died in 2012.

He published more than 300 research articles and chapters and maintained an h-index above 120, a figure extraordinary for a researcher whose main instrument was the word list. He died on 11 September 2023 at a nursing home in Mississauga, Ontario, from complications of a stroke, aged 96.

Major Contributions to Psychology

Episodic and semantic memory

In a 1972 book chapter, Tulving argued that the long-term memory then being modelled as a unitary store had to be split. Semantic memory holds general knowledge — that Paris is the capital of France, what a word means — with no record of where or when it was acquired. Episodic memory holds personally experienced events, bound to a time and a place, and retrieved with a distinctive subjective quality of re-experiencing. He elaborated the case in Elements of Episodic Memory (1983), one of the most cited books in the field.

At the time the proposal was theoretical, and it was contested. What made it stick was the neuropsychological evidence that followed: patients with medial temporal lobe damage can lose episodic memory while retaining much of their semantic knowledge, and neuroimaging showed partly different networks for the two kinds of retrieval. The distinction is now routinely applied in clinical assessment of Alzheimer's disease and frontotemporal dementia.

Encoding specificity

Working with Donald Thomson, Tulving showed in 1973 that memory failure is frequently a retrieval problem rather than a storage one. The encoding specificity principle holds that a cue is effective to the extent that it overlaps with the memory trace as laid down at study; if the cue is absent at test, the information may be available but not accessible. This explained why recall is better in the environment where learning took place, why recognition can sometimes be worse than cued recall, and why weakly encoded targets can be retrieved by strongly associated cues — a regularity Tulving and S. Wiseman described as the Tulving–Wiseman function. He called the act by which a cue brings a stored trace to mind synergistic ecphory.

Memory, consciousness and mental time travel

From the mid-1980s Tulving linked his memory systems to kinds of awareness. Episodic retrieval, he argued, is accompanied by autonoetic consciousness: the rememberer is not merely aware that something happened but mentally re-locates themselves in the remembered episode. He coined mental time travel for this, and later added the crucial point that the same machinery runs forward — that the principal evolutionary payoff of episodic memory is the ability to simulate and plan for a future that has not happened.

The evidence he found most compelling came from patient K.C., a man who, after a motorcycle accident, had largely intact semantic memory but a profound episodic deficit. K.C. could not recollect his past and could not imagine his personal future, and Tulving treated the pairing as the signature of a system whose loss removes the self from time.

To measure the subjective side of this without relying on introspective essays, Tulving introduced the remember/know procedure: on each recognised item, a participant reports whether they remember the study episode itself or merely know the item was presented. The paradigm became a standard tool in cognitive psychology, neuroimaging and patient work, intended to separate recollection from familiarity.

Brain imaging and the HERA model

Tulving was among the first memory researchers to adopt functional neuroimaging. His PET work in the early 1990s produced a striking regularity: episodic encoding tends to engage left prefrontal cortex, episodic retrieval the right. He proposed the hemispheric encoding/retrieval asymmetry (HERA) model for this, together with the notion of an "episodic retrieval mode" — a sustained mental set, distinct from the retrieval attempt itself, that produces the right-lateralised signal. He also advanced a novelty-encoding hypothesis, tying the efficiency of new episodic formation to dopaminergic responses to novelty.

Major Publications

  • Tulving, E. (1962). Subjective organization in free recall of "unrelated" words. Psychological Review, 69, 344–354.
  • Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization of Memory. Academic Press.
  • Tulving, E., & Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80, 352–373.
  • Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104, 268–294.
  • Tulving, E. (1983). Elements of Episodic Memory. Oxford University Press.
  • Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science, 247, 301–306.
  • Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53, 1–25.

Impact and Legacy

Tulving's awards include the American Psychological Association's Distinguished Scientific Contribution Award (1983), election to the National Academy of Sciences (1988) and the Royal Society (1992), the Canada Gairdner International Award (2005), appointment as an Officer of the Order of Canada (2006), induction into the Canadian Medical Hall of Fame (2007), and the American Psychological Foundation's Gold Medal Award for Life Achievement in Psychological Science. The vocabulary he introduced — episodic memory, encoding specificity, autonoesis, mental time travel — is now ordinary language in memory research, and the SPI (serial-parallel-independent) model he developed with Daniel Schacter remains a standard account of how the systems interact.

His claims have also been contested on several fronts, and the arguments continue.

Is the episodic–semantic boundary a system boundary? Critics in the connectionist tradition, James McClelland among them, argue that the two are better described as points on a continuum: episodes lose their contextual detail over time and are gradually transformed into decontextualised knowledge, so that one's vivid first driving lesson becomes, decades later, simply knowing how to drive. On this reading the "systems" are emergent properties of a single learning mechanism shaped by the statistics of experience, and Tulving's own acknowledgement that semantic memory depends on episodic memory for its acquisition — the serial component of SPI — is read as conceding part of the case.

Is episodic memory uniquely human? Tulving held that it was, on the grounds that autonoesis requires a self extended in time. Since Clayton and Dickinson's 1998 work with scrub jays, which showed that the birds remember what food they cached, where, and how long ago, "episodic-like" memory has been demonstrated in rodents, apes, dolphins and cuttlefish. Suddendorf and Busby objected that such behaviour can be produced from semantic knowledge — knowing the date of your birth is not remembering being born — and Tulving, while conceding that "interesting analogues" exist in animals, maintained that they do not demonstrate autonoesis. He proposed a behavioural test, the "spoon test", to settle it, and the question remains open.

Does the remember/know procedure measure what it claims to? The paradigm assumes that remember and know report distinct processes, recollection and familiarity. Signal-detection critics have argued that the two response categories can be generated by a single strength dimension divided at a decision criterion, which would make the procedure sensitive to response bias and to confidence rather than to two memory systems. Tulving defended the dual-process reading, and the dispute drove a large methodological literature on how the judgement should be modelled.

None of these criticisms has dislodged the central achievement. Tulving's own verdict on his method, given in a third-person autobiography he wrote for a lifetime achievement award, was that "ignoring accepted practice has been Tulving's major problem". Reviewers of his work have generally treated that as a description of how the episodic–semantic distinction came to be proposed at all.

Learn More

References

  • Tulving, E., & Donaldson, W. (Eds.). (1972). Organization of memory. Academic Press.
  • Tulving, E. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80(5), 352-373.
  • Tulving, E. (1983). Elements of episodic memory. Oxford University Press.
  • Tulving, E. (1985). How many memory systems are there? American Psychologist, 40(4), 385-398.
  • Habib, R., Nyberg, L., & Tulving, E. (2003). Hemispheric asymmetries of memory: The HERA model revisited. Trends in Cognitive Sciences, 7(6), 241-245.
  • Clayton, N. S., & Dickinson, A. (1998). Episodic-like memory during cache recovery by scrub jays. Nature, 395(6699), 272-274.
  • Haggbloom, S. J., et al. (2002). The 100 most eminent psychologists of the 20th century. Review of General Psychology, 6(2), 139-152.
  • Suddendorf, T., & Busby, J. (2003). Mental time travel in animals? Trends in Cognitive Sciences, 7(9), 391-396.

Last updated: 2026-10-27

Menu
Loading