It may come as a surprise to postdocs running mice in water "mazes, poking molluscan neurons in a dish, or flashing computerized test icons in front of bored students—but memory does have a life outside the laboratory. It is this potential detachment from reality, particularly in dealing with human memory, that has prompted Neisser (1978) to deliver in the first International Conference on Practical Aspects of Memory a signal presentation, entitled 'Memory: what are the important questions?'. Neisser claimed, more or less, that the bulk of experimental research on human memory, since its first days some one hundred years earlier (Ebbinghaus 1885), meant very little as much as memory in real life is concerned. He further went on to argue that 'if X is an interesting or socially significant aspect of memory, then psychologists have hardly ever studied X' (ibid.). Neisser's scholarly provocation stirred emotional polemics (e.g. Banaji and Crowder 1989; Conway 1991; Loftus 1991), itself a potentially interesting topic for research on everyday memory (Roediger 1991). It also epitomized, and contributed to, a deflection in the direction of human memory research.
It is difficult to underestimate the importance of the first controlled paradigms for quantifying human memory (Ebbinghaus 1885). Without them, memory research would have not been promoted to the rank of a respectable, quantitative, and reproducible science. In parallel, the mainstream research on animal learning, while looking for reproducible methods and quantifiable variables, has also become increasingly dependent on artificial laboratory "paradigms ("classical and
"instrumental conditioning). Many of these paradigms expose the experimental animal to settings and demands far remote from the ecological niches and from the problems that these niches pose (Boakes 1984). The effectiveness and popularity of these animal learning paradigms have tempted prominent investigators to adapt them to human use (e.g. Woodworth and Scholsberg 1954), and even to go as far as to declare that most, if not all, the riddles of human memory will ultimately be solved by understanding the behaviour of the "rat (Munn 1950). There were, of course, early attempts to investigate human memory in its natural settings (e.g. Galton 1879; Colegrove 1898; Thorndike and Woodworth 1901a,b; "capacity, "transfer). Furthermore, over the years the mainstream 'laboratory approach' to memory research was not without its ardent opponents, who claimed that most of the studies miss the social and ecological context, function, and complexity of memory in real life (Bartlett 1932). However, the challenge did not gain much momentum until the 1970s.
Real-life memory research was at first formulated as an opposition, devoid of a formal consensus on its precise definition (Klatzky 1991). It is known by multiple names, which refer to the same or similar notions: 'everyday memory' ('an awkward phrase', Neisser 1991); 'ecological memory' (Bruce 1985); and 'real-world memory (Cohen 1996). In essence, the majority of the traditional research on memory throughout the first half of the twentieth century has focused on the structure and the 'syntax' of memory, i.e. how the system operates and what are the formal relations between "stimuli and actions ("algorithm, "model). Furthermore, this research was dominated by the role of the investigator, who in many respects shaped the behaviour of the "subject in artificial settings by choosing the problems, the constraints, and the "reinforced response. In contrast, real-life memory research concentrates on the subject in its natural environment, on phenomena relevant to daily life, and on familiar types of stimuli and responses. It emphasizes better the 'semantics' of the learned behaviour, i.e. its content, function, context, and meaning. Here is a selection of 'real life' questions: How accurate are our autobiographical reminiscences? Should we trust eyewitnesses? What shall we do to remember better which name belongs to which face? How do we keep a mental record of what we "plan to do later in the day ("prospective memory)? How do experts differ from novices in their everyday "skill? Or, closer to academic life, how much do we remember after attending a classroom lecture (not too much, and unfortunately or not, jokes rather than facts are remembered best; Kintch and Bates 1977).
Criticism of research on real-life memory focuses on what is considered by the critics as lack of "controllability, rigorousness, and reproducibility of the "methods and the generalizability of the conclusions (Banaji and Crowder 1989). But the reduced control of extraneous variables is the price many investigators are ready to pay in exchange for the ability to tackle naturalistic, complex behaviours, and unveil new memory phenomena (Baddley 1981; Klatzky 1991). Explicitly or implicitly, the modern real-life memory movement did exert a significant impact on human memory research.1 The development of novel techniques that permit on-line analysis of the conscious brain in normal subjects has in parallel extended and enriched the repertoire of methodologies available in the investigation of human memory (e.g. virtual reality, "functional neuroimag-ing). As a consequence, many bona fide real-life phenomena are currently being dealt with at the forefront of human memory research, both inside and outside the laboratory (Cohen 1996; Goel et al. 1997; Maguire et al. 1997).
Selected associations: Context, False memory, Flashbulb memory, Observational learning, Prospective memory
1The parallel in the field of animal learning is the etiological approach (Tinbergen 1969; Lorenz 1981). For a review, see Camhi (1984); for more recent variants, see Eichenbaum (1996) and Chiel and Beer (1997); also *birdsong, *imprinting.
Was this article helpful?