Phases (phasis, Greek for 'appearance') are distinct states in time and space. "Acquisition, "consolidation, retention, "retrieval, and "forgetting are all phases of memory. Some authors use the term 'stage' instead of phase to refer to these different periods of memory. 'Phase', however, is more suitable, because in addition to denoting a distinct state (definition 1), including in development (definition 2), it also connotes the recurrence of states (definition 3). This is important, because memory phases could recur. For example, items in memory shift back and forth between active (retrieved) and nonactive (stored) states (Lewis 1979; "persistence), and, upon retrieval, might even be reacquired and reconsolidated (Sara 2000; Nader et al. 2000). The expression ("performance), endurance, and susceptibility of the "engram to interference are common "criteria in the "taxonomy of memory phases. Our knowledge, however, about the number, nature, and transition of memory phases is a function not only of the particular memory system and memory task, but also of the sophistication of the research "methods employed.
A particularly popular distinction of phases in a memory trace, once formed, is between 'short-term' memory (STM) and 'long-term' memory (LTM). That memory can be short or long lived, was probably noted by the first human who forgot the items on his daily hunting list but still remembered the way back to the cave. More formal distinctions emerged in experimental psychology. The results of both contemplation and experimentation gravitated toward "models of two to three memory phases. Here are a few examples. The German physiologist Exner concluded that 'states of mind' vanish, if not caught by "attention, within a few seconds, and dubbed these first few seconds in the life of a memory 'elementary memory (cited in James 1890). James preferred the term 'primary memory', as opposed to 'secondary memory', or 'memory proper', which is '... the knowledge of a former state of mind after it has already dropped from consciousness' (ibid.). Meumann (1913) advocated a three-phase distinction: immediate, temporary, and permanent. Hebb (1949), drawing from earlier data and concepts (Lorente de No 1938; Hilgard and Marquis 1940), returned to the basic two-phase type of models, and proposed a 'dual trace' hypothesis as well as mechanism: a reverberating, transient, unstable trace, that 'carries the memory until the growth change is made' and memory is stabilized in the long term. The introduction of the specific 'short-term' vs. 'long-term' terminology is accredited to Broadbent (1958; "attention), who spoke about 'short-term' and 'long-term' storage systems (see also Peterson and Peterson 1959).
The current terminology of elementary temporal memory phases refers to 'sensory', STM and LTM. Sensory memory of visual and auditory information is termed 'iconic memory' and 'echoic memory', respectively (Sperling 1960; Efron 1970). Sensory memory lasts for less then a second (Figure 54) to a few seconds at most. The duration of STM and LTM depends on whom you talk to. STM lasts for a few minutes for neurologists (Sacktor and Mayeux 1995), up to a few hours for cellular neurobiologists (Goelet et al. 1986; Dudai 1989, 1997b). LTM, according to this coarse classification, is for neurologists memory that lasts for more than a few minutes, and for neurobiologists, by convention, memory that lasts for more than 24 h. In what is regarded as a cellular analogue of learning, "long-term potentiation in a brain slice, the limits of 'long term' are pushed backwards to 1-3 h only. Similarly, consolidation of LTM could last anything from a few hours ('cellular consolidation', which requires "protein
Fig. 54 The shortest memory phase. In a 'classic experiment, Sperling (1960) presented a group of 'subjects with nonsense 'stimuli composed of letters (or letters and numbers) in various arrangements (a). The stimuli were each presented for 50 ms only. The subjects were requested to report all the items in each stimulus immediately afterwards. When the number of items was higher than four, the subjects never reported all the items correctly (heavy curve in b). The subjects were then presented again with stimuli consisting of letters in three symmetric rows (a, middle and left). They were told that a tone would come immediately after the stimulus, and that this tone would be either high, middle, or low. If high, only the upper row should be reported; if middle, only the middle row; if low, only the bottom row. Under this *cued, partial report mode, the amount of information reported was two to three times larger than in the whole report mode. The availability of this information declined within less then a second (b; the straight diagonal is the theoretical curve; the numbers on the other curves are the onset of the tone in seconds after the visual stimulus). This short-lived memory is called 'iconic'.
synthesis) to weeks or longer ('system consolidation', Dudai 1996; "hippocampus). Behavioural, "neurogenetic, pharmacological, and cellular analysis unveils additional, 'intermediate' memory phases, between 'short' and 'long' (Rosenzweig et al. 1993; Tully et al. 1994; Winder et al. 1998;Wustenberg et al. 1998; Sutton et al. 2001). In "real-life, 'long' itself is evidently not an homogeneous phase, as some LTMs are irretrievable already within a few days, whereas others linger for many years, up to a lifetime (e.g. EBT 1923).
Different memory systems ("taxonomy) may have different types of temporal phases, probably moulded in evolution to comply with the specific functional demands imposed on the particular system. For example, in mammals, "declarative memory is characterized by multiple LTM phases, ranging from weeks (hippocampal-dependent) to years (hippocampal independent), whereas nondeclarative memory appears to be consolidated faster (McClelland et al. 1995; Shadmehr and Brashers-Krug 1997a). Such heterogeneity probably stems from the different properties of the particular circuits that subserve the different memory systems, but not from different building blocks at the molecular "level.
Why are there short- and long-term phases of memory? And are these phases serial or parallel? The favoured answers to the first question are three: (a) transient STM phase(s) provide the organism with the ability to hold information indispensable for ad hoc tasks but superfluous in the long run (in this respect, 'short-term' fulfils the role of '"working memory'); (b) transient STM phase(s) provide the organism with a better opportunity to evaluate, prune, classify, and rearrange information before the decision is taken to store it 'permanently' (McClelland et al. 1995); (c) there are phases because biological memory is incapable of operating otherwise, due to the structural and functional constraints of the machinery.1 As to the second question, namely are the short- and the long-term phases of memory serial or parallel, information from both cellular and system studies suggest that short- and longer-term memory could unfold in parallel; further, there are situations in which LTM is intact but STM is faulty (Shallice and Warrington 1970; McCarthy and Warrington 1990; Emptage and Carew 1993).
Finally, all learning and memory processes could themselves be regarded as a phases in the overall continuous process of the development of the organism. Recent findings in cellular and molecular biology support such view, which has been promoted well before promoters, "immediate early genes and transcription factors were even dreamt of: 'Growth and learning are one continuous process, to the earlier phases of which we give the one name, and to the later... we give the other' (Holt 1931).
Selected associations: Acquisition, Consolidation, Retrieval, System, Taxonomy
1The reader who is already familiar with 'paradigms will recognize in possibilities a and ¿the Panglossian paradigm,and in can alternative.
Was this article helpful?