Working memory - the workplace of cognition

In recent years, the notion of short-term memory has been developed into the concept of working memory. Working memory is a system providing temporary storage and manipulation of the information that is supposed to be essential for creating important links between perception and controlled action. Reisberg (1997: 143) states: “Working memory, it seems, deserves its name: it is indeed the workplace of cognition. But if working memory is to serve this function, then it must be capable of storing and working with all the diverse content that we can contemplate and attend – pictures and words and smells and abstract ideas, to name just a few.”

Johnstone (1984) distinguishes between short-term memory and working memory and introduces an explanation for the distinction between them. For remembering a set of numbers, like phone number, this process occurs in our short-term memory and no processing is taking place to recall it in the same order within a matter of seconds. However, to memorize them backwards, a working process must take place and the short term memory is now called working memory. Essentially, short-term memory and working memory are the same space but the use defines the name.

Johnstone (1984) introduced a definition of working memory from the educational viewpoint as “that part of the brain where we hold information, work upon it, organize it, and shape it, before storing it in long-term memory for further use.” The working memory space is very limited in terms of both its capacity (amount of information it can hold) and its duration (length of time it can hold information). Furthermore, working memory space depends on the age of the individual, and as Miller (1956) showed from memory experiments, the average capacity is about seven plus or minus two ( 7 ± 2 ) separate chunks. Chunking is the process of grouping into units which could be a single number, a letter, or many pieces of information, and the nature of the items plays a major role in the capability to recall.

Baddeley and Hitch (1974) suggested one of the most influential models of working memory. Based on their findings, they conceived that working memory as a multi-component system; the central executive is aided by two peripheral and independent systems – the phonological loop and the visuo-spatial working memory – that temporary store verbal and visuo-spatial information. This theoretical framework illustrates a development of earlier models of short-term storage (such as the model of Atkinson & Shiffrin, 1968).

It is arguable that the central executive system is the most important of the three components of the working memory model because it controls the allocation of resources between the phonological loop and visuo-spatial working memory. The central executive is considered to be multi-functional and complex, and there is a considerable debate about the precise nature of its function (Bull & Espy, 2006). Miyake et.al, (2001) proposed that visuo-spatial working memory has a closer relationship with the central executive than the phonological loop, and visuo-spatial working memory can store more information as a whole. Although the central executive is the most important, it is the least understood component of the working memory (Baddeley, 2006).

The phonological loop is a slave system that specialises in processing the language-based information. Baddeley (1999) proposed that the phonological loop can be divided into two subcomponents: Passive phonological store is able to hold verbal information, for example before the memory trace vanishes or is refreshed by the articulatory control process (the second component), which is an active phonological rehearsal mechanism. Bull & Espy (2006) clarified the characteristic of the phonological loop as follows "Information held in the phonological store is subject to decay, unless it can be refreshed by sub-vocal rehearsal, a process akin to repeating under one's breath the information one is trying to retain. Sub-vocal rehearsal, then, can be disrupted by secondary tasks that also use the verbal resources of the PL". Visually presented material (e.g. written material) can be also converted into an articulatory code by the articulatory control process and then it will be transferred to the phonological store (Baddeley, 1997).

Many researchers have used this phonological loop characteristic to study its processes, using the secondary or dual-task methodology (Lee & Kang, 2002; Furst & Hitch, 2000; Logie et.al, 1994). There are basic phenomena associated with the phonological loop as follows (Logie, 1995; Baddeley, 2006): The phonological similarity effect is the clearest phenomena associated with the phonological loop, and it refers to the fact that recall of a series of words or letters is more confusing when these words or the letters sound alike. For example, sequential recalling for a series such as “Mat, Cat, Fat, Rat, Hat, Chat” is more difficult than a sequence such as “Bus, Clock, Spoon, Fish, Grate, Man”. This phenomenon results from the confusion of the similarity of the verbal items that comprised in the phonological to one another (Baddeley, 1966a; Conrad, 1964).

Irrelevant Speech refers to the disruption that occurs from presenting irrelevant speech, while the recalling process is held. Salame & Baddeley (1982) discussed the disruptive effect of irrelevant speech. This kind of speech involves the conveying of ideas not relevant to the task in hand. The irrelevant speech accesses directly the phonological store, thereby disturbing its contents. Word length effect is a phenomenon where recalling sequences of long words such as “University, Aluminium, Hippopotamus, Mississippi, Refrigerator” is less well done than short words sequences such as “Pen, Book, Chair, Greece, Nail” (Baddeley et.al,1975). The interpretation of the word length effect is that long words take a longer time to say or rehearse, allowing a greater degree of trace decay, thus will be less well retained (Baddeley et.al, 1975; Logie, 1995; Baddeley, 2006).

Articulatory Suppression: is a dramatical collapse in retaining a verbal sequence when subjects are concurrently required to repeat aloud an irrelevant speech sound such as “the, the, the” or “hiya, hiya, hiya” (Levy, 1971, 1975; Murray, 1965, 1968). The Visuo-spatial working memory is a second slave system of working memory which parallels the phonological loop for processing visual and spatial information. Evidence from a number of sources now proposes that the visuo-spatial working memory may involve two subcomponents: one for maintaining visual information and the other for spatial information (Pickering et.al, 1998. Pickering et.al, 2001). Hue and Ericsson (1988) found visual similarity effects in instant recall of unfamiliar Chinese characters. Frick (1988) argued that images in visual-spatial memory are unparsed and uncategorised. He reported when visual confusion errors occur in retention of letters there appears to be independent degradation of parts of the letter, so the letter ‘P’ might be recalled consequence of the similarity effect as an ‘R’. However, the attention that has been paid to study this slave system is much less than the phonological loop. 

Baddeley (2006) justified this because of the absence of a rich and standardized set of stimuli such as those provided by the languages, visual information is much more difficult to quantify. He continued , "It seems likely that the visuo-spatial system will play a crucial role in the acquisition of our visual and spatial knowledge of the world: What color is a banana? How does a bicycle work? How do you play a DVD? How can I find my way around my hometown? Whereas we have many tests of language at the levels of phonology, individual word meaning, and text comprehension, we appear to lack well-developed measures of visuo-spatial world knowledge." However, Baddeley draws attention to the compensating properties for studying the visuo-spatial working memory and the important of vision and visual attention.