Working memory. Working memory utilises short term memory but refers to our ability to process and maintain one set of information whilst completing a task (Baddeley, 1992). For example, remembering the order of number to dial when a phone number is read out to us. One of the most frequently published games designed to train working memory is a variation on a pair matching task, where players are required to remember the location of an increasing number of paired items on a grid (see Figure 17 for an example from the game Fruit Smiley Brain Games (Hatch Media LLC., 2013). This game is based on the Paired Associates Learning task which is often used in neuropsychological assessments (Xxxxxxxx, 2008). The user is required to hold the position of multiple cards in their memory whilst continuing to uncover new cards. The pair matching task is increased in difficulty by increasing the number of images to match or including a timer. Pair matching games can be played single player or competitively (e.g. Best Memory Matching Game; Dainty Game., 2014). ` Figure 17. Images from ‘Fruit Smiley Brain Games’; a pair matching game designed to train memory. An alternative working memory task is a computerised version of the ‘N-back’ test. In the classic N-back test, a sequence of stimuli is presented and user is asked to identify when the current stimulus matches the one from n steps earlier in the sequence. In the N-back test from the Lumosity training series (Lumos Labs Inc., 2014), a series of symbols are presented and the user needs to press ‘Yes’ or ‘No’ to identify if the symbol matches the previously presented symbol (see Figure 18). In the Lumosity example, the task is timed and requires the user to remember 1 symbol previously (1 N-back) but the intensity of the task can be raised by increasing the number of numbers back the user needs to remember (2 or 3 N-back). Figure 18. N-back working memory task from Lumosity brain training
Working memory a multicomponent model for the exec- utive functions The multicomponent model of working memory proposed by Xxxxxxxx (1986) is one of the most actual and influential models of executive functioning. This model was designed with the aim to develop the initial modal conception of short- term memory (Xxxxxxxx & Xxxxxxxx, 1968). The main prediction of the initial model suggested that short-term memory (STM) was the main entry towards long-term memory (LTM) encoding, with the capacity to encode information in LTM depending upon the time of residence of an item in STM. This assumption has been challenged in several ways. For example, in a study by Xxxxx and Xxxxxxx (1973), a long sequence of words was presented with the instruction to recall the last word beginning with a pre-specified letter, e.g. with the letter B and the following list of items: ball, cabbage, blue, apple, cloud, basin, etc., the response would be basin. At the end of the task, subjects were asked both to answer the initial instruction, and to recall all words beginning by the specified letter. While they had no particular problem to recall the last word beginning with a pre-specified letter, they showed a poor level of performance to recall the other words of the sequence, which were thought to be held in STM during a certain amount of time. Moreover, no relationships were found between the time of maintain of the words in STM and the final performance. This result suggested that learning in LTM is not directly dependent upon the time of residence of an item in STM but could rather depend upon the manner items were processed to achieve a given task. Hence, relationships between STM and LTM seem more complicated than those predicted by the initial predictions geared from modal views of STM. A more explicative view was taken by the “levels of processing” approach (Xxxxx & Xxxxxxxx, 1972). Accordingly, LTM learning rather depends upon the nature of the processes of encoding than on the time of residence of an item in STM. The memory trace durability is assumed to be larger when the en- coding processes are deeper. A continuum of levels of processing starting with a superficial and peripheral level of processing to an elaborated semantic level was proposed. For example, processing a word at a visual level only (peripheral level), was not expected to lead to an efficient LTM encoding, while processing the meaning of a word was expected to lead to a durable trace in LTM (semantic level). This effe...
