Atkinson and Shiffrin’s (1968) multi-store model suggests that memory is a flow of information processed through three stores (Myers, 2010). The first system called sensory memory (SM) processes mainly visual and auditory information from our environment (Myers, 2010). It is made up of several components associated with each sense and filters information experienced by the senses – iconic SM which receives visual information has a very short retention time of merely a half of a second while echoic SM which receives auditory stimuli has a slightly longer retention time of three to four seconds (Sperling, 1960; Myers, 2010). If attention is brought to sensory information, it is transferred to the next system called short term memory (STM) (Atkinson & Shiffrin, 1968). This system processes visual and auditory information received from SM for approximately one minute and also temporarily uses retrieved information from the long-term memory for problem-solving (Myers, 2010). Miller’s (1956) study has shown that STM is not only limited in retention duration but also in capacity having on average a storage capacity of seven bits of information whether verbal or numerical (Myers, 2010). The last system is long term memory (LTM), often referred to as the ‘permanent storage’ (Atkinson & Shiffrin, 1968). As its name suggests, it has unlimited capacity to store a vast amount of information such as motor skills, language, autobiographical and factual information (Baddeley, Eysenck, & Anderson, 2009).
The three basic memory stores are distinct from each other in terms of encoded information, capacity and duration of information retention (McLeod, 2007). One of the strongest evidence to support this distinction lies in Murdock’s (1962) experiment – when presented with a list of words, the tendency was that the participants would more likely recall the first words (primacy effect) and the last words (recency effect) than the words in the middle of the list (Myers, 2010). This is known as the serial position effect whereby the first words are recalled since they have been transferred to LTM and the last words were still accessible in STM (Myers, 2010). However, the middle words were present for too long to be in the STM but not long enough to be encoded in LTM, giving evidence that STM and LTM are two different and separate stores (McLeod, 2008).
Research on patients suffering from amnesia also support the multi-store model (Groome, 2006). One classic case is reported by Corkin (1968) about a patient HM who suffered from anterograde amnesia – He was unable to form new memories and lost part of his existing memories (Groome, 2006). However, despite his inability to create new memories, he was still able to have a conversation thus indicating that his STM processes were intact with normal capacity and duration (Wickelgren, 1968 as cited in Groome, 2006). He lost the capacity of only one store hence supporting the idea that STM and LTM are separated (Groome, 2006). Another case is a patient known as KF who suffered damaged to his STM – he was still able to temporarily recall visual information but could not process auditory information which made conversation difficult (Myers, 2010). On the other hand, his LTM processes were retained, again providing evidence that STM and LTM are separated systems (Myers, 2010).
Although KF’s case study explicitly provides strong evidence for distinction of STM and LTM, it also questions the idea of having only one STM (Groome, 2006). The damage in KF’s STM seemed to have only affected the echoic and...