Everyone probably has heard about the concepts of short-term memory (STM) and long-term memory (LTM). The STM and LTM model, also known as the two-processor Multi-Store Model of memory (MSM), was initially announced by Atkinson and Shiffrin in 1968. They argued that memory could be stored in several distinct parts of the brain and that these separate memory processes operate in a single, uniform, sequential way. The prototype distinguishes three main types of cognitive function:
- Sensory memory/buffer
- STM
- LTM
The input enters the modality-specific (responsive to vision, hearing gustation, tactual, and olfactory senses) sensory buffer that consists of two principal types of stores: visual (iconic) memory and auditory (echoic) memory. The sensory stimuli are believed to last up to a few seconds; however, it persists mostly less than a second. After that, only the relevant traces of information are passed from sensory memory to STM, where attention is required for the transfer to occur.
Short-term memory capacity was thought to be 7 +- 2 chunks of information (Miller’s Magic Number, 1956). However, the assumption mentioned previously was challenged by Cowan (2010), who carried out a memory experiment with the so-called ”running span procedure”. In contrast to Miller’s study, Nelson Cowan did not let the participants know the length of the random numbers’ sequence before reading it to them, which meant that contestants were less likely to develop memorization strategies. According to Multi-Store Model, STM is a ‘gateway’ from which the information moves to LTM after approximately 6-12 seconds (some references provide 15-30 seconds). Nevertheless, the intense rehearsal (rehearsal loop) is substantial for the transfer to happen. Some information is displaced by other messages or is lost due to the lack of practice.
Psychologists believe that once the message reaches the last, potentially of unlimited capacity memory store, it may remain there from a few minutes to a lifetime. Critically, since memory is a hypothetical construct, this idea is open to discussion and debate.
The question arises whether the MSM is not outdated. There is a shred of substantial evidence in favor of this model from the XX century – it is worth mentioning two principal pieces of research here: the case study of Henry Molaison, HM patient, by Milner (1966) and the Glanzer and Cunitz’s cognitive experiment based on free recall (1966). However, as more and more data have been obtained, the MSM prototype appears to be over-simplified nowadays. It does not explain memory distortion, why some things got printed on our memory with little (or even no) rehearsal, and the reason behind the learning difficulties despite a lot of practice. In addition, Logie (1999) has challenged the concept of STM being a gateway to LTM.
All in all, the MSM of memory is a simplistic conceptualization of memory systems’ architecture and function. Nonetheless, it was a steppingstone in the improvement of that matter in a cognitive approach.
Sources:
John Crane. “Multi-store memory model.” InThinking. IB Psychology, http://www.student.thinkib.net/psychology/page/22793/multi-store-memory-model. Access: 18 Dec. 2020.
Psychology Unlocked. “Atkinson & Shiffrin’s Multi Store Model of Memory (AKA: Two-Process Model) | Memory.” Youtube, 27 Nov. 2016, http://www.youtube.com/watch?v=7G9IK_mUmRE. Access: 18 Dec. 2020.
IB BioPsychology 