Understanding the Atkinson-Shiffrin Multi-storage Model

  • by

What is memory?

Essentially memory is a psychological mechanism that allows us to encode information, store information and retrieve information. Memory is essential for learning, assisting in the formation of your identity and capable of influencing behavior. index

Atkinson-Shiffrin Multi-storage Model

Remember that this is a model of how memory forms – models can be changed, altered or improved. The Atkinson-Shiffrin Multi-storage Model has three major components to it:

• Sensory memory
• Short Term memory
• Long Term memory

So how do these components help to formulate memories in an individual? To understand how these components work it is important to know about them in more depth.

Sensory Memory

This is the term given to the senses that store information from the environment. When you are reading a book, the text on the pages is getting processed by your visual senses. If you are listening to a lecture about quantum physics, your auditory senses are being used. There are two major types of sensory memory that you will be focusing on in VCE psychology, this includes:

• Iconic Memory
• Echoic Memory

Iconic Memory

This type of memory is involved with the vision field only. Anything that you see visually can be stored as iconic memory, this information will not necessarily be further processed as that depends on whether or not attention is given on this stimulus. The storage capacity of iconic memory is very large and covers the whole visual field and the storage duration is very small – information stays in the iconic memory for no longer than 0.25 seconds.

Example: when you are sitting in a café, the visual field around you will be stored in the iconic memory. Although you may see a lot of detail around you, it will get stored in your iconic memory but not necessarily stored in short term memory – unless attention was used.

Practical Example of iconic memory

Look at the following picture for 5-10 seconds. Don’t cheat!


Now answer the following questions.
1) What was the character in the 2 column on the third row?
2) What was the character in the right, bottom grid doing?
3) How many male characters were there?

As you can see – it is quite difficult to remember the image in front of you for a long period!

Echoic Memory

This type of memory is registered by the auditory system only. All the sounds including the pitch, noise, tone and rhythm are stored in the echoic memory. Just like the iconic memory, echoic memory has an unlimited capacity – you can hear everything simultaneously! However again, to prevent this information overloading your brain, echoic memory is very short lasting – only between 1.5 to 5 seconds.

Short Term Memory (STM)

So how does memory form? Remember that the sensory memory is very large and most of it will not enter our memory. For example, when you go to the shops you obviously don’t remember all of the things that you saw or heard – a large chunk of it is lost information!

The Atkinson-Shiffrin model says that information is only transferred into the short term memory when we give attention to the sensory memory. For example, we are more likely to remember a screaming child in a restaurant or shop because we give more attention to it compared to other noises.


The short term memory has a longer storage duration of 18-30 seconds. However this memory can be lengthen depending on whether or not we rehearse the information. This is one of the ways in which information in the short term memory can be transferred into the long term memory.
For example:
when someone new is being introduced to us – we can forget their name after approximately 18 seconds, however if we repeat the name to ourselves a few times then the name can be remembered for a longer time. Or maybe we need to remember a phone number – by trying to rehearse it a few times visually, this image can stay in our short term memory for a longer time.


The short term memory, unlike the sensory memory does not have an unlimited capacity. The capacity is only limited from 5 to 9 chunks (7 ± 2 chunks). Chunking breaking individual bits of information into sections that you can remember.
For example:
Consider the number 0403198719662009 – you can remember all 16 numbers individually by looking at each number as a separate part, however it will be difficult to do this as the short term memory is only limited to usually a maximum of 9 chunks. Instead you can break the information into more meaningful date such as:

04/03/1987 my bday
1966 mum’s birthday
2009 brother’s bday

Chunking is a great mechanism to break down information into bits that are easier to process!

Information in the short term memory may become encoded to the long term memory only if enough rehearsal was done to information in the STM. Information that is not rehearsed will decay and not reach long term memory – for example you may remember the screaming child in the store for a while, but unless you repeat the story many times the information will eventually decay.

Long Term Memory (LTM)

This type of memory is permanent unless disturbed via amnesia, brain injury or other complications such as Alzheimer’s disease.

So what makes this information more permanent?

It is basically the way in which information has been encoded. The information in short term memory can be transferred into long term memory by repetition of information. Long term memory is more permanent as it is used in areas of the cerebral cortex, rather than just the hippocampus which is involved with processing short term memories and newly formed memories.

Types of Rehearsal

There are two major mechanisms that can be used to transfer information into the long term memory these are known as:
• Maintenance rehearsal
• Elaborative rehearsal

Essentially these two mechanisms differ by how an individual repeats the information. For example, say that you receive a phone number from a friend. If you repeat the number, 0401305399, mechanically, several times, you are using maintenance repetition. Maintenance rehearsal takes a lot of work and it doesn’t say in the long term memory for very long – unless you practice rehearsing the number many times.

On the other hand, elaborative rehearsal is when you use information that is already known (in the long term memory) and use it to give meaning to the information in the short term memory. For example, the number 0401305399 can be remembered more efficiently if you break it down into

04/01 your anniversary
30 your address
5399 your PIN number

As you can see it is much easier to use elaborative rehearsal compared to maintenance as the information is more related to things you already know – all that it required is a bit of imagination!

Proof for the Multi-storage Model

The multi-storage model is best supported by the serial position effect. This is a test where a large series of objects is shown to an individual and they are asked to remember this list. It was found from the experiment that the first 6 items shown were remembered well. This is because at the beginning the individual is using maintenance rehearsal to remember the items. This section is known as the primary effect. However as the amount of objects, or data increases, it becomes more difficult to keep track of all the objects and therefore the % of recall decreases for the middle objects.

serial position

It is also found that the objects at the end of the list are remembered more effectively, this is because the information is still stored in the short term memory. Therefore if the individual is tested at the immediately of the test the individual will still remember the last items. This is known as the recency effect.

It was also found that after distraction or a time lag after the serial of objects, individuals don’t remember the last items well – this supports the idea that short term memory only lasts for approximately 30 seconds.

Limitations of Multi-Stage model

Remember that models are not fixed! They are simply used to explain certain phenomena – they can be changed, improved and discarded. Some of the limitations of the multi-stage model include:

It is too simple: case studies have shown that some people can completely stop storing information about their personal lives but still remember facts. The multi-stage model predicts that if the short term memory capacity is damaged – no information should be remembered!

It only focuses on the process: it does not consider that some things are easier to remember than others – for example why is it that we can remember one definition easier than another? The multi-storage model doesn’t focus on this!

It only focuses on conscious rehearsal: there are certain things that we may remember without consciously repeating the information. What about some random event during school and your childhood? Or what about an annoying song that you REALLY did not try to consciously remember?

Leave a Reply

Your email address will not be published. Required fields are marked *