Memory Mondays, Ch. 3: Short-Term Memory

Welcome to Memory Mondays, where I read a textbook on memory and talk about what I Iearned. If you like your cognitive psychology neatly summarized, with a healthy dose of unnecessary commentary and bad humour, this is the series for you!

Short-term memory is not what many people think it is.

It is not about remembering something from a few days ago, or even a few hours ago. That’s long-term memory. It is very small, and very brief, and while it is part of working memory, it is not the same as working memory. That’ll be another chapter.

It is brief retention of information, and can be divided into two parts: verbal and visuo-spatial.

Verbal Short-Term Memory

The model of verbal short-term memory discussed the most in the textbook is the phonological loop, which is part of the multicomponent model of working memory proposed by Baddeley and Hitch. The former, it is worth noting, is one of the authors of the textbook and specifically of this chapter. While I could be skeptical of how balanced the presentation of different models of short-term memory is, and I certainly finished the chapters on short-term and working memory with more understanding of and support for the multicomponent model than the alternatives, I do think it is the more broadly used and accepted model. It was the only one discussed in any detail in the cognitive psychology textbook I read previously, and aligns with my understanding of dual coding. Rather than a researcher using this chapter to defend and promote a model he is heavily invested in, I instead see an expert using his understanding of experiments and studies to explain our current understanding of short-term, of which his model forms the core.

The phonological loop has two components: short-term store and articulatory rehearsal process. The main method of measuring verbal short-term memory is the digit span, which is the longest sequence of numbers you can remember and repeat back without errors. When you initially hear (or read) the numbers, they sit in your short-term store and rapidly fade. You can maintain this memory by using articulatory rehearsal: basically, by saying the numbers to yourself inside your head, over and over. Like a loop.

The importance of this rehearsal process can be demonstrated by using articulatory suppression, saying something aloud over and over. It is easy to demonstrate this effect for yourself. Try remembering a string of seven numbers. Now try, while saying “rabbit” again and again, aloud. Harder now, isn’t it? Even if you read the numbers, they are stored in verbal, not visual, short-term memory.

The word length effect also demonstrates the role of rehearsals. Our verbal memory span for a list of words declines as the words become longer. This is because we are rehearsing in real time, and longer words take longer to say, leaving more time for other words in the list to fade in the short-term store and be forgotten.

The main shortcoming of the phonological loop model is the lack of clarity on how order is stored in memory: “the problem of how a system like the brain that processes events in parallel can preserve serial order” (pg. 75) is a challenging one. However, the model explains effects such as the word length effect, and links well with other aspects of working memory.

Alternative models have been proposed to address questions not sufficiently answered by the phonological loop model. For example, the Object-Oriented Episodic Record (O-OER) model tries to account for the irrelevant sound effect, which is the disruption of verbal short-term memory by speech or music. These models also have their own flaws. For the O-OER, it places verbal and visuo-spatial short-term memory in the same system, which is not supported by experiments or neuropsychological evidence. Other models, such as the Scale Invariant Memory, Perception, and Learning (SIMPLE) model (not really a simple acronym) are fairly recent and still being worked on. Perhaps the advantage of the multicomponent model is its age; it has been developed for longer.

Visuo-Spatial Short-Term Memory

The other main component of short-term memory stores visual and spatial information. Separately. They are separate. Visual is the “what:” golden brown, on a white plate, flecked with bits of garlic and shiny with butter. Spatial is the “where:” off-centre on the placemat, in front of the salt and pepper (I may be writing this during dinner). First, they are measured through different tasks (like how digit span measured verbal short-term memory). Visual short-term memory is measured using pattern span, where a grid of squares are coloured in a pattern which the subject needs to remember; the number of squares increases until accuracy breaks down. Spatial memory is measured using the Corsi span task, where the experimenter taps a pattern on an array of nine blocks and the subject imitates the pattern; the length of the sequence increases until accuracy breaks down. Secondly, they are disrupted through different interfering activities (like how verbal short-term memory is disrupted through articulatory suppression); Corsi span is reduced when a spatial tasks is used, and pattern span is reduced if a visual task is used.

Even though they are separate, they are connected. When you scan the scene in front of you, you are taking in and using both visual and spatial information. You need information about the “what” and the “where” to plan and carry out action.

There are a couple differences between visual short-term memory and visual long-term memory. First, there is a difference in encoding time, which is the time needed to create the memory; visual long-term memory needs longer exposure to the stimulus. The second difference, much more interesting, is capacity. Visual long-term memory has a much larger capacity than visual short-term memory. Think about pattern span, which is short-term memory and quite small, compared to your ability to visualize your house (or apartment or dorm room) and its contents, which is drawing on long-term memory. Unfortunately, the chapter did not compare verbal short-term memory and verbal long-term memory, but I suspect there are similar differences in capacity.

Primacy and Recency Effects

The last areas of short-term memory to cover are the primacy and recency effects. If you are given a list of items to remember, in any order, your memory will be best for items at the beginning of the list (primacy effect) and for items at the end of the list (recency effect). These effects likely occur because the earlier items are held in long-term memory and benefited from additional rehearsal, while the later items have not yet faded from short-term memory.

The recency effect is not limited to short-term memory. Even for long-term memory, we more easily recall recent items. This is beneficial, as recent information (e.g. about your environment) is more likely to be relevant to planning and decision-making.

One explanation for the recency effect lies in retrieval. An interesting analogy is used for this, and merits being quoted here:

Crowder (1976) likens the task of retrieving items from free-recall lists to that is discriminating telephone posts located at regular intervals along a railway track. The nearest post will be readily distinguishable from the next, while as the posts recede into the distance, the problem of separating one from the other becomes increasingly hard. This process can be seen in terms of a discrimination ratio, based on the temporal distance between the item being retrieved and its principal competitor, the one immediately before it. On immediate recall, the most recent item has a considerable advantage, but with increasing delay, discriminating an item from the one before becomes less and less easy. (pg. 77-78)

Real World Applications

Research on short-term memory is still working on studying more authentic scenarios and complex situations outside the laboratory. However, the research needed to start by analyzing simple tasks, such as digit span, and situations that are easier to control and offer opportunities for replication by other researchers. These simple tasks also allow particular aspects of short-term memory to be isolated, which is much harder to do in more authentic scenarios, where stimuli are likely to include verbal, visual and spatial elements. Once there is robust laboratory evidence, the theories can be tested in real world situations, where more variables are present and thus conclusions are more open to different interpretations.

The true relevancy of the models of short-term memory is revealed in the study of working memory, which will be discussed next week.

Next week: models of working memory


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s