Memories are made of this

Memory is very important. It is the basis of all human thought and learning. Without it, we would find ourselves in a world that is perpetually new and unfamiliar. Teachers everywhere make it their business to try to understand how memory works so that they can optimise learning for their students. Without at least a rudimentary understanding of how human memory works, teachers can make errors and students can become disadvantaged.

Over the years, many psychologists have investigated the nature of memory, and have formulated a number of theories about its functionality, scope and challenges. In my next few blog posts I intend to highlight some of the more important theories of human memory and offer some critical commentary on how they have influenced the design of learning experiences. We start off with some useful ideas around proposed cognitive architectures - how we image human memory is organised.

Dual Coding Model

Canadian cognitive psychologist Allan Paivio will probably be remember most for his work on Dual Coding Theory. Paivio's main argument is that we represent our learning through both verbal codes (logogens) and visual codes (imagens).  In other words, if you want to remember something, the best way to do so is to make sure you code it as both a verbal and a visual memory. Abstract words that have no pictorial analogue, says Paivio, are often harder to recall than words that you can associate with images. Consciously code (or make meaningful) words and images together, and this way, according to the theory, retrieval from memory will be easier.

This theory is similar in its architecture to that proposed by British psychologist Alan Baddeley. With his colleagues Baddeley proposed a version of Short Term Memory (STM) called working memory (WM). Baddeley argued that WM (our conscious dynamic memory) was composed of a Central Executive control component and two separate slave sub-components called the visual spatial sketchpad and the phonological loop. According to Baddeley, the visual spatial sketchpad can only hold one image at a time, whilst the phonological loop can hold up to seven (plus or minus one) auditory items simultaneously. The latter is time based, so has limitations if for example you wish to remember a large sequence of numbers. Working Memory is dynamic, ever changing, and relies on coding if content is to enter into Long Term Memory where it is stored for later retrieval. It has been proposed that unlike WM, LTM can store an infinite amount of memories - a theory that we will probably never be able to test.

Working Memory Model

Both Paivio and Baddeley saw the importance of identifying different kinds of memory and how they can be represented within cognition. Both also stressed that these different modalities of dynamic memory should be allowed to work together to strengthen long term, retrievable memory.

Learning in the WM needs to be coded consciously (made meaningful) for it to become a stronger more permanent memory, and learning can be strengthened further if both types of coding (visual and auditory) work in concert. This idea is exemplified in the classic 'show and tell' form of learning. If students strongly associate words and images together, then later a recurrence of either should activate the other.

Clearly, my commentary here is just a simplified version of the two models. Other components have since been added and models extended as research has progressed. There are also problems with these models of memory. For example, other sensory modalities have not been considered. What about tactile memory, or taste? Or perhaps our strongest and most evocative memory of all - olfactory memory? These modalities tend to be ignored in most memory models, but patently, they represent a large proportion of what we can recall about our personal experiences.

How can Paivio's and Baddeley's theories be applied in education? This link outlines Paivio's thinking around the application of his dual coding theory to teaching and learning. We already know that in the design of digital learning materials, both audio and visual content can be combined to reinforce learning. Video and film are prime examples of learning experiences that have impact because they combine audio and visual content. The addition of text can be helpful, but we need to be aware that overwhelming students with too much content presented in different modalities can also be counterproductive. More research is needed to discover what are the most effective combinations of text, audio and visual materials, and whether these vary according to individual learning needs and expectations, time of day, orientation of task, size of screen, colours, juxtaposition of items, and so on. We also know that WM has a limited capacity, and can be overloaded if not enough space is available to code effectively. In my next post I will examine cognitive load theory and its importance in the design of digital learning content.

Photo by Todd Martin
Graphics by Steve Wheeler


Memories are made of this by Steve Wheeler is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.