Working memory is the “bridge” between the information collected using our senses and stored in sensory memory, and long-term memory where information can be later retrieved. Working memory is commonly associated with consciousness, and is where we can work on multiple activities simultaneously. While not part of long term memory, by applying a computer model to working memory we would arrive at a parallel processing model because of this “multi-tasking” nature.
Working memory is the portion of memory which sensory data enters from the various sensory memory areas. For example, when information is seen visually, it is stored in the visual sensory memory area, and then brought into working memory for processing before it is determined if the information should be encoded into long-term memory.
Working memory is that part of the brain’s memory system where information is stored for a relatively short period of time, between 15 and 30 seconds, and where the brain can process pieces of information in more than one way or accomplish several tasks at once. This is why we can be listening to music, reading and still hear our spouse call our name.
The current research theories suggest that working memory has approximately seven information “slots” that can each accommodate some discrete piece of information, with each piece being of the same or different types of data. However, with this low number of available positions or slots, working memory cannot handle much information. Consequently, the theory of chunking was introduced based upon ongoing research.
Chunking is the process of grouping information into chunks or pieces that make sense to the learner. This allows us to handle much larger pieces of information than we might otherwise be able to handle. We can see this type of approach in computer technology. The basic element of data is called a bit, which can be either a 0 or a 1. We chunk bits together to form a byte (8 bits) and we chunk bits together to form words. (A word in the computer area is a construct based upon the processing width of the CPU. Words nowadays are often 64 bits or 8 bytes long.)
The question then is why do we care about working memory? By organizing learning in such a way as to be easily chunked, or for separate pieces to be easily identified as related, we can help working memory more quickly process the information and determine its relationship to information already stored in long-term memory, assisting the long-term memory encoding process.
Additionally, presenting the information in smaller sections helps the learner rehearse the information in working memory, which assists with the encoding of that information for long-term storage and makes it for easier recall of that information. Depending upon which model of long-term memory you want to subscribe to, this could increase the number of nodes, connections, or concepts that are created, which also makes it easier to recall the information later.
How we process information in working memory has an impact from a behaviorism perspective. Recall that with behaviorism, we are attempting to make some change in the learner’s behavior through a variety of methods like positive and negative reinforcement, shaping, fading, chaining, etc.
By presenting information to the learner in more easily processed chunks, we can improve their ability to process the information, establish relationships with previously learned information recalled from long term memory, and promote understanding of not only the new information but that previously learned information. We know that creating new relationships between information not only promotes further understanding, but strengthens relationships between discrete concepts and enhances recall.
Because we have utilized these pathways, we can expect to see a change in the learner’s behavior, all things being equal. If the learner decides that our assumptions are invalid from their perspective, they may choose to not assign the same weight or importance to the information that we did, thereby breaking down the learning and memory processes.
It is important to understand both behaviorism and cognitive information processing for the reasons already presented, but also because it helps us as instructional designers consider what behavior we want to change, which could be a demonstration of newly acquired knowledge, and therefore how the information should best be presented to the learner to assist them in integrating the new knowledge into long-term memory and therefore be capable of exhibiting the desired behavior.