Forgetting is complicated. For one thing, how much you forget depends on the content or task. Some things are easier to forget than others. We store things in various memory systems and retrieve them with various levels of success.
We are not good at remembering nonsense syllables and isolated words. Ebbinghaus shows that although verbal learning is linear, forgetting verbal learning is nonlinear. Learning is a function of time spent. The longer you study the more you learn. But you forget the majority of facts very quickly within the first few hours. After the first day, the decline in recall is more gradual.
We are better at remembering names and faces. But that is complex. Bahrick showed that memory for names and faces depends on the task. We are very good picture recognition. Given a picture, we can often tell if it is of a former classmate or not. Given a name, we quite good at deciding if it is a former classmate. We seem to have a preference for faces. In contrast, college teachers remember the names of their students more than their faces.
We are fairly good at picture and name matching. Given a name, we can often find the correct name from a list. Given a picture, we can often find the correct picture from an array. We are not very good at recalling names from the past in free recall. We are not much better at recalling a name given a picture (picture cueing).
Bahrick et al (1975) studied nearly 400 people who varied in age from 17 to 74. They found that this picture-name connection can last for years but somewhere between 40-50 years old, memory of schoolmates takes a dip. During this same time frame, we lose some of our foreign language knowledge too.
Learning a foreign language is similar to learning Ebbinghaus’ syllables. The better the initial learning, the longer the knowledge remains. As everyone knows colloquially, we tend to forget a language we don’t use. The forgetting starts rapidly before it tapers off. Some memories remain. They seem to last forever. Bahrick calls these memories permastores (like mental permafrost). They are examples of stable, long-lasting memories.
Nonsense syllables and foreign languages are part of semantic memory, our long-term memory of world knowledge and general information. Learning this type of information requires conscious thought but it soon becomes automatic. It took a while for you to learn your name but now it has become a part of you.
Semantic memory is declarative. It is knowledge you know that you know. Knowledge you can affirm (declare). The other type of declarative memory is episodic memory. These are the memories of events, experiences and autobiographical information. They are the stories of your life.
Semantic and episodic memory systems work together. Semantic memory stores that Austin is the capital of Texas, while episodic memory tells you whether or not you have ever been there.
Episodic memory is interesting because it appears to decrease at a rate of about 5% per year (Linton, 1975). This is similar to forgetting how to ride a bike or the lowered performance of not dancing. You remember your first day of school or what you did on your summer vacation. But over the years these events are less recallable.
In contrast to declarative memory, implicit memory (also called procedural memory) is automatic. You don’t have to consciously think about riding a bicycle, driving a car or executing a series of keyboard commands. Indeed, you probably can’t explain how you do these tasks. You just do them. You have learned “how” to do something, instead of learning a “what.” You use non-declarative memory to brush your teeth, put on your clothes and walk across the street.
Implicit memory is “muscle memory,” although it is not stored in the muscles. It uses the cerebellum and the limbic system of the brain. The focus is on the continuous movement of muscles. We use implicit memory to hop, twirl and play video games. These complex skills are very resistance to forgetting. Two years after flying a virtual airplane, subjects had very little forgetting (Fleishman & Parker, 1962). Any loss is quickly recovered with a few warm up trials. You really don’t forget how to ride a bike.
Patients with amnesia keep their implicit memories. They can still write, conduct music and dance. Their old skills are not forgotten. Also, they can learn or improve on implicit memory tasks. They can get better on these tasks. The non-declarative systems remain even after the declarative memory systems no longer work.
Although continuous movement are easily retained, the same is not true for discrete motor movements. Discrete skills are used in typing and playing the piano. Each strike is a separate event. Each key stoke is independent of the next. There is no remarkable retention for discrete skills. (Baddeley, 1974).
What about mixed tasks involving both continuous and discrete components, such as giving CPR. Evidence is that the “survival” of patients drops from 100% to about 15% in a year (McKenna & Glendon, 1985). If you’re going to have a heart attack, make sure you’re surrounded by people who have renewed their CPR certification in the last year.
Clearly, the amount of forgetting is related to the type of material. In general, we don’t forget continuous movements. We do less well with discrete or mixed tasks. We do fairly well with names, faces and episodic memories. We do very poorly on remembering isolated facts.