Memories are synaptically linked to other memories, and these links can be made in several ways.
Transient links are made all the time. For instance, when preparing to take a holiday we typically think about where we are going, what the weather might be like, what clothes we should take, what money or documents we might need, etc. Our thoughts, in this example, might seem to be occurring more or less at random, jumping from weather to clothes to money, but they are not at all disconnected. If we really wanted, we could find the exact memory item that triggered the mental jump from weather to clothes, from clothes to money, and so on. There is always a path followed by our thoughts from one memory to the next; it creates what is often called a train of thought.
Sometimes, we make memory links in play. We build connections between memories in our thoughts and so make associations between events that may never have actually taken place. We do this, for example, when we daydream, or when we plan how we would spend the millions we might win in a lottery. Once chains of thought have been built in this way, they are often revisited. Replaying any chain of thought strengthens it, which makes future recalls easier.
A more definitive linkage, discussed in Solving Problems, occurs when we problem solve. Permanent links between previously unlinked memories are likely to form whenever significant relationships are found.
Memory chains can be short or lengthy. Many older folk, for example, can recall exactly where they were and what they were doing when they first learned of President Kennedy’s assassination. This is likely to be a short sequence, robustly formed in association with strong emotions, but not often part of a longer sequence. On the other hand, a journey once taken, or detailed plans to build an elaborate addition to the house, for instance, might be stored as very lengthy chains of linked memories.
We all possess many useful, short and long, memory chains, assembled during daily living or taught to us by others. They form actual mental structures, just like the ones we have been calling the neural networks that store discrete memories, but longer and more complex. Whenever activated, they give rise to thoughts and actions that usually play out in succession. As more-or-less permanent memory structures, they minimize the amount of thinking to be done (and therefore the amount of energy expended) when there is little novel to be considered in the triggering situation. Memory chains are important enough to deserve their own name: I term them constructs.
- Recall that in The Mind we defined memories to include such elements as facts, theories, opinions, personal experiences, emotions, past thoughts, ideas, etc.
- Something like the following was once said by an assembly-line worker: “Everyday I comes in, and I switches on the machine. Then I marries the Duke. . . .”
- Easier, but often less accurate. Over time, the accuracy and truth of any event held in memory can become unwittingly modified. Memories become erroneously linked (as witnesses’ differing statements make obvious). What we think happened may not in fact have happened at all. A wished-for fantasy (for example, that the girl or boy next door had a crush on you) can later be remembered as reality, and is a disorder that has been called the False Memory Syndrome. (See Elizabeth Loftus & Katherine Ketcham, The Myth of Repressed Memory: False Memories and Allegations of Sexual Abuse [New York: St. Martin’s Press, Inc., 1996] for a discussion of this syndrome.)
Severing the link between the left and right hemispheres of the brain can also cause false memories. The left hemisphere (which searches for and provides explanations of observations recorded primarily in the right hemisphere) when disconnected can be shown to invent reasons for witnessed events, because the mind needs to resolve the conundrum such unexplained events pose. See Michael S. Gazzaniga, “The Split Brain Revisited,” in the July 1998 issue of Scientific American, 50–55.
- For example, any long-partnered individual can often predict how their companion will respond or behave, because many responses stem from a “hardwired” neural network chain.