Verified by 
We know that memory makes mistakes. We also know that some personal memories can be vivid and accurate over long periods of time. How do we account for this difference in the reliability of memory?
Two Levels of Memory
Inaccurate and accurate memory coexist because of a fundamental distinction in personal memory: primary memory and integrated memory. Primary memory is the representation of the original experience in the form of visual images, sounds, smells, tastes, emotions, and bodily sensations. Integrated memory is then constructed over time from the images in primary memory as well as related memories and general knowledge.
If we are encouraged to access and describe our primary memories, then our recall remains faithful to representations of our original experience. In casual remembering, however, we usually draw on integrated memories.
When Memory Makes Mistakes
Integrated memories blend different memories and incorporate beliefs about the world and our own self-concepts. For example, if we think of ourselves as consistent over time – as many people do – we remember our past self as more similar to our present self than it actually was. We then shape the earlier memories to be more consistent with our present self. We have a consistency bias. This is especially true with political views.
Alternatively, if we believe we have changed significantly, then our memory can exaggerate the differences between our past and present selves. People who have seriously transgressed in the past often remember themselves as very different people during those past transgressions. We can also have an inconsistency bias.
Importantly, integrated memories are especially responsive to the retrieval context. If we recall personal events with a theme in mind, that theme influences the information we select from these memories.
If we are remembering to make a point or to contribute to a conversation or to answer a particular question, then our recall may be biased. The outlines of remembering will be accurate, but the details we select will bolster particular traits or themes in our lives – happiness, courage, injustice, accomplishment.
When telling others about a family trip, for example, if we want to convey a narrative of fun, we selectively remember the happy moments and set aside the inconveniences, arguments, and discomforts. While remembering a time when we overcame difficulties to achieve success, we will highlight the difficulties and emphasize our efforts in overcoming them.
How to Remember Accurately
To maximize accuracy, we need to access our primary memory images, independent of specific themes and narrative goals. We should then treat our memories as subjects of careful study.
We begin by retrieving a primary memory. One way to do this is by concentrating on the perceptual qualities of the remembered event – in particular, visual images or particular sounds. What sights and sounds come to mind? Examine one of these sights or sounds and recall as much of the surrounding memory as you can.
We can also enter primary memory through emotion, allowing ourselves to re-experience our original feelings. More holistically, we can immerse ourselves in memory images of the setting. Any of these retrieval strategies will encourage further exploration of the primary memory.
When recalling the birth of our daughter more than 40 years ago, for example, I can call upon integrated memory, relating the events of that night in mostly narrative form, without accessing the vivid images and strong emotions of that time. If I concentrate, however, I can access the vivid images and emotions represented in primary memory, which feel as immediate as if the birth happened last week.
For more comprehensive recall, we can follow our first analyzed memory as it leads to other memories and describe those linked memories with the same thoroughness as the initial memory. In this way, we are practicing reflective introspection – a method that stretches back to the 19th century and continues today as the basis for contemporary research on autobiographical memory.1
Free Recall and Cued Recall
It also makes a difference if we are recalling freely or if we are asked questions.
If survivors of trauma are allowed to recall freely, their remembering can be highly accurate – in some cases more accurate than the documented record. However, when these same people are asked specific questions, some of the answers can be inaccurate – if they’re not in primary memory. To answer such questions, people must consult integrated memory or make educated guesses, using the questions as retrieval cues. That’s why witnesses to a crime should first tell their own memories in free recall, uninterrupted, before being asked specific questions.
Finding Truth in Memory
Our memories are not creative inventions that change kaleidoscopically every time we access them. We have not evolved to misremember the world. When initially remembering an event, we form primary memories consisting of perceptual images, emotions, and physiological experiences. We then use these primary memories to form integrated memories, bringing together similar specific memories, general knowledge, and self-concepts – while also taking into account retrieval demands. The main function of these integrated memories is to be useful to us in remembering the past.
If we want to maximize memory’s accuracy, we can access the representations of our original experience and bypass most of the constructive processes that shape integrated memories. We can treat our memories as subjects of study, describing them carefully and comprehensively.2 This effortful, extensive recall is something we can all do when remembering our past.3 We truly can depend on our memories.
References
1. The landmark self-studies of Hermann Ebbinghaus in the late 19th century founded the field of memory research. This tradition continued into the 1970s and 1980s with the heroic self-studies of Marigold Linton and others, who diligently recorded and analyzed their own memories.
2. Ross, M. (1989). Relation of implicit theories to the construction of personal histories. Psychological Review, 96 (2), 341-357.
3. Winograd, E. (1994). The authenticity and utility of memories. In U. Neisser & R. Fivush (Eds.), The remembering self: Construction and accuracy in the self-narrative (pp. 243-251). New York: Cambridge University Press.
