What Makes Memory Retrieval Work—and Not Work?

Source: Karolina Grabowska/Pexels

Sometimes, we search for information in long-term memory and find it—a name, a movie title, or a vivid example to support a general conclusion. Other times, we're unable to recall what we believe we should know—only to remember it later when we’re no longer searching for it. And sometimes, memories come to mind unbidden, without focus or effort.

What’s actually happening with memory retrieval?

The Act of Retrieval

Memory retrieval requires activating mental representations in long-term memory and copying them into short-term memory, where we consciously experience them. To bring a memory into consciousness, we actively search our long-term memories by sending in retrieval cues that connect to what we want to remember.¹

The underlying idea is that long-term memory is analogous to actual physical space. Information is stored somewhere and then retrieved. We can think of retrieval cues as traveling along pathways to the desired information in memory.

Why Does Retrieval Fail?

Long-term memory is our subconscious, a vast collection of mental representations of names, images, concepts, and specific episodes in our lives. When we can’t retrieve a memory, it’s not because the memory is missing; it’s because we can’t find an effective pathway to that memory.

Contextual Breaks. Finding our way to a memory depends on the context in which we first encoded what we want to remember. This context can be external and sensory—a particular smell, a song, the play of light on a person’s face. Or it can be internal—a mood, a physiological state, or a sequence of thoughts.

We fail to retrieve a memory because of a discrepancy between the retrieval context and the encoding context.

When we’re unhappy, we find it difficult to remember happy events. We can walk into a room and forget why we came in because we made our decision in a different location. We were also cognitively focused on a specific action, something to accomplish, and not on remembering that action later. We don’t often remember dreams because their encoding context is almost entirely internal, and upon awakening, this context vanishes.

Recreating the encoding context during retrieval makes us more likely to remember. When we forget why we entered a room, returning to the original location revives the encoding context and the original purpose.

Interference. When similar memories are more prominent than what we want to recall, they can sidetrack us, sending us down the wrong retrieval pathway. We may struggle to remember an acquaintance named Margot, trying out other possibilities, such as Marlo or Margaret. And when we do find “Margot,” we may think of the actor Margot Robbie, which can then hide the full name we’re searching for.

Source: Katya Wolf/Pexels

Overgrown Pathways. A memory can remain vivid but inaccessible because the pathways to that memory are overgrown with disuse. This is helpful with unpleasant events we don’t want to think about. But it also prevents us from thinking about events that would be helpful to recall. Keeping a journal can provide effective retrieval cues that we would otherwise lose track of.

Delayed Retrieval. Sometimes, we remember after a significant delay. This can be particularly frustrating with names or when we think of something we should have said during a conversation—two hours later.

Delayed retrieval occurs because searching unsuccessfully along ineffective retrieval pathways primes unused pathways nearby. If some of these nearby pathways connect to the memory we want, priming makes them more likely to be activated later, leading to successful recall. Delayed recall often occurs unexpectedly because retrieval processes continue operating without our direct awareness.

This is especially noticeable when we’re in the tip-of-the-tongue state, feeling like we’re on the verge of remembering. We may or may not be close to recalling, but we spend more effort trying to retrieve our desired memory and are therefore more likely to recall it later.

Source: Edmund Dantes/Pexels

In conversation, we’re required to retrieve pertinent information quickly to keep the flow of communication going, so we can’t wait too long for an effective retrieval pathway to appear. If we're frustrated by delayed recall after a completed conversation, it's helpful to know that this can happen to anyone. It's due to the complexity of quickly searching the warehouse of long-term memory.

Memory Essential Reads

Can You Trust Your Memories?

Trying Harder Increases Cognitive Effort and Enhances Memory

With names, delayed retrieval can be resolved by drawing on our reliable memory for faces. One or two devoted sessions of practice associating the face (the retrieval cue) with the name (the sought-after memory) can solve a particular naming problem for good. It's okay if the association is unusual as long as it works for us.

Mnemonics and Retrieval

We use mnemonics to help retrieve names and collections of information. The mnemonic “HOMES” assembles the first letter of each Great Lake into a word so we can use each letter as a retrieval cue. Similarly, many people encode SOH-CAH-TOA to remember the formulas for sine, cosine, and tangent. I learned “Oscar Had A Heap Of Apples,” which stands for OH-AH-OA: opposite over hypotenuse; adjacent over hypotenuse; opposite over adjacent.

To remember the standard biological taxonomy in biology, we can encode “Dear Kate, please come over for great spaghetti” (or any comparable saying): domain, kingdom, phylum, class, order, family, genus, species.

The paradox with mnemonics is that they actually add information to what we want to remember: Instead of remembering only the names of the Great Lakes, we need to remember “homes” and the names of the lakes. I had to learn about Oscar and his apples in addition to knowing the formulas for sine, cosine, and tangent. This paradox clarifies how mnemonics assist memory.

We can easily encode and store an indefinitely large amount of information in long-term memory. Our limitation is with retrieval.

Mnemonics work by giving us only one thing to retrieve. An easily remembered word or phrase that points to what we want to remember. Mnemonics tell us that the secret to remembering is to streamline retrieval.

Involuntary Memory—When Retrieval Has a Life of Its Own

Once in a while, a memory intrudes into consciousness, unrelated to our ongoing activities. The intrusion occurs when a distinctive perception or mental image suddenly reactivates unused retrieval pathways.

Even as retrieval pathways become inaccessible, the memory representations remain vivid and detailed over many years. So when an unused pathway is unexpectedly accessed by a particular smell or an evocative image, a memory we haven’t considered for years can return abruptly, with surprising clarity and detail.²

Improving Retrieval

Involuntary memories teach us that retrieval is enhanced by focusing on sensory details at the time of the events: the tart-bitter taste of grapefruit at breakfast, the tone of our mother’s voice, and the feeling of full-bodied laughter. These sensory details can then retrieve more fully elaborated memories.

Personal memory is also augmented by visiting places from earlier in our lives. With its abundance of retrieval cues, a place is truly a universal petite Madeleine, calling up long-forgotten memories.

The Persistence of Memory

When we can’t remember something we believe we know, we should keep in mind that it’s still in long-term memory, and it will stay there. To recall it, we just need to find an effective retrieval cue.