Visual Memories Linger

If you want people to remember, give them an image.

Posted Nov 26, 2020

In the early 1960s, Canadian psychologist Allan Paivio and his colleagues noticed that people recall words better on psychological tests when those words elicit visual mental images. People often remember concrete items better than abstract ones, maybe because specific, material terms “more easily evoke imagery” (Paivio 1991, 347). When many people hear words, the “click of comprehension” comes when they see something in their mind’s eye (Paivio 1991, 108). Paivio’s experiments indicate that “picturable material is easier to remember than less picturable material” (Paivio 1991, 258). For memory, the meanings of words may matter less than their ability to inspire mental pictures.

Katie, "Dual Coding Theory." March 16, 2011.
Allan Paivio
Source: Katie, "Dual Coding Theory." March 16, 2011.

Paivio went on to propose the dual coding hypothesis, which maintains that people encode memories in “independent but partly interconnected” word- and image-based symbolic systems (Paivio 1991, 326). Seeing a frog brings to mind the word “frog” (in whatever language one speaks), and hearing or reading “frog” may evoke a mental picture of a frog. Paivio’s evidence for the cognitive role of visual mental imagery played an important part in the imagery debates between 1975-1995 when psychologists argued over whether the visual mental images people experience reflect the neural processes underlying cognition. It surprised Paivio that, given how evolutionarily recent written language is, people encode anything verbally at all (Paivio 1991, 343). He hypothesized that language, which aided human cognitive development, commandeered skills that had already evolved, such as fine visual and auditory analysis. Maybe the word- and image-based coding systems were interdependent, and language was so successful because words can call forth images.

In the 50 years since Paivio proposed dual coding theory, our understanding of memory has evolved considerably. Metaphors for memory often draw on current technologies, and in the digital age, references to "storage" — of paper files in file cabinets, for instance--make less and less sense. Brain imaging experiments now suggest the widespread neural activity of people consciously experiencing memories. Like a vivid, imagined scenario, an episodic memory (linked to a particular time and place) involves the partial recreation of a complex pattern of bodily feedback (Barsalou 2008). Rather than comparing memories to objects stored in warehouses, one might compare them to songs like the Doors’ “Light My Fire”: a recognizable pattern of excitement that is never performed the same way twice.

Novelists and poets have not needed fMRI experiments to know that evocative language lingers in the mind. Words that inspire mental images “stick” — a bad metaphor, since no mind is a sticky, planar surface with finite space to occupy. Better said, image-inspiring words create songs that beg to be sung over and over again. Fiction-writer Flannery O’Connor wrote that “the nature of fiction is in large measure determined by the nature of our perceptive apparatus. The beginning of human knowledge is through the senses, and the fiction writer begins where human perception begins. He appeals through the senses, and you cannot appeal to the senses with abstractions” (O’Connor 1969, 67). To keep readers immersed in a story, to make them care about characters, don’t write about love, forgiveness, and the future. Write about a blue blanket.

"Toni-Morrison-SM-2" by cjdrexel. Creative Commons. Public Domain.
Toni Morrison
Source: "Toni-Morrison-SM-2" by cjdrexel. Creative Commons. Public Domain.

Toni Morrison ends her novel Jazz with the narrator contemplating her characters. She depicts Violet and Joe, who have had a troubled history, lying under an old quilt, and she reports their plan to buy a lovely new blanket:

“… they played poker just the two of them until it was time to go to bed under the quilt they plan to tear into its original scraps right soon and get a nice wool blanket with a satin hem. Powder blue, maybe, although that would be risky with the soot flying and all, but Joe is partial to blue. He wants to slip under it and hold on to her. Take her hand and put it on his chest, his stomach. He wants to imagine, as he lies with her in the dark, the shapes their bodies make the blue stuff do. Violet doesn’t care what color it is, so long as under their chins that avenue of no-question-about-it satin cools their lava forever” (Morrison 2004, 224).

As these scarred characters face their future together, Morrison gives them —and the reader — something to see and touch. Readers can share the characters’ sensations as they discuss the blue color, as Joe pictures their shapes under it, and as Violet imagines cool satin against her chin. Morrison crafts paragraphs brilliantly, and this one progresses toward the word “forever.” In the immediate context, Violet is imagining a time when their love will be more affectionate than sexual, but both characters are thinking about a shared future. I first read Jazz in 2017, and through the three battering years since then, I have remembered the blue blanket. The way Morrison evoked its look and feel kept it alive in my mind as though I had experienced it personally.

When Paivio conceived of dual coding, he emphasized visual mental imagery, but imagery draws on every sensory modality, usually in combinations (Starr 2013, 78). Memories, like the experiences they encode, consist of everything a body senses at the time. Today, Paivio's hypothesis might be adjusted to say that words become more memorable if they evoke multisensory imagery with a strong visual component.

When I advise students seeking academic jobs, I urge them to describe themselves in a classroom so that their interviewers can see them teaching. The evidence from literature matches that coming from labs: Language that evokes imagery of any kind is more memorable. In this time of remote learning, it may be hard to remember life in classrooms, but human memory and imagination are working fine. Memories and imagined scenes come from sensations, and it takes sensations to draw people into them.


Barsalou, Lawrence W. (2008). “Grounded Cognition.” Annual Review of Psychology 59: 617-45.

Morrison, Toni. (2004). Jazz. New York: Vintage-Random House.

O’Connor, Flannery. (1969). “The Nature and Aim of Fiction.” Mystery and Manners. Edited by Sally and Robert Fitzgerald. New York: Farrar, Strauss, Giroux.

Paivio, Allan. (1991). Images in Mind: The Evolution of a Theory. New York: Harvester-Wheatsheaf.

Starr, G. Gabrielle. (2013). Feeling Beauty: The Neuroscience of Aesthetic Experience. Cambridge, MA: MIT Press.