Iconic memory
Iconic memory is visual sensory memory. It holds a brief, detailed image of what you just saw for well under a second before it fades. George Sperling showed in 1960 that it briefly stores more than you can report, and Ulric Neisser gave it its name in 1967. Only what you attend to moves on to short-term memory.
Drafted with AI by the StudyPDF team. We check dates, numbers and research claims against official sources and read pages on a sample basis. How we write these pages
In more detail
Iconic memory is the visual part of sensory memory, the first stage in the Atkinson and Shiffrin model. It works automatically, holds a lot, and keeps it only very briefly. That short overlap is why a waved sparkler leaves a trail of light and why the world does not go black every time you blink.
Sperling flashed a grid of 12 letters for 50 milliseconds. Asked to report all of them, people named only about 4. Then he played a tone right after the flash to say which row to report. People got most of any row they were cued to, which means about three quarters of the grid was briefly available. When the tone came about a second later, the advantage was gone. The image had faded before people could read it all out.
For studying, the lesson is that looking is not learning. A diagram you glance at fills iconic memory for a moment and then disappears unless you attend to it and work with it. That is why you can look at the Krebs cycle ten times and still not be able to draw it. Redrawing from memory forces the image into working memory and from there into long-term memory.
Sperling's 1960 experiment in numbers
| Condition | What people did | Result |
|---|---|---|
| Whole report | Name every letter of a 12-letter grid shown for 50 ms | About 4 letters |
| Partial report, tone right after the flash | Name only the row the tone pointed to | Most of any row, so about 9 of 12 letters were available |
| Partial report, tone about 1 second later | Same task, later cue | Advantage gone, no better than whole report |
Example
Tom stares at a labelled heart diagram for two minutes before his anatomy quiz. He closes the book and can picture the shape, but he cannot say where the pulmonary veins enter. Each glance filled his iconic memory for a moment, but he never turned the picture into something he could recall. The next day he covers the labels, draws the heart from memory, checks it and fixes the gaps. After three rounds he can label it without looking.
How to use it
- 1
Look at the diagram with a goal: which parts, in which order, connected how.
- 2
Cover it and redraw it from memory, even if it is rough.
- 3
Compare with the original and mark what you missed or put in the wrong place.
- 4
Redraw only the missing parts, then the whole thing once more.
- 5
Repeat the next day and a few days later, without looking first.
How StudyPDF does this
When your slides contain labelled diagrams, StudyPDF can build image occlusion flashcards from them: the labels are masked and you recall each one, which is redrawing from memory in card form. The diagrams come from your own slides, and the decks export to Anki's native Image Occlusion.
Sources
See also
Common questions
How long does iconic memory last?
Well under a second. In Sperling's experiments, the extra information people could report was gone when the cue came about one second after the flash, and many estimates for the usable image are a few hundred milliseconds. Echoic memory for sounds lasts longer, about 3 to 4 seconds.
What is the difference between iconic and echoic memory?
Both are sensory memory. Iconic memory holds what you see and fades in well under a second. Echoic memory holds what you hear and lasts about 3 to 4 seconds. That makes sense, because speech unfolds over time and you need a few seconds of it to understand a sentence.
Put it to work on your own course
Upload your lectures and ask Bo. Cited answers, plus flashcards, exams, and more built from your material.