Recently, I’ve run into the claim that images are processed 60K times faster than text. And, folks, it’s a myth. More over, it’s exemplary of bad practices in business. And so it’s worth pointing out what the situation is, why it’s happening, and why you should be on guard.
It’s easy to find the myth. Just search on “images processed 60K times faster than text”. You’ll get lots of citations, and a few debunkings. Most of the references are from marketing hype, selling you visual support.
The origin is suspicious. It’s always cited as coming from 3M, Polishing Your Presentation. Which is problematic, because when you go to that paper, you find the quote, but not a legitimate citation. Instead, there’s a vague statement about “findings from behavioral research” with no citation! Bad form.
A study done jointly between 3M and the University of Minnesota about presentations also is potentially a source. With only one small catch: it doesn’t mention 60K at all! Instead, it does conclude that “Presentations using visual aids were found to be 43% MORE PERSUASIVE than unaided presentations.” Which is hardly controversial.
Yet this is another zombie, like learning styles, that won’t die! It’s been researched by several folks, including Alan Levine and Jonathan Schwabish. No one seems to be able to identify a real piece of evidence. And it just doesn’t make sense!
In use, words are practiced enough to be recognized as a whole, serving as icons; they’re not repeatedly processed from letters into words. Second, images need parsing, too, and contextualization between the image and the current semantics.
Sure, we have many more neurons devoted to image processing than auditory, but that’s not only due to a sensory primacy (e.g. distance capability), but also the richness of the visual field. And more doesn’t equal ‘faster’. Yes, we’re processing in parallel, but nerve firing rates change based upon activation, not modality.
And this means that we have to have our ‘hype’ shields up. We need to evaluate any claims by several methods. Who else is saying this? Not pointing to the same (bad) data, but what convergent evidence is there? And what vested interest do the promoters have? And, importantly, does it make causal sense? Is there a plausible scenario when you dig beneath the surface features?
And, if you don’t want to read research published in the original academese, find those who you can trust. Those who’ve demonstrated a consistent ability to cut through the hype and the research, and bring good interpretations and debunk the myths. You can see my list of mythbusters here.
So, please, practice professionalism, avoid the hype, and use good principles in design and practice.
Thanks for tossing the signal flag, Clark. The 60000 times faster assertion is a zombie that will not die.
For the record the 3M reference often cited (I’ve seen it used in textbooks and research papers) is a PDF of a promotional brochure.
I have a direct email from Doug Vogel, the author of the 3M and the University of Minnesota study, who assured me his work had no connection to the claim.
I have it nailed back to a 1982 Business Week article by Philip Cooper
https://cogdogblog.com/2015/03/dialed-back-to-1982/
He currently teaches/works at MIT Sloan School. I’ve called/emailed but no answer. I’ve asked people in Boston to go knock on his door.
If anyone has the lead, it’s him. If you or anyone of your readers can nail this down, there is a huge $60 prize sitting on the table.
If the 60,000 times faster claim were true, don’t you think we would have already come up with adequate diagrams and illustrations to accelerate school learning? I mean, it’s like looking at pictures turns our brains into quantum computers? We could have 9 year old Phd’s as our educational standard. Wow, at my age, I could have at least 7 doctorates and countless certifications! I probably would have fulfilled my secret life-long dream to be a robotic-rocket-submarine-and-gene-therapy designer by now. “Fake news” is not limited to the political arena alone; in fact quacks have been making ludicrous claims for centuries, legitimizing their claims with “science”.
Do people really believe this nonsense?
I heard this claim mentioned in documentary. I think it could be easy to create this data in a limited context. For example, an image of a dog vs. a lengthy description. Big increases sound impressive if the original is small number. For example, maybe they had to identify 20 attributes to “pass” and could go back to view the prompt until they did. i.e. re-reading vs. a glance.
All of this to say, this is why sources are so important. I was looking for one and found you.
But that isn’t why I am writing. I know there are a lot of studies around image reconstruction by measuring brain activity during a visual input. They are surprisingly good. My question is, have they tried this with reading? I have some theories on how this would turn out. I checked but couldn’t find anything.
Zoe, for more on trying to track down the 60K topic, I’d look to Alan Levine (https://cogdogblog.com/2015/03/dialed-back-to-1982/). As to reading brain activity during reading, I don’t know about any such research, not because it doesn’t exist, but because it’s not an area I track deeply. I’d ask some reading or perception researchers. Good luck!