cc licensed ( BY-NC-SA ) flickr photo shared by Ric e Ette

I am sure you have across this statement in blog posts, presentations promoting visual communication styles; it often cited as a “fact” (yep, I am using those kind of quotes):

Research at 3M Corporation concluded that we process visuals 60000 times faster than text.

In May 2012 I started digging back the layers to try and find the source of this research. By July all I had found was circular references, dead ends, and just whiffs of a trail. There is still a significant cash prize ($60, what do you think I am made of? that seems generous) on the table for the person who connects me to the actual research.

The 60 clams are sill in my wallet.

This week, Dean Shareski gave it a valiant effort by calling out (scare quotes coming) a Social Media “Scientist” who used it again, as “Fact”

Here is where the fact is asserted

b2 crap

She even is helping cement more truthiness by making the claim tweetable


No I won’t tweet that. You should not either.

A scientist, IMHO, backs conclusions with facts, observations, data, valid references to support it. That is how science works. So let’s follow the trail; the link provided leads us to “We process visuals 60,000 times faster than text ““ Here’s WHY” (SAP Visual Enterprise, Dec 2011)

Visuals help to increase efficiency and effectiveness of communication. That is nothing new, especially for regular readers of this blog. But did you know that in fact we process visuals 60,000 times faster than text?

Even I was a bit surprised when I saw that number in an article written by Mike Parkinson, founder of Billion Dollar Graphics (BDG), a company that offers organizations and individuals tools to be more successful using effective visual communication. This article titled “The Power of Visual Communication”, can be found on the BDG website together with some other interesting information about visual communication.

The WHY gets less interesting to me, as the claim is left hanging in the wind. Oh let’s keep following it, where we get to Mike Parkinson’s Billion Dollar Graphics The Power of Visual Communication. Someone with that much money surely has an answer. The article is chock full of stuff, and at least does have real bonafide references at the bottom.

But there is no citation at all for the claim; it appears once again as an assertion, an assumed, a fact

So visuals are processed 60,000 times faster than text, graphics quickly affect our emotions, and our emotions greatly affect our decision-making. If most of our decisions are based on relatively quick intuitional judgment and emotions, then how many decisions are influenced by visually appealing, easily digested graphics? The answer is no secret to advertisers.

Dead end, again.

You end up seeing stuff like from the Visual teaching Alliance “Professional Development for Primary, Secondary & University Educators/Administrators” surely they provide solid citations, it is cited as a FACT (all caps theirs, not mine):


You find stuff like in Teaching New Literacies in Grades 4-6: Resources for 21st-century Classrooms


But when you read Burmark’s book (the segment found in Google Books), you do not find the reference, just the repeated refrain


It goes on and on. Even when cited properly in MLA format, when you go to the “source” the 2001 3M document you get is not a research summary, not product of their R&D, but a sales brochure:

Did you know that visual aids have been found to improve learning by up to 400 percent? Did you realize that we can process visuals 60,000 times faster than text? Would you guess that the average person only remembers about a fifth of what they hear?

These findings from behavioral research confirm our daily experience: we rely on all our senses to bring ideas and concepts to life.

That’s it. It is not a source, but yet another in 60,000 chains of repetitions of the claim.

I hand it to Dean, he left a comment on the Social Social Scientists post and then the twitter conversation got a bit heated

This is the SCAREQUOTES “improvements” the author added:


It should be noted that some people have discrepancies with the above stat. If you’d like more information on this, please see the comment below by Dean and follow his link for more research.)

F**** yeah, people have discrepancies with it, cause there is zero evidence to support it. Is that a “Low Down”? This is the justification for repeating it

Thanks for sharing your opinion, Dean. I too have heard this fact both in speeches from Facebook, LinkedIn and Instagram at SMWNYC this past February as well as in text (as linked). In any case, it makes sense that we process pictures faster than text ““ and whether that’s 60,000 times faster or 6,000 times faster it’s hugely important for brands to understand that in our noisy, fast-paced online world it’s important to utilize images, as well as “listen” to them for visual literacy.

In lieu of a direct link to a reference that supports the claim, we have a better “scientific” verification “peeches from Facebook, LinkedIn and Instagram”

The search goes on. I have yet to find proof that it exists yet that is not enough to be certain it doesn’t. Darren Kuropatwa got the closest at least finding the name of Jenn Manalo, Sr. Product Specialist, 3M Corp, who in a 1998 presentation mentioned the 60,000 number as a fact.

What seems likely was there was some internal 3M research. The one most closely associated with it was a 1986 paper by Vogel eta al Persuasion and the Role of Visual Presentation Support: The UM/3M Study which concluded:

Presentations using visual aids were found to be 43% more persuasive than unaided presentations.

This is not even in the neighborhood of the first of the 3M (2001) statements “Did you know that visual aids have been found to improve learning by up to 400 percent?” although “more persuasive” should not be equated to improving learning, and the numbers are wrong. 43% more persuasive is a bit less than 1.5 times more persuasive.

You will find this one repeated too, though it is not as powerful as the 60,000 one, but at least this one was a research study. But if you take this as a fact, and extrapolate a 1980s university study that compared a group of students who watched presentations, some with images, some without, to 2014 as a direct “Fact”… well., lemme show you some Arizona beach front property. It was one study, with media and participants 30 years ago.

I did get in touch by email with the study’s author, and as someone in the middle of 3M sponsored era of similar work, Doug Vogel was pretty darn close to what was going on:

The research that I did as a PhD student at the U. of Minnesota was involved with persuasion and the working paper that you found is actually the most complete description of the work (even beyond that which was ultimately reported in my thesis). I have not seen the 2001 3M publication but my research had nothing to do with visual processing speed.

I do have an inside contact, my ds106 colleague Rochelle Lockridge who is a strategic analyst for 3M. She has been asking the department head at 3M who manages this area of research, and has not yet been able to locate the source of the original 60,000 time research (Rochelle is still inquiring, all it means is that no one at 3M she has asked seems to have any recollection of the study).

The ironic thing is that it is actually a rather ludicrous claim in the first place. What do we mean by “information being processed” be it visual or text? Does it mean recognizing visual patterns, facial recognition which studies do show the brain can do very fast. Is that information? What does it mean to be “processed”? Recognized or understood” Is it conceptual? Do we see knowledge visually?

It’s almost like claiming that rock music can be processed more quickly than than primary colors. The comparison is just… pointless.

But if you blog it, cite it, tweet it 60,000 times; if “Facebook, LinkedIn and Instagram” say it is so, does that make it a fact?

***** no.

I am not invested in the proof or non proof of the claim, it is this willingness of people to repeat something as fact without going any farther then the previous blog post, status update, or Buzzfeed listicle to repeat it.

I have no illusions that this will ever stop. I applaud efforts to point this out to others, but it is tireless, thankless work. In the rush of being a social media maven, there’s little time to bother with pesky things like backing up your claims.

If course I believe visual information is important.

It’s why I teach digital storytelling, why I take so many photos, why I present on it.

But I do not lean on a baseless claim masquerading as a fact to make the assertion. Do you?

How does it feel to be set free?

cc licensed ( BY-NC ) flickr photo shared by Mindsay Mohan

If this kind of stuff has value, please support me by tossing a one time PayPal kibble or monthly on Patreon
Become a patron at Patreon!
Profile Picture for CogDog The Blog
An early 90s builder of the web and blogging Alan Levine barks at on web storytelling (#ds106 #4life), photography, bending WordPress, and serendipity in the infinite internet river. He thinks it's weird to write about himself in the third person. And he is 100% into the Fediverse (or tells himself so)


  1. Nice post. A few years ago I tried to track down the Toffler quote about the illiterate of the future that appears on his website. Turns out its a graphic designers mashup. Existed in parts but not a published whole. Don’t quote on that as they say. Can I have $60 clams please.

  2. Hi Alan,

    I look after the SAP Visual Enterprise blog that has apparently elicited your anger! Feel free to ask me anything (AMA for fellow redditors).

    My predecessor posted this particular item in 2011 and (apparently) did not do any fact checking. When I was first alerted to this oversight I decided to review the literature in the area. I also decided to leave the original post up for posterity and in the hope that it would stimulate conversation ~ not antagonize you.

    Alan, you asked what is “information being processed”. To be honest, the answer doesn’t reliably exist. Generally, the layman would consider two primary mental processes to comprise “information being processed”.
    1) Perception: stimuli from the five senses interpreted by the relevant part of the brain (cortex) ~ “what is this?”
    2) Processing: what meaning can be drawn from what I perceive ~ “what does this mean?”

    Neuroscience researchers generally consider the ‘formation of a response to stimuli’ (typically an fMRI test) strictly as the end of “information being processed” ~ i.e. the subject has processed enough information to formulate a response. E.g. with pattern recognition when the brain recognizes a pattern (perception ~ “what is this?”). Behavioral decision theorists would go with the period from when the subject is stimulated to when the subject acts. E.g. with facial recognition you could be asked whether you trust that person (processing ~ “what does this mean?”). However, if you really wanted to get into it I would suggest reading Rene Descartes, Immanuel Kant, Karl Popper, and, Michael Polanyi (philosophers) if you would like to really get into thinking about “information being processed”. It’s messy.

    As to the issue at hand, Tony Parkin commented that this 60,000x issue is peculiarly widespread among “facts” relating to “visuals”. For example, Edgar Dale’s ‘Cone of Learning’ as presented in his book “Audiovisual Methods in Teaching” (1969) never had percentages associated with it ~ he was merely spatially depicting his conception of a comparison between different teaching tools in the spirit of his book. See this PDF for a coherent explanation and introduction to the forces at work:

    People “know” that visual information is “better” than textual information, however, the reality is vastly more complicated than that as implied by the picture on pg.10 of the above PDF. These fantastical extensions of observational and/or conceptual reports are (1) easy to communicate, (2) valuable, (3) convenient, and (4) remarkable (in the marketing sense) which is why people persist in parroting them. By comparison, academic journals require you to take the time to read a large body of knowledge, and, there are expensive pay-walls which many people are disinclined to get past.

    If you do read academic research you will find the inconvenient reality that the answer to text vs. visuals is….. IT DEPENDS (not very surprisingly and rather anti-climatically). The core issue appears to be ‘what is being communicated?’.

    If we apply this to the 60,000x issue, due to the brain’s structure, text is (conceptually) comparatively more difficult to process as it uses more parts of the brain to do it. This is because we have to use our visual cortex to make sense of the letters and there specific arrangement and then move this piecemeal into the auditory cortex via working memory which, using its linguistic prowess, reconstructs (sequentially processes) the meaning of the text in the ‘language of the brain’ (i.e. memory) before we can do anything with it. By comparison when a visual is processed (simultaneous processing) by the visual cortex we already have it in a form that is usable. The question is whether that specific form is usable for our specific purpose.

    For example, imagine I buy some IKEA furniture and I want to assemble it. The text + static image instructions are frustrating as I have to turn text into procedural-motor actions (physically move stuff). Instead I go straight to YouTube and search for my product and bam, there’s someone who shows me step by step how to assemble it and what to watch out for. Why do I prefer visual information in this case? I’m performing a spatial task and visual information communicates spatial information more effectively than text + static images. A meta-analysis published recently supports this by concluding that, for instructions communicating a procedural-motor task, functional animation is 106% more effective than text + static images (Source: Höffler, T. N. & Leutner, D. (2007) Instructional animation versus static pictures: A meta-analysis, Learning and Instruction, 17(6), pp. 722-738.).

    Now, if I wanted to know the precise length of one of the screws, no matter how you constructed an image or animation I would rather have the actual number. However, if you gave me that number in the context of a picture of the screw and a couple of lines articulating the spatial relationship between the number and the screw that would probably be even more helpful as then I would know what the number referred to ~ i.e. context.

    In sum, visuals are good for providing spatial context (size comparisons, how to do manipulate a part into position, what a thing looks like, etc), text is great for specificity and precision ~ as long as you know what the author is getting at. What the world needs is more of this ~ bloggers such as ourselves to honestly interpret complex academic research.

    The last thing I would like to say is that I read a story (which for the life of me I cannot find) which mentioned that the 60,000 number came from comparing the file size of some text (few bytes) to the file size of a (somehow) comparable image (few megabytes) and then, all things being equal (time taken), a person can process visual data 60,000x faster than text as measured in quantifiable computational terms.

    1. Thanks Baden, for such a detailed response, maybe some of the clearest writing on the topic; especially on the nuances if processing (let’s let the philosophers off the hook). I’m not angry at your blog, it did nothing that countless other sites do. I appreciate deeply that you preserved the original knowing the topic was problematic (maybe it could be edited with a note?) thanks again for perhaps the most informative comments ever

  3. You’re most welcome Alan. A comment is a good idea, I’ll do that today, perhaps something like this:

    “It should be noted that some people have [identified] discrepancies with the above stat. This, perhaps, reflects the results of several extensive investigations which conclude that there is no actual research supporting said stat.”

  4. Kia ora, Alan

    The problem is not confined to “processing” visuals or texts (whatever that means). Nor is it a new one, but it is mercurially amplified by social media. The rash publishing of inaccurate and spurious information has been going on for centuries and even within so-called scientific circles. It follows the principle that if enough people are grouped together with one belief over a matter at issue or just the appearance of that concomitant phenomenon, there will be an automatic further accumulation of followers leading to an ever increasing group size. “Safety in numbers” is a phrase once coined to describe the security of many people following the same mission or principle and often no check is put in place to validate the process contributing to the growth of the group. Another is, “to them that hath more shall be given”. Examples of this are legion among the people of the world. Alas, it has always been that way.


      1. OK, Alan, you’re an optimist.

        Looking at it your way, there must be a finite possibility you’d be right. It’s just simple statistics, isn’t it?

        You’d have to go along with accepted statistical principles, of course and that would be following consensus belief. You wouldn’t want to do that without testing it, would you?

        Is that what you call ‘scrutiny in numbers’?


    1. Kia ora ‘Sisyphean Writer’

      I thought the post was not so much about uncovering the source of a fact but about the way it is communicated, wherever the origin may have been, in a cascade fashion of huge magnitude without regard to its source or validity. I stand corrected if my interpretation is erroneous.


Leave a Reply

Your email address will not be published. Required fields are marked *