Do Goldfish have a longer attention span than American Internet Users?

If what you read on the Internet is to be trusted – and let’s face it the Internet never lies, then I have around eight seconds to convince you that this article is worth reading because most of you have an average attention span that’s slightly worse than that of a goldfish.


With that picture of a goldfish in bowl added to the mix that’s about eight seconds’ worth of content and we’ve now ditched all the fish-brained slackers, so if you’re still here then hopefully you might be interested in finding out why this whole idea that Internet is somehow altering people’s brains and reducing their attention span over time is nothing short of a load of foetid dingo’s kidneys.

Where shall we start..?

Well, how about with the article which most recently brought this particular claim to my attention and the claim one of the reasons that the Internet is allegedly “killing” satire is the diminishing attention span of the audience:

The second big contributor to satire-blindness is our diminishing attention span. The average American attention span in 2000 was 12 seconds; in 2013, it was eight seconds. This is less than the average attention span of a goldfish (nine seconds).

The provenance of this article is actually quite interesting. The author, Arwa Madahwi, is billed as a “strategy director” with a company called Contagious Communications which, in turn, describes its role in business life as helping “brands and advertising agencies understand and adapt to shifts in marketing, consumer culture and technology”. As supporting “evidence” for this claim, Madahwhi’s article supplies a link to an article/podcast on another marketing website which, in turn, cites a historian at Harvard’s Business School as its primary source:

Publishers are now rolling out shorter books faster and faster, and Harvard Business School historian Nancy Koehn says it’s largely because we’re getting less attentive and more anxious:

“The average American attention span in 2013 was about 8 seconds. The average attention span in 2000 was 12 seconds. And then get this kicker – the average attention of a goldfish is 9 seconds.”

Hopefully you’ll notice a couple of things here.

One is the complete lack of any credible scientific sources or links back to primary evidence to back up any of these claims, which is sadly far from unusual these days.

The other that even this very short chain of sources cited here reads like part of the passenger manifest of the Golgafrincham B-Ark and that’s true of the vast majority of people and other online sources who’ve been busily disseminating this claim. They’re people who work in marketing, online communications and public relations. People who have one very obvious thing in common; they spend much of their working lives trading in bullshit.

So let’s pick this apart and we’ll start with easily the most glaring howler in all this, which – somewhat ironically – happens to be the claim that the average attention span of a goldfish is just nine seconds.

Is that actually true?

You know what, I’ve absolutely no idea… and neither it seems does anyone else, least of all the people who make claims about the attention span of goldfish, and that’s because, to the best of my knowledge, no has ever bothered to try and measure their attention span. The great urban myth relating to the cognitive performance of goldfish, which circulated for many years before it was roundly debunked by researchers at the University of Plymouth in 2003, was that they have a very short memory span – usually cited as just three seconds – but memory and attention are two entirely different things.

So, right from the off, the goldfish comparison turns out to be utter nonsense and with that out of way we can safely turn to the question of our own species and address the question of what we do – and don’t – know about our own attentional capabilities.

Now, to begin with, we need to understand that there are two basic types of attention we need to bear in mind.

One of these is focussed attention and this is the type of attention that we’re most concerned with because this what enables us to concentrate on performing a specific task over a significant period of time. If you sit down to read a book, watch a film or write an article, as I’m doing here, then that requires focussed attention. You need concentrate on what you’re doing over a period to time in order to complete the task successfully.

The second type of attention is usually referred to as transient attention and that’s a very different mechanism, one which serves a very different purpose. Transient attention is reactive, it’s an automatic response triggered by sudden changes in your immediate environment such as an unexpected and unfamiliar noise or a sudden movement that you just happen to catch in your peripheral vision. It’s an evolved mechanism that acts over a very short period to time – a few seconds at most – to draw your attention away from whatever it is you happen to be doing at the time towards the source of that sudden change in your immediate environment for just long enough for you to decide whether it and/or its apparent cause requires your focussed attention – because it presents some sort of potential threat (e.g. a predator) or opportunity (e.g. food)  – or whether its a wholly benign change that can be safely ignored.

So, as a species, we are easily distracted by sudden and unexpected changes in our immediate environment for what are entirely sound evolutionary reasons. Were that not the case then our species wouldn’t be here because our distant ancestors – and we are talking very distant here as in going all the way back to the first organisms that developed sensory capabilities – would have quickly become extinct due to competition from other organisms, the one’s with the sensory capabilities that our (hypothetical) ancestors lacked.

So that’s the basic framework for attention. Transient attention is short-lived, reactive and is triggered by sudden environmental changes while focussed attention operates over significant periods of time and requires a degree of motivation and conscious effort and its the latter that we are primarily interested in whenever we talk about measuring people’s “attention span”.

So, what do we actually know about focussed attention and, in particular, how long it can be sustained in the absence of significant distractions?

In truth the answer is rather less than most people think we know, especially teachers and people who work in marketing and advertising.

To start with, we need to go back to the days before the Internet (and Internet statistics) and look at what people though they knew based on psychological and educational research and the prevailing view back then, so far as the attention span of older teenagers and adults was concerned, held that when performing tasks involving a concentrated flow of information over a period of time, such as attending a lecture and taking notes, people could sustain their concentration on such a task for around 10-15 minutes before their attention would waver for a couple of minute or two, allowing the brain to “reset” itself in to state where it was ready focus on the next block of information for around another 10-15 minutes. That was certainly the prevailing view when I was attending psychology lectures at university over 25 years ago and, in fact, there was one lecturer in particular who made something of performance, in the introductory lecture to the course he taught, of informing students that all his lectures were structured around around 10-12 minute blocks of information with little Q&As, visual aids and other odds and ends in between because research had shown that this was the best way to ensure that students were able to maintain their focus and concentration throughout the entire lecture.

This idea, that our species’ peak attention span lasts for around 10-15 minutes at most but can be sustained, effectively, over much longer durations by inserting short breaks and changes in tempo or in the type of activity every 12-15 minutes, giving the brain time to reset its focus, emerged during the 1970s from studies of student behaviour in lectures, particularly note-taking and quickly found its way into the literature on teaching practice and teacher training where its been widely and uncritically repeated ever since. Well uncritically, at least, up until 2007 and the publication of a review paper by Karen Wilson and James H. Korn of Saint Louis University – “Attention During Lectures: Beyond Ten Minutes” – in which the authors posed the obvious question: How good is the evidence that underpins this idea?

And the answer to that question, it turns turns out, is “not very good at all” or rather, as the paper’s abstract notes:

…we reviewed several types of studies including studies of student note taking, observations of students during lectures, and self-reports of student attention, as well as studies using physiological measures of attention. We found that the research on which this estimate is based provides little support for the belief that students’ attention declines after 10 to 15 min. Most studies failed to account for individual differences in attention.

Please do take the opportunity to download and read the full paper (which I’ve linked to above). It’s not particularly long (5 pages including references) nor is it overly heavy on psychological jargon so it’s reasonably easy to follow without too many forays into looking things up on Wikipedia if you’ve not got any kind of background in psychology at all.

If, however, you must insist on a tl;dr summary of its contents then what Wilson and Korn found is that although there are some functional limitations to our attentional capacity – we know from a wide range of other research, for example, there are limits to how much information we can process through working memory in any given space of time and that too much information at once will lead to cognitive overload (and if you want to know more about that then one of the best place to start is Chabris & Simons’ “The Invisible Gorilla“) – but beyond that not only can there be there considerable differences in attention span between individuals but those differences can depend on a wide range of individual factors including personal motivation, physiological and psychological arousal, presentational style, personal energy levels (i.e whether we’re feeling alert or tired or, in the case of some university students, hung over), etc.

The bottom line is simply that if we’re feeling good about ourselves and alert and full of energy and we’re presented with a task were motivated to carry out and its something we find genuinely interesting and compelling then we can focus our attention very effectively on that task for very long periods of time and not just in terms of minutes but hours; and if every so often during that task our mind takes a brief wander to “reset” itself ready for another period of intense concentration then this often something we don’t even notice as we’re doing it. On the other hand, if we’re tired and the task we’ve given to do is one we find dull and boring and its something we don’t really care about then our minds will tend to wander all over place and we’ll find it difficult to concentrate on that task because it just doesn’t engage our interest and we’re not in the least bit motivated to see it through.

That the reality here and it has a number of significant implications for the claim that attention spans are somehow getting shorter over time not least because if you are going to assert that people’s attention spans are getting shorter over time then the absolutely the first thing you need to sustain that claim is an accurate measurement what people’s attention spans used to be, for example, thirty or forty years ago. If you don’t have that baseline measurement – and Wilson and Korn’s review strongly suggests that we don’t, or rather it shows that what we thought was the baseline for the last forty years or so has turned out to be unreliable – then its impossible to make any kind of reasonable comparisons or reliably measure any changes over time.

We can’t say, for sure, that people’s attention spans have become shorter over the last forty years because we don’t know, with any degree of reliability, what people’s attention spans were forty years ago and that, in turn, means that any other assertions that might be predicated on the assumption that people’s attention spans have been getting shorter over time, such as the suggestion that the growing use of screen technologies and/or social media are somehow changing young people’s brains and causing their attention span to diminish over time have to be treated as, at best, highly questionable if not simply as a matter of as yet unevidenced conjecture.

Okay, so that’s the general picture vis-a-vis actual research into focussed attention but what about the claim the average attention span of American’s has fallen by a third between 200o and 2012 to just eight seconds. Where did that claim originate and what, if anything, is it actually based on?

Well, the nearest I’ve been able to get to tracking down the original source of this claim turns out – inevitably – to be a marketing brochure produced by a Chicago-based online marketing firm called Zocalo in which you’ll find the following quote:

With the 24-hour news cycle and the proliferation of round-the-clock social media updates, consumers have almost come to expect to hear and see information as it happens. The average time a user spends to digest information has decreased significantly over the past twelve years – from 12 seconds in 2000 to 8 in 2012 (less than the average attention span of a goldfish).* This “hurry up and post” culture is the fundamental reason that, across all channels, we’re seeing a shift to fewer words and more eye-catching visuals.

* Source: The Associated Press, 1.30.2012

Okay, now we at least have a definite source – Associated Press – and some useful context; the eight second figure that others have been citing as a supposed measurement of people’s average attention span is actually a measure of the average amount of time that Internet users allegedly spend digesting information as they encounter it online and from there we can perhaps reasonably infer that the actual figures that AP were citing (12 seconds in 2000 and 8 seconds in 2012) are derived from Internet metrics and analytics data.

So now we know that eight seconds is actually the average amount of time in 2012 that Americans spent looking at individual web pages before moving on to do something else (most likely look at another web page) compared to twelve seconds, which was the average back in 2000 – but what, if anything, is this actually telling about the average attention span of Americans and how this may or may not have changed over the course of a twelve year period from 2000 to 2012?

Not a damn thing.

If you want an explanation for why, on average, American’s spent only eight seconds viewing an individual web page before moving on to something then almost certainly the best and most accurate answer is likely to found in something called Sturgeon’s Law, an adage that was originally coined back in the 1950s by the legendary science fiction author and critic Theodore Sturgeon and which, in its general form, posits that “ninety percent of everything is crap”.

As we’re dealing, here, with the Internet it’s probably more accurate to modify that statement to one which posits that the vast bulk of information and content published on the Internet is absolutely no interest whatsoever the vast majority of Internet users. Eight seconds, the figure incorrectly cited by so many marketing websites as a measure of the average attention span of American Internet users is in reality the average amount of time it takes Americans not to digest information online but to decide whether or not the contents of a web page is of the least bit of interest or relevance to them and whatever it is they might actually be looking for at any given time .

You could, in fact, say that this is average operational speed of an American bullshit detector and you certainly wouldn’t be too far from the truth.

What ever else eight seconds might be, one thing it isn’t is a measurement of people’s average attention span, even if they are Americans. That claim, like so many others you’ll readily come across online, is not just wrong it’s not even wrong.