By Elliot Harris

Ponders user experience. Find him at the twitters @StubbornMinion

You’ve got a great new feature. How do you let your users know?

This post originally appeared as a guest blog on the excellent Nathalie Nahai’s website. Nathalie has since removed the blog section of her site however, and so I’m reposting it here.


Naturally, having spent months researching, designing, and developing a new Awesome Thing for your website, you’ll be wanting to shout about it. ‘What an achievement!’ you say high-fiving yourself, eager to inform the netizens of this fabulous new gift imminently to be bestowed upon them. But did you stop to ask yourself –

Wait! Hold on a second! Before we begin, you might appreciate this street photography of Tokyo. Oh, and you know what’s better than escalators? Slides! Also Nathalie would like to take a moment to tell you all about her super-informative podcast. Check it out! All of it!

Look, I’m sorry about that. You came here expecting a blog post about promoting website features, and I totally interrupted your experience to tell you about utterly brilliant but nonetheless unrequested things. Well, I think they’re brilliant. Perhaps you don’t.

And frequently that’s the problem with website messaging: we want to communicate, but we don’t want to interrupt, and we don’t know who’s interested. Yet oftentimes we’ll distract all our users with the same message, with little thought as to who among them needs to see it, or when.

Promoting a new feature

Let’s imagine you own a freakishly popular video sharing site whose content is predominately amateur video recordings of majestic animals in the wild. And you named it ZooTube.

As the owner of ZooTube, your stats tell you that most of your users come only to watch adorable animals going about their daily business. (Giraffes playing croquet, elephants hosting dinner parties, that sort of thing.) Very few users upload their own videos. If you then deployed powerful new video editing tools, should you promote them to all your users?

Probably not, unless your research suggests that the reason so few people upload videos is because the editing tools are insufficient. Otherwise, you’ll succeed only in interrupting the experiences of a majority of people for whom those tools couldn’t make the slightest bit of difference.

Now, not to sound too Six Million Dollar Man, but we have the technology! Those users who have uploaded videos are the ones more likely to be interested in your new editing tools. You could target those users’ accounts, making your promotion more overt for them, and more subtle for others.

blog-youtube

I honestly wrote all of this before I happened to login to YouTube (you may have heard of it), and was greeted with the message in the screenshot above. I have never uploaded a video. ‘Looking for your Video Manager?’ I had never even heard of the Video Manager!

At least the popup doesn’t appear in a lightbox, which would have prevented me from using the rest of the site. YouTube knows that this message isn’t that important.

Taking care of new users

Ah, now a new user lands on your website. You’ve got no idea what they’re interested in! Also, your site has lots of complex parts, so you think you’d better explain everything before the user gets all confusled. (That being a scenario in which a person is bamboozled by their own confusion.)

This user needs to get a sense of the service you’re offering, and a well-designed homepage will achieve just that. You could even produce a brief video introducing your service in broad terms, for those who’d rather sit back and watch than piece things together themselves.

But be careful of trying to communicate every single feature. Try instead to convey the essence of the product, sufficient to pique the user’s curiosity.

The reason for this caution is that our short-term memory is limited. The headline-grabber on this subject tends to be George Miller’s ‘magical number’ of seven plus or minus two; that being the number of unrelated items that can simultaneously exist in short-term memory (Miller, 1956).This number continues to propagate, though it was later demonstrated to be closer to four plus or minus one (Broadbent, 1975). However, to think in terms of numbers of items is to oversimplify the variously complex factors of perception and attention that ultimately affect short term memory.

The bottom line is that the capacity of short-term memory is both limited and volatile, and often remembers only those things relevant to the individual’s goals at the time. To illustrate this, earlier on you read two examples of animals doing unusual things. Can you remember what they were?

If you can’t, consider that so far there have been several things to remember; that your attention while reading this has been moved to other topics since the animals were mentioned; and that in the context of reading about promoting features on a website, you probably didn’t think that elephants playing croquet was an important point to remember. (Or were they giraffes?)

With these limits in mind, instead of throwing the kitchen sink of features at users the moment they arrive, wait a while, and then nudge them in the direction of those features you think they may not have noticed but might be important to them, based on their actions to date, and your research.

Conclusions

All together, aim for a more graduated, contextual, and discriminating approach to helping your users. Consider their individual experiences, needs, and goals, and tailor your messages to match.

Try as hard as you can to solve problems first through good, research-based design. Like a joke isn’t funny if you’ve got to explain it, if you’ve got to walk users through your user interface, it’s probably not a good one.

Don’t rely on users remembering numerous tips, because short-term memories are limited. If you’re nudging them in the direction of features, don’t reveal all those features in one go. If you’re walking them through a task, avoid giving them all the instructions at once; take them through it one step at a time.

And in general, try talking to users softly before wielding the megaphone.

Bibliography

Broadbent, D. E. (1975). The magical number seven after fifteen years. In A. Kennedy, & A. Wilkes, Studies in long-term memory (pp. 3-18). John Wiley & Sons Ltd.

Miller, G. A. (1956). he magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review , 63 (2), 81-97.

 

Transport for London: Your user experience going down the Tube

You’ll hopefully forgive me for that little play on words in the title. I don’t mean to suggest that London Underground is all that bad. Far from it, in fact; I think the Tube is an amazing service, and though we Londoners often complain about its foibles, we also know that many of us depend on it. That dependency is a testament to the Tube’s utility and fundamental place in London life.

But I think we all know that things could be better. TFL clearly know this as well, as they work to bring about more frequent, reliable, and capacious trains.

All of those improvements are measurable, quantifiable things, and they’re very much worthwhile. But there’s more to the experience than that: there’s the emotional, qualitative experience to consider as well. Of course, these two things are linked – a quantifiably better service would doubtless be perceived as being superior, by subsequently happier customers. But that’s not the whole story.

On the Tube, opportunities for frustration are numerous, and the passengers demanding. A delay of just minutes, a lack of information, misinformation, just missing your train; all these things can (and frequently do) evoke scoffs and unhappy faces from passengers.

And evidently, some passengers take out their frustrations with the service in less-than-constructive ways. Aside from a desire to improve the user experience of the Tube for the sake of customer satisfaction, addressing or managing these frustrations may also result in fewer incidents of staff abuse.

So what could be done to improve how passengers feel? In this blog post, I’d like to focus on an area that, for the most part, could be addressed right away: communication.

The passenger hears many announcements during their journey, and the delivery of these announcements often reveals a staff that is as frustrated and tense as the passengers. Difficult though it may be – and I do very much feel for TFL’s staff, who have to deal with myriad problems and customer personalities – everyone would benefit from announcements made in clear, non-robotic, warm voices. This is a customer service issue, and in restaurants, hotels, and shops, the customer expects a professional-yet-friendly tone from the staff, the likes of which is often missing from TFL’s announcements. Here for example is an announcement that I recorded at Leicester Square.

 

 

The voice you hear is loud, aggressive, and impatient. It creates a tense atmosphere. An announcement like that can only have a negative effect on a passenger’s mood.

Advancing the negative feelings further, passengers on the train can on occasion hear both a driver announcement and a platform announcement simultaneously. This talking over each other makes it harder to hear the information being relayed, again adding to the tension and frustration, and leaving the passenger feeling confused and bombarded – hardly a peaceful sensation.

The easiest solution here would be for the driver to wait until the platform announcement has finished, though it may also be necessary to limit the duration of platform announcements so as not to keep the driver – and thus the train – waiting for too long.

However, there’s perhaps a technological solution to be employed here, so that platform announcements are piped into standing trains, and vice versa. This would make all announcements mutually exclusive, and thus easier to hear for both those on the train and those on the platform. And whilst it’s true that not all messages are relevant to both the passengers on the train and those on the platform, irrelevant information is a price worth paying to ensure that the announcements can be heard at all, and to create a calmer, more peaceful atmosphere.

That is, assuming the passengers are paying attention in the first place. Many passengers, myself included, have headphones in as they travel, meaning that announcements are often missed. Particularly the beginning of them, during which those of us with headphones must first hear the faint noise of something being said over whatever it is we’re listening to, and then remove our headphones, by which time the critical information may have passed.

I wonder also about those who are hard of hearing. Though I’ve not done any research on this (as always, please chime in via the comments if you have a perspective to share), is it possible that, due to the volume of ambient noise, passengers with hearing aids turn them down, because otherwise the cacophony is too much to bear? And so, similar to those wearing headphones, these passengers would also struggle to catch an announcement when it is made, because they aren’t aware that it’s being made.

I wonder then if it would be possible to give a visual indicator just prior to an announcement, to prompt these people into listening. A non-invasive yet noticeable way of doing this would be to change the lighting. The driver would hit a button, the lighting would change, then after a moment’s pause – allowing people to remove headphones and such – the driver would make their announcement. (This would in some circumstances work even for those with their eyes shut, as a change of lighting can be noticed through the eye lids. Though this effect would be less noticeable when the train is in daylight.)

It’s easy to focus on numbers and forget about the less tangible issues at play. Those issues can often seem less important, partly because they’re harder to measure and therefore easier to dismiss. But I would argue that if TFL cares about customer satisfaction enough to improve the easily quantifiable things, it should also consider how it communicates with passengers. Currently, it all too often encourages bad moods and dissatisfaction to proliferate. Good communication can take the edge off a bad situation, and poor communication can make a bad situation even worse. Similarly, you can make the trains bigger, faster, and more reliable, but undermine all that good work by creating a tense, frustrating atmosphere.

Quantifiable measures are important, but they are not the whole experience. How are your customers feeling?

Amazon Cart

Recently, Amazon launched a Twitter feature that allows users to respond to a tweet containing an Amazon product link with the #AmazonCart hashtag, which will then add the item in question to the user’s shopping basket.

As Ars Technica wrote, the utility is minimal, and the business goals quite clear. And fair enough, because they doubtless thought this through and knew what they were aiming for.

However, I wonder if, with a slight tweak, this feature may actually have been a little more useful. What if instead of adding the item to your shopping cart, it was instead added to your Amazon wish list?

Adding the item to your shopping cart is an action that is likely to be followed up immediately, or in the near future. As you cannot complete the purchase via Twitter, you’re going to need to visit the site to buy the thing. And frankly, you’re probably already there, because you likely wouldn’t be buying the item sight unseen, or else you already knew about it and therefore didn’t need to see it on Twitter to be encouraged to buy it, as the Ars Technica article describes.

Why would you choose to tweet about something to add it to your shopping cart rather than click on the link, only then to have to go to the Amazon website anyway in order to complete the transaction?

On the other hand, let’s say you’ve just seen something awesome on Amazon. You want to mark it for the future, but you also want to share it with people on Twitter, because you think it may be of interest to them. You tweet a link to the world saying how great the product is. A few of your followers notice, check out the link, and respond with their own tweet, adding the hashtag #AmazonWish.

The difference between these two scenarios is that whereas the first is purely mechanical, the second is also aspirational. The purely mechanical approach fails because it’s simply a poor mechanic for adding an item to a shopping cart. Making that problem worse is that once the item is in your cart, it’s going to be there the next time you want to make a purchase on Amazon. If you don’t want to buy it at the time, you’ll have to remove it, at which point it’s both in your way, and then no longer associated with your account. This is why the wish list exists in the first place: it’s a place to store things off to one side, which you may purchase, or have purchased for you, at a later date.

And that’s why #AmazonWish would be more useful than #AmazonCart – because it’s non-committal, and it appeals to a genuine, already existing use case, in which people share products they’re interested in with their friends. (Possibly whilst mentioning that their birthday is coming up, hint hint, nudge nudge, and so forth.)

So there are usability issues around adding something to the shopping cart rather than adding it to a wish list, but there’s also a matter of perception. To the mind of the user, adding an item to the shopping cart is a merely a part of the checkout process, and one which a Twitter hashtag is doing nothing to expedite.

On the other hand, the sharing and discussing of interesting products on social media is a use case that already exists. Only now, there’s a convenient hastag if you’d like to favourite that item for future reference and possible purchase.

So might it be useful to have a hashtag that adds Amazon items to your wish list? #AmazonWish

Supermarkets: Part two

Those self-service checkouts at supermarkets: constantly frustrating, yet I can’t resist using them. But reflecting on the subject led me to thinking about other parts of the supermarket experience, and how they might be improved to the benefit of both the business and the consumer. A couple of ideas occurred to me.

1) Checkout Bonus Points

Have you ever played Fruit Ninja? If you have, you’ll likely remember the unbounded glee of slicing through a perfect cocktail of fruit, and reaping the combo bonus points. If you haven’t, you may get a sense of it from this video.

What if supermarkets offered you bonus points on your loyalty cards for certain items, or combinations of items?

2) Companion Chef

I remember approximately 15 years ago shopping in Safeway with my dad, using those handheld barcode scanners to scan each item as we put it into the shopping cart. I think this technology still exists in Waitrose, but otherwise seems to have died off. Would it be possible first to revive this idea, then take this to the next level by using NFC chips in products rather than barcodes? This would make the process less fiddly (one need only put the reader and the item close together, rather than hunt around for a barcode to scan). The shopper would then use a bespoke handheld device provided by the supermarket, or an app on their phone, to scan in the items.

With that being the case, let’s say you’ve just scanned an item. At that moment, your own personal chef pops up and says, ‘Hey, the ingredients you’ve just bought would make a great curry if you added some lemongrass, and guess what? It’s in the next aisle over!’ Using a phone app, or being able to login to a bespoke device, presents an opportunity for the software to learn your tastes, and then introduce you to other things you may like, similar to how music, book, and other media recommendation algorithms work today.

So you’ve read those ideas and you’re thinking, ‘Why on Earth would they do any of that?’ For me, there are two reasons.

The first reason is to encourage discovery. Supermarkets often reconfigure their shelves to force shoppers to spend more time exploring the store, and to re-examine the inventory. With this in mind, imagine that, on the bottom of the payment receipt, the shopper is informed that had they bought a different kind of pasta sauce, they would have achieved the Perfect Pasta combo, and have thusly received an extra 50 points on their loyalty card. The shopper wants those points, so they make sure to buy the alternative sauce next time. Turns out they prefer the taste of it, and what do you know, the markup is a little higher. How about that?

I suspect some of my readership may be laughing at this point. But games are a big business, they’re a good way to motivate people, and the nice thing about this approach is that anyone who doesn’t care can just ignore it, in the same way that many shoppers already ignore the ’10 extra points when you buy Product X’ coupons that already exist, whereas others painstakingly scan several of these coupons as they go through the checkout, eager to gain every extra loyalty point they can.

Anyone who does want to play the checkout bonus points game can be directed to a section of the supermarket website, where details of the various combos could be found. Perhaps some would only be hinted at, leaving the shopper to figure out the clues. The possibilities for introducing shoppers to new products here are endless! OK, maybe not endless. But they’re at least several.

In a similar fashion, extra points for certain items could encourage healthy eating. This may be something that’s detrimental to the business, but there may yet be a means of turning that into a positive. However, rewarding people for buying healthy items would be a positive step to tackling public health. It’s true that people could game the system, scanning their loyalty card only when they buy healthy things, but most people would not go to that level of trouble, even if they spotted the workaround.

Who says you can’t have fun at the supermarket?

Part one of my UX exploration of supermarkets can be found here.

Supermarkets

I may be romanticising this, but so far as I understand it, there was once a time in which the purchase of groceries necessitated human contact. More than this, it required bartering: careful negotiation, ever-vigilant of your opponent, warily wading through the pricing swamp, scanning for the many traps into which one may be lured, then suckered.

In many ways, very little has changed.

Where once this contractual dance was undertaken with a human, in today’s 21st century nightmare hell ride the shopkeeper has transferred their fiendish mind to a computer. No longer able to set the prices of things themselves (for they have no hands with which to label), they instead wish to induce the shopper to opt for suicide over plum tomatoes and baby spinach.

I think we’re probably all on the same page here: I am of course referring to the blasted self-checkouts at supermarkets various. Oh, don’t bother conjuring an image of a particular one; they’re all pretty much alike. Astoundingly, they appear to be all based on the same software, and yet each implementation has its own particular eccentricities and means by which to draw the ire of the hapless soul who dared to use them. (That usually being me.)

‘Are you using your own bag today?’

‘Why yes, Computer. I am.’

‘Wonderful! Then naturally you’ll want me to weigh it.’

‘But of course! I’d be delighted if you could …wait. What?’

‘Just pop your bag on the scales there.’

‘Sigh. Fine.’

‘Have you done it yet?’

‘Yes, damn it!’

‘I don’t feel anything.’

‘Well it is only a plastic bag. How sensitive are the scales?’

‘Sensitive enough to weigh vegetables, I’m certain.’

‘But not plastic bags?’

‘I fear not.’

‘So what now?’

‘I’m going to call for the nearest human helper person.’

‘But I specifically avoided the humans!’

‘And with good cause. Now your misery is double.’

This, dear reader, is but one of several pathways which lead to madness. To be clear, every one of the pathways leads to madness, it’s just that there are several of them. Computer may just as easily have decided that the bag was too heavy.

And what’s with the whole weighing thing, anyway? I understand weighing fruits and veg to determine their price, but everything scanned (or weighed) must then be placed into the ‘bagging area,’ wherein they are weighed (or weighed again), for reasons which elude me. I’ve already indicated that I wish to pay for the item. What business is it of yours, Computer, what happens to that item afterward? What are you trying to accomplish here? Surely if I planned not to pay for an item, I simply wouldn’t have scanned it in the first place. The process of weighing what I’ve already elected to pay for seems only to ensure that I then remember to take that item with me. Well, thanks and all, but frankly I’ll do without your help.

You know what’s a hazard to one’s health and the health of others? Doughnuts. That’s why they’re security tagged. Perhaps you’re wondering how that could be, or if failure to diffuse and remove such a tag could result in an accidental ingestion of a sugar-dusted, jam-filled security beacon which, when carried outside the confines of the supermarket, would alert the authorities just in time for them to observe a not entirely figurative taste explosion play out from your colon. I guess that wouldn’t look so good for Tesco.

Prevention of such a PR disaster must surely be why Computer insisted on summoning one of its indoctrinated human minions, whose soul it ingested, and whose lifeblood was used to make highly unstable baked goods, to come to my aid. The man told me to ignore Computer; I told him about the taste explosion. Sympathy was not forthcoming.

Marks & Spencer does a good line in malevolent, computerised shopping assistants. The genius of theirs is that Computer will dangle freedom before your eyes like a carrot (along with your carrots), before beating you to death with a stick (usually cinnamon). In my case, it informed me that an item couldn’t be scanned, so a human would be summoned. Bravely however, I persisted. It scanned! Yes! But alas, no. Computer was still quite sure something was wrong, and offered no way of me telling it otherwise. In the end, I had to wait for a thoroughly redundant human, and was visibly irked when they arrived.

M&S is an interesting example for other reasons. In the Hammersmith branch, there are eight self-service checkouts, and six regular checkouts. When fully staffed, the human checkouts beat the computers every single time. Their line always moves quicker, whilst I’m stuck waiting to be served by some evil motherboard with an attitude problem.

But is it really Computer’s problem, or my own? Why do I avoid the regular checkouts when I know them to be quicker?

I may be romanticising this, but so far as I understand it, there was once a time in which the purchase of groceries necessitated human contact. Those times are increasingly lost. Perhaps there’s no place in a hectic 21st century London, where people are afraid to talk to each other in general, for unnecessary human contact. It may take longer, it may make one’s blood boil, but for some reason, we persist to self-checkout. Seems I’d rather be tortured by slightly new yet eerily familiar bouts of ineptitude time after time, than to have to talk to a stranger.

Well damn it, I’m rebelling! Today I shunned the self-checkout line in favour of enlisting the time-tested bar code skills of an actual person. It was a momentarily awkward experience, and is but a small first step for this antisocial and technologically battle-scarred man. There is a long way to go before recovery, but every little helps.

Part two of my UX exploration of supermarkets can be found here.

Why the details matter: Polishing the user experience at VUE cinemas

Those of you who don’t know me may be shocked to learn that, at times, I can be quite the pedant. (Those who do know me are no doubt laughing at the understatement.) Some like-minded people find it amusing. (It’s generally meant in jest.) Others don’t care. But I imagine that several find it massively annoying, and think me quite the arse for doing it. But despite this, I can’t help but feel that, sometimes, it matters.

The following observation is, I think, not pedantry for its own sake.

Back in January, I visited one of the VUE cinemas in London, walked up to their bank of ticket machines, and was confronted with three options, shown in the photo below.

A separate option to Buy Tickets Today? Buy Tickets lets me choose the date, including today. Why the redundant option?
A separate option to Buy Tickets Today? Buy Tickets lets me choose the date, including today. Why the redundant option?

At the bottom you’ll see the option to ‘Pickup Tickets’, which seems fairly self explanatory. There’s also ‘Buy Tickets’, which I understand equally well. But what does ‘Buy Tickets Today’ mean? Of course I want to buy them today – I’m at the machine, aren’t I? I don’t want to have to come back to buy them tomorrow.

You’ve probably figured it out; what they really mean is ‘Buy Tickets [for] Today’. And frankly, I figured it out too, and so would almost everyone, if not actually everyone. So, is it really a problem? I’m going to argue that it is.

There are many reasons why the label is confusing. The first, as I’ve already mentioned, is the truncated language, which omits the word ‘for’ and thus changes the meaning. However, for reasons I’m about to explain, even if the label read ‘Buy Tickets for Today’, the button would nonetheless remain both redundant and unintuitive.

It’s redundant because selecting the ‘Buy Tickets’ option allows the user to buy tickets for any day, today included. So making them choose between the two buy ticket options is entirely without purpose.

It’s not intuitive because users don’t initially distinguish between buying tickets for today, and buying them for other days. When someone walks up to the ticket machine, all they’re thinking is that they would like to buy tickets.

This would be simple enough to verify; all VUE would need to do is to stop a few people as they approach the machine and ask them what they’re going to do. I would contend that regardless of whether they were buying tickets for that day, later that day, or for a future date, they would all simply say that they’re going to ‘buy tickets.’

(I cannot, of course, speak definitively for the ways in which everyone thinks. Everyone is different. However, my experience, and education, of human cognition teaches me that in the main, people are unlikely to distinguish between these two slightly different tasks. They are the same task: buying tickets. Choosing the exact date for which one wishes to purchase those tickets is a sub-task, a detail.)

The information architecture isn’t mapping to the goal in the user’s head, and the more the user has to translate between the goals in their head and the options afforded to them by the user interface (UI), the harder it is for them to use the software. (This is commonly referred to in human-computer interaction as the gulf of execution.)

But to go back a few paragraphs, it’s true that this isn’t really stopping anyone from buying tickets. In the end, people who are momentarily confused (such as myself) ultimately figure it out. I’m not seeing a horde of perplexed cinema patrons wandering around the car park tapping on thin air, the nonsensical UI having induced them into a stupor, all still trying buy tickets on the haunting touch screen display in their minds.

But I feel that, a momentary confusion though it is, that moment is everything. For some people it won’t matter, but for others, it creates a noticeable discord, which ultimately reflects badly on VUE, giving as it does the impression that we, the customers, aren’t being thought of, and ultimately making us feel a little inconvenienced, or aggrieved, or perhaps irked. However slight that feeling may be, it’s there, and it’s avoidable.

Now whilst we’re all here, a couple of other things cropped up during this same visit of mine. One of those was that a machine was out of order. (See the photo below.) Fair enough; things go wrong. But given that the VUE staff have noticed that the machine wasn’t working (indicated by the sign they’ve put on top), why have they decided to leave the machine on? I can only imagine that there’s no easy way to turn it off, because otherwise there’s no reason for customers to see the scary error message.

They know the machine isn't working, so why not turn it off instead of letting people see this?
They know the machine isn’t working, so why not turn it off instead of letting people see this?

Alternatively, websites often inject a bit of personality, humour, and above all empathy into their error pages, for when things go wrong. (Such as Twitter’s over capacity page.) How about the same thing here?

After I made it past the broken machine and the confusing buttons, the two tickets I purchased, along with my receipt, spewed out of the machine so fast that they all ended up on the floor. Mildly annoying, slightly inconvenient, easily solved.

This happened whilst I was busy returning my credit card to my wallet. My reflexes are, alas, not cat-like.
This happened whilst I was busy returning my credit card to my wallet. My reflexes are, alas, not cat-like.

It doesn’t have to be this way! (He cried from his soapbox.) It’s possible to create experiences without the inconsistencies, the confusion, the inconvenience, the lack of forethought into what people may actually go through when they use your services. And as the tagline of this blog says, those experiences may be digital, like the confusing UI, or they may not, like the tickets falling to the floor. But in either case, the approach to solving these problems is the same, and the details always matter. Maybe not to everyone, maybe not all the time. But to some people, and often enough.

Above all, the question is this: Why settle for an only passable user experience, when you could have one that’s delightful?

Visual perception and user interfaces: Here’s looking at YouTube

Human beings: we’re not as capable as we’d like to think. Oftentimes ineffective design is not really that bad as such, it’s just that it gives our species too much credit.

Here’s a famous example of our visual deficiencies. Have a look at the video below.

Did you spot the bear the first time you watched that? I certainly didn’t. We’re so focussed on counting the passes that we’re oblivious even to a sizeable, not to mention thoroughly unrealistic, dancing bear. If someone had explained the video to you in theory and then asked if you thought you would have seen the bear, you probably would have said that yes, you would have. And understandably so. I mean how could you not spot a dancing bear?

Here’s another well known example, this time via the wonder who is Derren Brown, and based on an experiment undertaken as part of a study entitled Failure to detect changes to people during a real-world interaction (Simons, D.J., Levin, D.T., 1998).

Again, if someone put such a scenario to you, you would probably say that you, or indeed anyone, would surely spot that the tourist asking for directions had changed. It’s just ridiculous to suggest anything else. And yet, there you have it: a lot of people didn’t notice.

These examples illustrate two related concepts. The dancing bear is an example of inattentional blindness, which is a failure to detect an unexpected change (the dancing bear) within your field of vision, due to your attention being consumed by other tasks (counting the number of passes).

The tourist experiment is an example of change blindness, which is a failure to detect a change (the tourist swapping for another person) due to a visual disruption (the door). Often, the disruption could be a momentary distraction, or a loss of attention (a blink, a loud noise, etc). The crucial point here is that change blindness requires a comparison to memory; something has changed whilst you were not looking at it. Therefore, when you go back to look at the changed thing (the tourist), you have to remember how that thing looked before the disruption, in order to realise that it has now changed. (The moonwalking bear, in contrast, is happening whilst in plain sight. It’s just that your attention is consumed.)

Let’s talk about your very own retina display. (By which I mean, your actual retina.) As many people know, inside the retina there are rods, and there are cones. The rods are used primarily in low light (such as at night, outdoors), and are not sensitive to colour. But in daylight, or artificial light, as far as your vision is concerned it’s mostly about the cones.

How many cones are there, exactly, and how are they distributed across the retina? Prepare yourselves, because here comes a diagram.

retina

source: Webvision

The numbers in red indicate the density of cone receptors. We can see that within the fovea, at the centre of the retina, the number of receptors drops off dramatically as we move toward the outer edges. Then, once we’re beyond the edges of the fovea, the drop is even steeper, relatively speaking.

All of this is to say that your vision is profoundly better at the centre of the retina (the fovea), where there are more receptors. There are six million cone cells in each eye, but they are far more densely packed within the fovea. The fovea is 1% of the retina, but your brain’s visual cortex devotes more or less 50% of itself solely to processing your foveal vision.

But it gets worse. Whereas the cells in your fovea map 1:1 to the neuron cells that begin the processing of visual data, outside of the fovea multiple receptors connect to a single neuron. So that’s more data (the receptors), being funnelled down the same bandwidth (a single neuron). There’s going to be some data loss. In computing terms, this is a lossy compression; we essentially have JPEGs for vision.

So then: We’re likely to miss seemingly obvious things when we’re concentrating on something else (inattentional blindness). Our short term memories aren’t the best, and so momentary distractions will mean that we can miss even the most obvious of changes (change blindness). To make matters worse, our vision is far more effective in the tiny 1% of our retina called the fovea than it is on the periphery.

No wonder we miss so much.

You’re probably wondering where the UI part comes in. Well, I went looking for a study to illustrate how these deficiencies manifest themselves in human-computer interaction, and I found this: The Case of the Missed Icon: Change Blindness on Mobile Devices (Davies, T, Beeharee, A, 2012).

The study is particularly interesting because it focusses on mobile. These issues have long since been illustrated on desktop computers, but not much work has been done in the mobile domain. To quote the authors themselves

“In a mobile context, research into change blindness is limited. One could argue that the display size of a current standard smartphone does not allow for attention towards changing items to be lost. This assumption is based on an expected greater coverage of the smaller device using foveal rather than peripheral vision.”

The study conducted two experiments with 17 male and 12 female participants between the ages of 18 and 24.

In the first experiment, the participants were presented with a menu of icons arranged in a grid, in the style of an iPhone. A number of visual disruptions where then invoked at a random interval. Those were a) no disruption; b) a flicker; c) a change in orientation; and d) a push notification appearing on screen. Simultaneous to the disruption, one of the icons in the menu was changed. Did the participants notice?

As you can see from the results, the disruptions caused fewer changes to be detected. This is an example of change blindness. The study also points out that the position of the changed icon in the menu made little difference to whether or not the change was detected.

The number of icons did affect the detection rate, however. More icons in the menu decreased the rate of detection. Cognitive load: it’s a real thing, people!

In the second experiment, participants played a driving game. Their primary task was to control the speed and direction of the car whilst avoiding collisions with oncoming traffic, and collecting stars to gain points. However, they also had a secondary task: to adhere to the changing speed limit.

A change in speed limit was displayed to the participant in two ways. Either an icon with the new speed limit would appear for 3 seconds, and then disappear (called ‘direct insertion’), or there would be an icon indicating the current speed limit visible at all times, which would update itself whenever there was a change in speed limit (called ‘gradual change’). Both types of icon could be presented at either the top or the bottom of the screen.

The average response time, for those notifications which were noticed at all, was 3.877 seconds. Not what I would call speedy, especially given that the participants are playing a game, in which reaction times are a factor. (If time were not a factor, the consequences of taking more time to notice a change would be less severe.)

This second graph demonstrates how many notifications were noticed. In all, 34.5% of notifications went completely unnoticed. Why? It’s the dancing bear; the participants were so focussed on the primary tasks that they missed an otherwise noticeable change right in front of their eyes.

The study goes on to suggest things a user interface designer can do to mitigate these problems, such as reducing simultaneous on-screen activity; amplifying changes (with movement, colour, or, if you really must, shaking); reducing the complexity (and thus the cognitive load) of the UI; and positioning changes close to or within the user’s foveal vision.

There are many design principles in the field of human-computer interaction (HCI) which are implicitly understood by good designers, and the ones identified above are no different. We already understand that there’s the potential for users to miss on-screen changes.

However, we tend to brush these concerns aside when we’re pushed to do so by other factors. We know that these principles are sound, and we often adhere to them, but we don’t know why they’re sound, so we allow ourselves to break them. I think that the reason we’re able to do this is partly because these concepts always seem a bit hypothetical without knowledge of the evidence.

The study cited here is but one small piece of a mountain of such evidence, particularly when you include desktop computing as well as mobile. So the next time you find yourself thinking, ‘But surely the user will see that‘, remember: we’re not as capable as we’d like to think.

If you’re after some good reading material on this and other HCI / UX matters, I thoroughly recommend Jeff Johnson’s Designing With the Mind in Mind. The clarity of his explanations, particularly of the retina, was extremely helpful in writing this entry.

Special thanks to Dr. Ashweeni Beeharee for granting me permission to reproduce the graphs from The Case of the Missed Icon, and for his general support.