Human Hardware: You’re Looking, But Are You Seeing?
So far in the Human Hardware series, we’ve dealt with working memory (a just released study confirms the limitations of this. See my note at the end of this column), differences between sexes, and Dunbar’s number (Parts One and Two). Today, I want to explore how we can look right at something and not see […]
So far in the Human Hardware series, we’ve dealt with working memory (a just released study confirms the limitations of this. See my note at the end of this column), differences between sexes, and Dunbar’s number (Parts One and Two). Today, I want to explore how we can look right at something and not see it. This is particularly true in online engagement, whether it be a home page, a portal, a landing page, or a search results page. You’ve probably heard it referred to as banner blindness. It’s caused by two hardwired limitations, one in our eye and one in our brains.
An eye tracking tale
Let me start by telling you what happened in one particular eye tracking test we conducted with one of our research partners. But first, I’ll quickly explain how eye tracking works. Our eye tracking equipment is from Tobii, a Swedish company. It’s a special monitor that tracks the actual physical location of the eyes through image sensors and reflects near-infrared light off the back of the eye to see where the eye is focused. This is called foveal vision, and I’ll explain more about that later. Through the use of tracking and analytic software, you can see what people looked at on the screen.
But, as we saw in the one particular test I referred to, knowing where people look doesn’t always tell you what they see, and that’s the topic of this column. Looking and seeing are two completely different things. This became apparent when we started analyzing the results of the study. In the study, we gave subjects a number of websites to look at. On the websites, we placed a test ad. Each subject was given a test task to accomplish on the site. After they finished the task, we asked them to complete a quick survey and in this, we asked if they remembered seeing an ad. We also asked them what the ad was about and which brand it was for. Almost no one remembered the brand or what the ad was about. Few remembered seeing an ad at all. By all standard recall measures, this ad was invisible.
But it became more interesting when we looked at the eye tracking data. Almost everyone looked at the ad. Some spent a significant amount of time looking at it in fact, a few seconds at least. In fact, of all the time they spent looking at the page, about 10 to 15% of their time was spent looking at the ad. But they couldn’t remember it. They looked at the ad, but they didn’t see the ad. And this is a key difference.
There are none so blind as those that will not see
Let me give you a test. Now, to make this work, you have to follow my instructions exactly. No skipping ahead! I’m going to send you to an online video that shows two groups of people, one with white shirts and one with black shirts, passing basketballs. If you’re a man, pick the white team. If you’re a woman, pick the black team. I want you to count how many times the basketball is passed by your team and remember it, then come back to this page. I’ve purposely split you up into genders because one gender has a significant advantage in this task, but I won’t tell you which one until you’ve completed the task.
Ready? Then click this link (will open a new window, so you’ll need to kill your pop-up blocker) and when the video loads (it’s about 6 Mb and it loads in a java player), you’ll see a bar with a play button. Click the button and start counting. See you in a moment.
Hmmm…are you sure that’s the video I was looking at?
Okay, got the number? Good. I don’t really care what the number is. And the line about men or women being better at this doesn’t have a shred of truth to it. I just wanted to focus you more intently on the task. My real question is, did you see the man in the gorilla suit? What man? What gorilla suit? Go back and take another look at the video. I’ll wait.
How could you have missed it? Don’t feel bad. Almost everyone does (70% plus). It’s amazing how you can not see a man walking right through the scene in a gorilla suit, and it’s an example of what’s called inattentive blindness. You were looking, but you weren’t seeing. Let’s look (well, see, actually) at what happened here.
First of all, you didn’t see it because of two limitations, one in your eye, and one in your brain. Let’s deal with the brain first.
We’re constantly scanning our environment to see if there are any immediate threats. This functionality is buried in the deepest core of our brains, an instinct that goes right down to the reptilian brain stem. Even people with almost complete brain damage still scan their environment, and their eyes will follow you if you enter the room. They’re looking, but they’re not really seeing. Because for you to be seen and recognized, you need to switch from unconscious scanning mode to conscious attention mode. And that requires shifting to the executive mind, in the prefrontal cortex.
The gatekeeper that decides which stimuli get through to the executive mind is called selective perception. It’s at the juncture of our unconscious and conscious minds. As the brain idles in its unconscious gear, it’s constantly picking up signals and determining whether or not they demand attention. Most of the things are filtered out before they can clutter up the limited space in our executive mind and working memory. There’s just not enough storage space or processing power to worry about all these things. But things can get our attention in two ways.
The high road and the low road in our brain
First of all, we process some stimuli, especially visual ones, through two paths. One goes straight from the visual cortex to the amygdala, which is the fire alarm of the brain. It alerts us to physical danger. Again, this is in the older part of the brain. The other takes a much longer route to the prefrontal cortex. This is why you’ll jump back from a garden hose in a dark shed before you realize it’s not a snake (everybody seems to use this example when explaining the concept, so I’ll follow suit). If you were depending on your rational brain to save your hide, you’d have been hit by a bus (or a rampaging mastodon) long ago. The rational brain, as powerful as it is, is just too slow. But, after you jump, you realize it was a garden hose and feel a little foolish. This is your rational brain catching up.
In order for something to get from unconscious scanning to conscious attention and be stored in our memory (unlike the gorilla and the banner ad in our study), it has to go through four phases. Stimulation, registration, organization, and finally, interpretation. First, we receive the stimulus through our eyes and the visual cortex. Then, the brain has to register the meaning of the stimulus. It has to recognize what it is, so it can decide whether to ignore it or not. If the stimulus passes this phase, we then compare it (organization) to our filing cabinet of memories, to see what it means to us. The appropriate memories are retrieved from long term memory and loaded into our working memory for conscious consideration. It’s at this phase where we switch from unconscious monitoring to conscious interaction. Finally, if the stimulus means something to us, we begin to interpret its meaning. The executive mind is in full gear and attention is acutely focused on the stimulus.
As something moves from unconscious scanning to conscious consideration, there’s a corresponding transition in brain power. Our brains are highly efficient, so processing power is shuttled to where it is most needed. This means, as we begin paying conscious attention to something, the unconscious processes slow down dramatically. We’ve all experienced this. We become highly engrossed in something and oblivious to the outside world. We can even walk into a post as we’re staring at a particularly attractive member of the opposite sex. As our modern brain kicks into gear (although it’s questionable if this is the part of the brain that’s working, given my example), the reptile brain checks out.
Here’s what happened with our gorilla example. I focused your attention on the task, counting the passes. I even upped the ante by leading you to believe that your gender might have the upper hand in this task. You started the task with your prefrontal lobes in high gear, and your unconscious monitoring systems were already shut down. Even something as absurd as someone in a gorilla suit didn’t make it through your task induced blindness. When I first heard about this, before seeing the video, I thought the gorilla would be fairly discrete in the background. When I saw how prominent it was (the cat was out of the bag, remember, so I wasn’t focused on the task), I couldn’t believe anyone could miss the gorilla. But obviously, I was wrong. Daniel Simons is the man responsible for the study, by the way. He calls the phenomenon inattentional blindness.
When we track engagement with web pages, exactly the same thing is happening. People are so focused on their task that they become literally blind to everything else, even if they physically look at it. Ads are particularly susceptible to this, because they often fail to pass the recognition – organization threshold. We unconsciously see it’s an ad, it’s not something we’re interested in, and so we block it from our consideration.
But there’s another factor that leads to this blindness, and it has to do with the physical structure of our eye. We have two types of receptors in our eyes, rods and cones. Rods need less light, so they give us our night vision, but they don’t perceive color (which is why everything becomes gray at night). They are also more sensitive to motion. Cones need much more light, but they can perceive color and pick up much greater detail than rods. The center of our retinal wall, the fovea, directly opposite the lens, is packed with cones but has hardly any rods. This is the part of the eye we use to focus our attention. This is what is tracked in eye tracking. Surrounding the fovea is the parafoveal part of the retina, which has far fewer cones but more rods. This is our peripheral vision. The structure of our eyes means that we often catch movement out of the corner of our eye better than when we’re looking directly at it. This makes sense, as there was a much better chance of danger coming at us from where we weren’t looking, rather than the relatively small spot we were looking at.
But although our eyes have two distinct “projection” areas, the brain actually does a little sleight of hand to blend the two together. If we watch a bird in flight, or a horse running, we see it as one fluid motion, rather than a jerky set of “snapshots”. This is because of the rods and our peripheral vision. The parafoveal receptors constantly send signals that allow small adjusting movements in the muscles that control our eyes. They almost act as early warning radar, picking up what’s on the fringes and readying the foveal part of the eye. The brain stitches the images together to give us a smooth continuous stream of stimuli. It’s an amazing feat.
But, if we start intently focusing on something, the brain slows down the channels with the parafovea. We narrow our field of vision, to allow us to concentrate on what we’re looking at. The more intently we focus, the more our peripheral vision goes into idle. That’s why you didn’t see the gorilla (unless you’re one of the 30% who did).
Now, obviously, this had dramatic implications for web design and advertising. The level of intensity of task engagement can make your target oblivious to your ad, a navigation bar, a call to action, or even a gorilla moving across a page. Seems hard to believe, but it’s true. When we design, we’re focusing on the elements we’re adding to the page. We’re looking at it with our fovea in full gear. How could anyone miss it? But, our prospects aren’t coming to the page with the same task. They’re looking, but they’re not seeing.
Note: In a response to one of these columns, Bill Slawski took exception to my interpretation of George Miller’s work on channel capacity. Miller’s study used a series of slightly different tones to explore our capacity limits to consider alternatives, but the objective was to determine the capacity of our working memory in judging alternatives, not to see how we determine between tones. Miller picked this particular example only because it was a unidimensional stimulus – only one aspect differed. Bill says, rightly, that it has nothing to do with comprehending written text. But the point is, to make decisions between written text alternatives, you have to load them (or the concepts they represent) into working memory, the same as Miller’s pitch tones.
When you understand how working memory functions, you realize there is a distinct difference between working memory, or our executive brain, and long term memory. The purpose of the column was to explore the limitations of the former, a vitally important concept to understand for advertisers and interface designers. Bill failed to draw this line of distinction in his rebuttal. As luck would have it, a just released study confirms the limits of working memory might be even more restrictive than first thought, allowing us to only compare 3 or 4 alternatives at any one time. This is exactly the number of search results we see considered at a time, so this aligns very well with our research findings at Enquiro.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.