by Dean Burnett
Other visual illusions are more subtle. The classic “is it two faces looking at each other or actually a candlestick?” image is possibly the most familiar. This image presents two possible interpretations, both images are “correct” but are mutually exclusive. The brain really doesn’t handle ambiguity well, so it effectively imposes order on what it’s receiving by picking one possible interpretation. But it can change its mind, too, as there are two solutions.
All this barely scratches the surface. It’s not really possible to convey the true complexity and sophistication of the visual-processing system in a few pages, but I felt it worth the attempt because vision is so complex a neurological process that underpins so much of our lives, and most people think nothing of it until it starts going awry. Consider this section just the tip of the iceberg of the brain’s visual system; there’s a vast amount more in the depths below it. And you can perceive such depths only because the visual system is as complex as it is.
Why your ears are burning
(Strengths and weaknesses of human attention, and why you can’t help eavesdropping)
Our senses provide copious information but the brain, despite its best efforts, cannot deal with all of it. And why should it? How much is actually relevant? The brain is an incredibly demanding organ in terms of resources, and using it to focus intently on a patch of drying paint would just squander them. The brain has to pick and choose what gets noticed. As such, the brain is able to direct perception and conscious processing to things of potential interest. This is attention, and how we use it plays a big role in what we observe of the world around us. Or, often more importantly, what we fail to observe.
For the study of attention, there are two important questions. One is, what’s the brain’s capacity for attention? How much can it realistically take in before it gets overwhelmed? The other is, what is it that determines where the attention is directed? If the brain is constantly being bombarded with sensory information, what is it about certain stimuli or input that prioritizes it over other things?
Let’s start with capacity. Most people have noticed attention has a limited capacity. You’ve probably experienced a group of people all trying to talk to you at once, “clamoring for attention.” This is frustrating, usually resulting in loss of patience and shouts of, “One at a time!”
Early experiments, such as those by Colin Cherry in 1953,10 suggested attention capacity was alarmingly limited, demonstrated by a technique called “dichotic listening.” This is where subjects wear headphones and receive a different audio stream (typically, a sequence of words) in each ear. They were told they had to repeat the words received in one ear, but then were asked what they could recall from the other ear. Most could identify whether the voice was male or female, but that’s it, not even what language was spoken. So attention has such a limited capacity, it can’t be stretched beyond a single audio stream.
These and similar findings resulted in “bottleneck” models of attention, which argued that all the sensory information that is presented to the brain is filtered through the narrow space offered by attention. Think of a telescope: it provides a very detailed image of a small part of the landscape or sky. But, beyond that, there’s nothing.
Later experiments changed things. Von Wright and his colleagues in 1975 conditioned subjects to expect a shock when they heard certain words. Then they did the dichotic-listening task. The stream in the other ear, not the focus of attention, featured the shock-provoking words. Subjects still showed a measurable fear reaction when the words were heard, revealing that the brain was clearly paying attention to the “other” stream. But it doesn’t reach the level of conscious processing, so we aren’t aware of it. The bottleneck models break down in the face of data like this, showing people can still recognize and process things “outside” of the supposed boundaries of attention.
This can be demonstrated in less clinical surroundings. The title of this section refers to when people say their “ears are burning.” The phrase usually used to mean someone has overheard others talking about them. It occurs often, particularly at social occasions such as wedding receptions, farewell parties, sporting events, where a lot of people are gathered in various groups, all talking at once. At some point, you’ll be having a perfectly enjoyable conversation about your mutual interests (football, baking, celery, whatever), when someone within earshot says your name. They aren’t part of your current group; maybe you didn’t even know they were there. But they said your name, perhaps followed by the words, “is a tremendous waste of skin,” and suddenly you’re paying attention to their conversation, rather than the one you are having, wondering why you ever asked that person to be your best man.
If attention was as limited as the bottleneck models suggest, then this should be impossible. But, clearly, it isn’t. This occurrence is known as “the cocktail-party effect,” because professional psychologists are a refined bunch.
The limitations of the bottleneck model led to formation of the capacity model, typically attributed to work by Daniel Kahneman in 1973,11 but expounded on by many since. Whereas bottleneck models argued that there is one “stream” of attention that hops about like a spotlight depending on where it’s needed, the capacity model argues that attention is more like a finite resource that can be divided between multiple streams (focuses of attention) so long as the resources are not exhausted.
Both proposed models explain why multitasking is so difficult; with bottleneck models, you have one single stream of attention that keeps leaping between different tasks, making it very difficult to keep track. The capacity model would allow you to pay attention to more than one thing at a time, but only so far as you have the resources to process them effectively; as soon as you go beyond your capacity, you lose the ability to keep track of what’s going on. And the resources are limited enough to make it look like a “single” stream is all we’ve got in many scenarios.
But why this limited capacity? One explanation is that attention is strongly associated with working memory, what we use to store the information we’re consciously processing. Attention provides the information to be processed, so if working memory is already “full,” adding more information is going to be difficult, if not impossible. And we know working (short-term) memory has a limited capacity.
This is often sufficient for your typical human, but context is crucial. Many studies focus on how attention is used while driving, where a lack of attention can have serious consequences. In many states, driving while physically using a phone is not allowed; you have to use a hands-free set-up and keep both hands on the wheel. But a study from the University of Utah in 2013 revealed that, in terms of how it affects performance, using a hands-free set-up is just as bad as using the phone with your hands, because both require a similar amount of attention.12
The fact that you have two hands on the wheel as opposed to one may provide some advantage, but the study measured overall speed of responses, scanning of environment, noticing important cues; all these and more are reduced to a similar worrying extent whether using hands-free or not, because they require similar levels of attention. You may well be keeping your eyes on the road, but that’s irrelevant if you’re ignoring what your eyes are showing you.
Even more worrying, the data suggests it’s not just the phone: changing the radio or carrying on a conversation with a passenger can also be equally distracting. With increased technology found in cars and on phones (it’s technically not illegal at present to check your emails while driving) the options for distraction are bound to increase.
With all this, you may wonder how anyone can drive for more than ten minutes straight without ending up in a disastrous wreck. It’s because we’re talking about conscious attention, which is where the capacity is limited. As we’ve discussed, do something often enough and the brain adapts to it, allowing procedural memory, described in Chapter 2. People say they can do something “without thinking,” and that’s quite accurate here. Driving can be an
anxious, overwhelming experience for beginners, but eventually it becomes so familiar the unconscious systems take over, so conscious attention can be applied elsewhere. However, driving is not something that can be done entirely without thinking; taking account of all other road users and hazards needs conscious awareness, as these are different each time.
Neurologically, attention is supported by many regions, one of which is that repeat offender the prefrontal cortex, which makes sense as that’s where working memory is processed. Also implicated is the anterior cingulate gyrus, a large and complex region deep in the temporal lobe that also extends into the parietal lobe, where a lot of sensory information is processed and linked to higher functions such as consciousness.
But the attention controlling systems are quite diffuse, and this has consequences. In Chapter 1, we saw how more advanced conscious parts of the brain and the more primitive “reptile” elements often end up getting in each other’s way. The attention-controlling systems are similar; better organized, but a familiar combination or conflict of conscious and subconscious processing.
For example, attention is directed by exogenous and endogenous cues. Or, in plain English, it has both bottom-up and top-down control systems. Or, even more simply, our attention responds to stuff that happens either outside our head, or inside it. Both of these are demonstrated by the cocktail-party effect, where we direct our attention to specific sounds, also known as “selective listening.” The sound of your name suddenly causes your attention to shift to it. You didn’t know it was coming; you weren’t consciously aware of it until it had happened. But, once aware of it, you direct your attention to the source, excluding anything else. An external sound diverted your attention, demonstrating a bottom-up attention process, and your conscious desire to hear more keeps your attention there, demonstrating an internal top-down attention process originating in the conscious brain.#
However, most attention research focuses on the visual system. We can and do physically point our eyes at the subject of attention, and the brain relies mostly on visual data. It’s an obvious target for research, and this research has produced a lot of information about how attention works.
The frontal eye fields, in the frontal lobe, receive information from the retinas and create a “map” of the visual field based on this, supported and reinforced by more spatial mapping and information via the parietal lobe. If something of interest occurs in the visual field, this system can very quickly point the eyes in that direction, to see what it is. This is called overt or “goal” orientation, as your brain has a goal that is “I want to look at that!” Say you see a sign that reads special offer: free bacon, then you direct your attention to it straight away, to see what the deal is, to complete the goal of getting bacon. The conscious brain drives the attention, so it’s a top-down system. Alongside all this there’s another system at work, called covert orientation, which is more of a “bottom-up” one. This system means something is detected that is of biological significance (for instance, the sound of a tiger growling nearby, or a crack from the tree branch you’re standing on) and attention is automatically directed towards it, before the conscious areas of the brain even know what’s going on, hence it’s a bottom-up system. This system uses the same visual input as the other one as well as sound cues, but is supported by a different set of neural processes in different regions.
According to current evidence, the most widely supported model is one where, on detection of a something potentially important, the posterior parietal cortex (already mentioned regarding vision processing) disengages the conscious attention system from whatever it’s currently doing, like a parent switching the television off when their child is supposed to take out the garbage. The superior colliculus in the midbrain then moves the attention system to the desired area, like a parent moving their child to the kitchen where the garbage is. The pulvinar nucleus, part of the thalamus, then reactivates the attention system, like a parent putting garbage bags in their child’s hand and pushing the child towards the door to put the damn things out!
This system can overrule the conscious, goal-orientated top-down system, which makes sense as it’s something of a survival instinct. The unfamiliar shape in your vision could turn out to be an oncoming attacker, or that boring office colleague who insists on talking about his athlete’s foot.
These visual details don’t have to appear in the fovea, the important middle bit of the retina, to attract our attention. Visually paying attention to something typically involves moving the eyes, but it doesn’t have to. You’ll have heard of “peripheral vision,” where you see something you’re not looking at directly. It won’t be greatly detailed, but if you’re at your desk working at your computer and see an unexpected movement in the corner of your vision that seems the right size and location to be a large spider, you maybe don’t want to look at it, in case that’s exactly what it is. While you carry on typing, you’re very alert to any movement in that particular spot, just waiting to see it again (while hoping not to). This shows that the focus of attention isn’t tied directly to where the eyes are pointing. As with the auditory cortex the brain can specify which part of the visual field to focus on, and the eyes don’t have to move to allow it. It may sound like the bottom-up processes are the most dominant, but there’s more to it. Stimulus orientation overrides the attention system when it detects a significant stimulus, but it’s often the conscious brain that determines what’s “significant” by deciding the context. A loud explosion in the sky would certainly be something that would count as significant, but, if you’re going for a walk on the evening of July Fourth, an absence of explosions in the sky would be more significant, as the brain is expecting fireworks.
Michael Posner, one of the dominant figures in the field of attention research, devised tests that involve getting subjects to spot a target on screen that is preceded by cues which may or may not predict the target location. If there are as few as two cues to look at, people tend to struggle. Attention can be divided between two different modalities (doing a visual test and a listening test at the same time) but if it’s anything more complex than a basic yes/no detection test, people typically fall apart trying it. Some people can do two simultaneous tasks if one is something they’re very adept at, such as an expert typist doing a math problem while typing. Or, to use an earlier example, an experienced driver holding a detailed conversation while operating a vehicle.
Attention can be very powerful. One well-known study concerned volunteers at Uppsala University in Sweden,14 where subjects reacted with sweaty palms to images of snakes and spiders that were shown on screen for less than 1/300th of a second. It usually takes about half a second for the brain to process a visual stimulus sufficiently for us to consciously recognize it, so subjects were experiencing responses to pictures of spiders and snakes in less than a tenth of the time it actually takes to “see” them. We’ve already established that the unconscious attention system responds to biologically relevant cues, and that the brain is primed to spot anything that might be dangerous and has seemingly evolved a tendency to fear natural threats like our eight-legged or no-legged friends. This experiment is a great demonstration of how attention spots something and rapidly alerts the parts of the brain that mediate responses before the conscious mind has even finished saying, “Huh? What?”
In other contexts, attention can miss important and very unsubtle things. As with the car example, too much occupying our attention means we miss very important things, such as pedestrians (or, more importantly, fail to miss them). A stark example of this was provided by Dan Simons and Daniel Levin in 1998.15 In their study, an experimenter approached random pedestrians with a map and asked them directions. While the pedestrians were looking at the map, a person carrying a door walked between them and the experimenter. In the brief moment when the door presented an obstruction, the experimenter changed places with someone who didn’t look or sound anything like the original person. At least 50 percent of the tim
e, the map-consulting person didn’t notice any change, even though they were talking to a different person from the one they’d been speaking to seconds earlier. This invokes a process known as “change blindness,” where our brains are seemingly unable to track an important change in our visual scene if it’s interrupted even briefly.
This study is known as the “door study,” because the door is the most interesting element here, apparently. Scientists are a weird bunch.
The limits of human attention can and do have serious scientific and technological consequences too. For example, heads-up displays, where the instrument display in machines such as airplanes and space vehicles is projected onto the screen or canopy rather than read-outs in the cockpit area, seemed like a great idea for pilots. It saves them having to look down to see their instruments, thus taking their eyes off what’s going on outside. Safer all round, right?
No, not really. It turned out when a heads-up display is even slightly too cluttered with information, the pilot’s attention is maxed out.16 They can see right through the display, but they’re not looking through it. Pilots have been known to land their plane on top of another plane as a result of this (in simulations, thankfully). NASA itself has spent a lot of time working out the best ways to make heads-up displays workable, at the expense of hundreds of millions of dollars.
These are just some of the ways the human attention system can be seriously limited. You might like to argue otherwise, but if you do you clearly haven’t been paying attention. Luckily, we’ve now established you can’t really be blamed for that.
_____________
* Some scientists have called this finding into question, arguing that this staggering number of smell sensations is more a quirk of questionable math used in the research than the result of our mighty nostrils.1