The Notification Game - How social media gets you hooked

A deep-dive into how behaviourist psychology on the web poses some fundamental questions about our behaviour around apps: Are we really in control?


There was once a woman, known to the annals of psychology as Karen. Since the age of 10 Karen had suffered the debilitating affliction of drug resistant seizures. You can think of these as electrical cascades that ripple through parts of the brain that should not ripple. By the age of 27 she had one avenue open to her, what we know as a callosotomy. A procedure in which the corpus callosum, the primary bundle of fibres connecting both hemispheres of the brain is surgically severed. Without this connection, the right and left hemisphere of the brain are isolated from one other. A curtain is drawn between these two ordinarily rather chatty parts of normal, neurotypical consciousness.

One thing you should know before we continue is that certain faculties of the brain exist in only one or the other hemisphere. Language, for example, is almost exclusively found in the left hemisphere. Additionally, the physical movement of each side of the body exists in the opposite side of the brain, left to right and right to left. So when I raise my left hand to drink my morning cup of builder’s tea it is the right hemisphere doing the heavy lifting.

The callosotomy was carried out without a hitch, everything went as planned and one day shortly after the procedure Karen sat in a quiet office with her doctor discussing ordinary medical matters. Everything was going normally. Her left hemisphere was quite busy employing her primary language centres discussing her rehabilitation when without Karen’s control her right hemisphere started slowly undoing the buttons of her doctor’s shirt with her left hand. Obviously, whatever piece of consciousness that silently occupies her right hemisphere was not satisfied with simply talking to her ostensibly handsome doctor.

At that moment there were at least two different motivations being expressed simultaneously by Karen. Or is it there was one action being expressed by Karen and another that she did not control? The quick among you will notice a question being slowly wheeled out from stage left.

Are you really in control?

Many of us would answer with a resounding yes. But are you really? Are you so sure that your motivations, desires, thoughts, or actions are completely at your behest? How would you know? Research and Karen’s naughty side appear to suggest that there are at least some parts of us that are not.

I’ve used this provocative story to highlight a strange all-but-fact about the human condition. One that is often overlooked in the cogs of our day-to-day lives. There are processes within each of us that we have no real control over but which exert control over us. And it gets worse. These mechanisms can be for lack of a better term hacked. And these mechanisms of influence, of behavioural manipulation, are no longer aphoristic or exclusively the domain of ancient Greek playwrights. We’ve studied them, mapped them, and are deploying them with startling effectiveness.

But before we go any further with this narrative I’d like to lay out some key aspects of the psychological framework we’ll be looking at today, though hundreds are actively being used and researched. This framework is a cornerstone of Psychology: Behaviourist Psychology, or more specifically Neo-Behaviourist Psychology.

Does Behaviourism Ring a Bell?

Behaviourist Psychology relies on a simple premise. We are drawn to pleasant things and we withdraw from things that are unpleasant. Conversely, we are also drawn to things that remove unpleasant sensations and we withdraw from things that remove pleasant sensations. The act of being drawn to or withdrawing from a thing is a response. We call the sensation, pleasant or not, the stimulus.

Let me muddy the waters a bit. The idea of pleasant stimulus does not necessarily mean happy feelings nor does unpleasant mean unhappy feelings. As I’m sure you’ve noticed, humans are an odd bunch. We can find social controversy exciting, find violence stimulating, and can even find sitting in an isolated, dark room while muttering and tapping at a laptop rewarding (don’t you judge me). It’s probably more accurate to think of positive behavioural stimulus as that which entices or interests and negative stimulus as that which repels or bores. After all, we perennially flock to horror franchises like Saw which are on the face of it terrible prospects and we all have that one moody, complex, and unavailable ex that we found oh so sexy. We are a complex bunch with complex tastes.

Now, one gem of genius of Behaviourist theory is that it takes this simple foundation and applies it over time to the realm of learning. Simply, a stimulus-response hook applied over time can lead to what we might understand as learning how to act in the presence of an object or stimulus. A second gem is that we can twin an arbitrary trigger with a stimulus then we can remove the original stimulus and still produce the same response. The textbook example of Pavlov’s dog illustrates the effect perfectly.

This seminal study conducted by Pavlov and his pooches goes like this. Every day Mr Pavlov would feed his dogs and ring a bell while he did so. They would chow down on the enjoyable stimulus while being conditioned to respond to the arbitrary trigger of the bell. Eventually, he could simply ring the bell and make his poor dogs salivate without any food being present whatsoever. Poor dogs. While this example seems trivial it has been shown time and time again to apply in truly myriad conditions to humans and dogs alike.

Now, I want you all to move your attention from this text and think back. Remember the last times you idly visited Facebook, Twitter, or whatever. Imagine your actions as you poke the app icon on your phone or click into the tab on your laptop. Got it? 
Hands up who, as their first or second action, clicked on their news notifications? Why did you click there? Was it conscious, did you actively choose to peruse this action? I recently gave a talk where I asked this very question. Nearly everyone held their hands up and looked sheepishly at the others in the room. They and you could have clicked anywhere.

Still sure you’re in control?

The Behaviourist Notification

Every social media app has one. It’s on your iPhone’s home screen, it’s on your screen when you log in to Facebook, and in all honesty many of us wake up to it. What we’re talking about here, dearest reader, is that little red circle enclosing a number, the social media notification. And this innocuous little symbol is has become a ubiquitous fixture of the modern online landscape.

So let's run a little experiment, just between me and you, I promise I won’t tell.

Take your phone out and unlock it. Now look at your phone’s dashboard and take a quick look at the apps you have installed. See the icons, the terrible wallpaper you should be ashamed of, and consider for a moment just how expensive this real estate is. Appreciate just how many hundreds of thousands of hours went into its design and presentation. Now hone in on the apps you have installed, the staggering amount of work that has gone into each of them. Their logo is their brand, it’s the first thing you see, it’s their identity. And if you think I’m talking in too grandiose terms about few pixels on the screen, then I would invite you to listen in on some management level discussions about their branding - this is serious business.

Now, do you also see something covering many of these absurdly expensive collections of pixels? Yep, you should have guessed it, those small red circles with a number in them. The notification is such a powerful behavioural manipulation technique that billion dollar companies are happy to cover their logo with it. So let’s take a close-up of the notification through a behaviourist lens and see what we find.

The anatomy of the notification

The stripped-down behavioural elements of a notification are deceptively simple: the stimulus is the experience and experienced gratification of whatever service the app provides, the response is your interaction with the app (your attention), and the arbitrary stimulus to be twinned with the gratification is the red symbol with a number inside of it. Seems simple right? Regrettably not.

Initially, the “at” notification symbol means little to you beyond being just another shade in the miasma of raw sensory input, it is at this point a truly arbitrary symbol. Your interaction with this app, your response, is at this point can be considered completely voluntary. This is your first exposure to the stimulus-response hook.

Over time your relationship to the app and its stimulants develops and deepens. The initial stimulus that took you to the app is replaced with your self-cultured bubble of interests, be they the twitter account of Donald Trump or a curated account of cat videos. This is where the learning aspect of behaviourism proper takes place. Each time you now interact with the app and visit the notification area you are learning to expect a certain experience and the contingent emotions involved with the experience. And each time you perform this sequence of actions that little red notification circle is being twinned with those rewarding or enticing emotions. Rewarding emotions that you yourself cultivated.

After a critical point, the hitherto arbitrary notification will have been twinned with the rewarding stimulus enough that even in the absence of the stimulus you will interact with the arbitrary notification as if this random icon actually produced those emotions in you. At the risk of alienating my readers, you are now reacting to Pavlov’s bell in lieu of actual food.

Through the behaviourist lens you were repeatedly exposed to both effective stimulus and arbitrary trigger enough times to be conditioned to respond the same way to either.

How Neo-Behaviourism Fits In

The neo-behaviourist framework extends these tried-and-true behaviourist principles. For us the most interesting element of this framework is the discovery of mediating factors between the stimulus and response.

In the behaviourist model the stimulus generates the response, there’s an automatic process that does the processing for you. However in the neo-behaviourist model the stimulus generates a tentative cognitive map in the mind of the individual then this map instructs the response. You can think of the intermediary cognitive map as a deliberate sober driver guiding her impulsive and basically unconscious friends, we’ve all been there. These intermediaries include things like availability, how much money you have, needs, difficulty, and the organism’s existing behavioural patterns.

The Neo-Behaviourist Notification

The underlying behavioural mechanism of the notification outlined above still exist. But there are just a few more psychological ingredients to throw into this bubbling pot. Previously, the functional elements of the behavioural hook of the notification symbol were pretty simple: the reward, the response, and the arbitrary trigger. And while the neo-behaviourist intermediary principals do complicate the matter, they also provide an actionable window through which the tech world can streamline or more accurately guide a behavioural hook to their own ends.

With this theoretical development previously extraneous information surrounding the hook become as important as the hook itself. How easily can the app be accessed, how emotionally rewarding and salient is the content on the app, and to what degree does the app need to fight for you in the ever more crowded economy of attention that characterises our modern digital landscape? All of these elements might hinder or aid the development of conditioning. If you happen to live in the exact centre of the Australian outback the chances are your connection might hinder the availability of the app but your isolation might mean that you’re more inclined to boredom and thus have some time to flick through Facebook. Intermediary variables are myriad, even in the Australian outback.

The neo-behaviourist notification invites UX designers and developers to streamline the user’s access to triggers and stimulus. Inspire them to ease access to the features of their app they want the public’s feet to fall and confuse those parts that they want to keep quiet.

A perfect example of this is the “delete my account” pages on social media sites. Almost every one requires byzantine navigation through menus you’ve never seen and the recital of the alphabet backwards while throat singing the British national anthem. They are designed to dissuade you from leaving them. Designed to throw up as many intermediary variable blockers as possible.

A Question of Influence

Well now, if I’ve done my job this whistle-stop-tour will have given you a lot to chew on. But there is one glaring question that remains — should we purposefully use psychological techniques that influence people in ways that they might not want to be influenced? Is it ethical to employ what has come to be known as “dark patterns”?

If you read this last paragraph aloud and if you listen closely you should be able to hear philosophy grads across the globe clear their throat and start explaining that this is an intimidatingly complex question. That we need to think long and hard about questions of agency, freedom, the limits of free-capitalism, and whether we should tease dogs with food in weird psychology experiments. And you know what? The philosophy grads are probably right.

During a recent talk I gave on this subject there was a clear philosophical divide during the question and answer period. One camp believed that these so-called dark patterns are pernicious logical extensions of the drive for profit and should be avoided. The other camp asked a simple but powerful question “Where are the companies that didn’t employ these techniques?”.

Another simple truth is that the cat is already very much out of the bag. Most if not all of the currently popular social media sites and apps do design and code with some degree of behavioural manipulation in mind. Push notifications bring you into the app with the implicit or explicit promise of social reward. Even the most intimate parts of ourselves, love and sex, are used as an emotional weight to get you into apps more often.

Additionally, while the idea of being influenced may feel intrinsically uncomfortable there can be places where this kind of manipulation serves as a force for social good. Take for example the existence of the nudge unit. These are professional psychologists who work for the government and who, amongst other things, design and execute campaigns to reduce our population’s smoking, binge drinking, and reduce the physical abuse of public sector staff. While the die-hard libertarians among you might still bristle at the idea of influence, many among us feel that these are forces of good in the world. Without them, we would be worse off.

Final Thoughts

The ethical-philosophical landscape of human influence is complex. But it is nevertheless one in which we find ourselves deeply embedded. Like much of philosophy there are no clear answers and no hard lines in the sand, that is unless you happen to go by the name Immanuel Kant. But perhaps if we’re more aware of how these techniques are used and what kinds of questions others have been asking. You might be nudged into making a more informed choice as a consumer or as a provider. Or maybe there will just be a sudden spike in the number of people buying dog food and bells from supermarkets.