Email is addictive
Like lots of people who sit in front of a computer all day, I am addicted to email. This worries me for two reasons. The first is the sheer strength of my compulsion. I must hit the ‘get mail’ button at least a hundred times a day. Sometimes, if I don’t have any new mail, I hit it again immediately, just to check. I interrupt my work to check my mail even when I know that I’m not going to find anything interesting and that I should just concentrate on what I am suppossed to be doing. When I come back to my office it’s the first thing I do. If I’m prevented from checking my mail for more than a few hours I get a little jumpy and remain that way until I have.
This is all rather sad, but the second reason I am worried by my email addiction is that I work in a psychology department and we’re supposed to understand how these things work. Now email isn’t a drug – it doesn’t deliver a chemical into your bloodstream. Yet it is clearly addictive. I’m a normal rational person (which is to say I’m just normally maladjusted) and I know that I don’t need to check my email as often as it do – certainly not immediately after checking it the first time for Goodness’ sake! – but still I am compelled. What’s going on, and can psychological science help me out?
Read more below the fold
Why is email addictive?
So, I think I’ve got an idea of what’s going on with email, and if I’m right it should provide some clues as to how I can stop myself being so addicted. The key is what psychologists call ‘operant conditioning’. This means the mechanisms by which behaviour is shaped by its consequences; how what we do depends on the rewards and punishments of what we did last time. This topic is the heart of behaviourism, that school of thought which dominated psychology for most of the last century. Many lab animals, and many person-hours, were recruited to help understand exactly how rewards and punishments could be arranged to influence behaviour. One suprising finding is that if you want to train an animal to do something, consistently rewarding that behaviour isn’t the best way. The most effective training regime is one where you give the animal a reward only sometimes, and then only at random intervals. Animals trained like this, with what’s called a ‘variable interval reinforcement schedule’, work harder for their rewards, and take longer to give up once all rewards for the behaviour is removed. There’s a logic to this. Although we might know that we’ve stopped rewarding the animal, it has got used to performing the behaviour and not getting the reward. Because ‘next time’ might always be the occasion that produces the reward, there’s never definite evidence that rewards have stopped altogether.
Email is addictive because it is a variable-interval reinforcement schedule
We’re animals – we have animal brains. All animal brains have the circuitry in place for producing operant conditioning. It’s a fundamental psychological process, and just the sort that can create behaviours what operate automatically, or in spite of our consciously telling ourselves we should do otherwise. Like me checking my checking my email. Checking email is a behaviour that has variable interval reinforcement. Sometimes, but not everytime, the behaviour produces a reward. Everyone loves to get an email from a friend, or some good news, or even an amusing web link. Sometimes checking your email will get you one of these rewards. And because you can never tell which time you check will produce the reward, checking all the time is reinforced, even if most of the time checking your email turns out to have been pointless. You still check because you never know when the reward will come.
I have just proved to myself how automatic my email checking behaviour has become. I am writing this in a hotel room which doesn’t have internet access. When sorting through my email (you don’t need a connection to delete email you’ve replied to, or are never going to reply to) I still hit the ‘check mail’ button at the rest points of the read-consider-delete cycle I am performing. My reflective self knows that there is no internet connection, so there is no way in hell I’m going to have new email – but that knowledge doesn’t filter down to the part of me hitting the ‘check email’ button. The habit, engrained in my mind by operant conditioning, is isolated from conscious knowledge, and in part from deliberate control; it can start without me thinking about it or even me wanting it to.
How can we design in solutions?
If operant condition is at the root of the problem, what’s the solution? Over a hundred years of experimental psychology has provided a rigorous characterisation of behavioural conditioning, and of the process by which reinforced behaviours disappear – known as extinction. By looking at each stage of the process by which a behaviour becomes conditioned, we can throw up ideas for addressing the problem of ‘unconditioning’ them.
Weakening the action-reward link
If a behaviour isn’t rewarded then it will gradually disappear. The problem is that we don’t want to remove the reward (email), so we need, instead, to weaken the strength of the link between the action and the reward. A simple delay would do this – imagine a five minute delay between hitting the check email button and getting new email. A delay is doubly-effective because the longer the delay the more likely you are to have email and so the more consistent the reward will be – and hence the less strong its reinforcing effect (I can see this is action when I go away for a week and check my email when I come back. I might have hundreds of emails, but there’s often nothing that seems very interesting. The combined effect just isn’t as rewarding as the anticipation of getting one hundred single emails). In theory, the opposite end of this ‘consistency strategy’ would be to check email constantly. If you check your email every second then the consistency of the reward increases – you consistently get nothing! The action-habit might extinguish because the action is now rewarded so infrequently relative to the number of times it is performed. How to do this practically? Have your inbox constantly visable, so that there is no jump from wondering if you have new mail/seeing a ‘new mail’ alert and knowing that it isn’t that important (although a down side is that if the new email is intrinsically distracting – such as a good joke, hot piece of gossip or invite out – it will now definitely interrupt when it arrives).
Removing the action all together, so that you cannot demand an email check, doesn’t solve the problem – we just move the association from the action of hitting ‘check mail’ to the action of opening your email client to see if there is any mail there. One strategy is to wait 30 seconds from the appearance of the alert before checking to see what kind of mail has arrived – if you can internalise the delay.
Weaking the stimulus-action association
Automatic behaviours such as email are contingent on environmental triggers – they need the right circumstance to become active, since they aren’t invoked by a smart, deliberate consicous process. So one way to decrease email checking behaviour might be to decrease the association between stimuli and the action. Moving or removing the ‘check mail’ button will stop it hijacking your action-stream when it is in your field of view. Similarly, so will stopping email alerts all together, but it leaves you with the problem of wondering what is going on while you’re attention is on another app (are you missing out on some important information?). Changing the consistency of the environmental trigger could work. If the ‘new mail’ appeared slowly, so it didn’t grab your attention, was in a different place each time, or was a different shape/colour/icon then your low-level monitoring systems wouldn’t be able to form an association between its appearence and automatically triggering the ‘check mail’ interrupt habit. When you wanted to check your email you could scan for the icon – it could still be perfectly obvious, but the lack of consistent appearence would prevent it getting wired into a direct channel in your brain.
I don’t think the solution is social – getting people to never email you urgent information wouldn’t work. There would always be invites for coffee that someone didn’t feel like knocking on your door or phoning you to make, but which they’d still like you to know about (and you’d still like to know, so you’ll be checking your email for them). Plus there’ll always be important information which isn’t urgent, but which we’re still going to want to know as soon as it is available (look at news – it rarely has any immediate consequence for our daily lives, but checking it is how many of us start the day).
Shifting the cost-benefit ratio
A classical approach to changing habits is to shift the cost-benefit pay-off. If checking or reading email is made harder to do, mayhbe we can increase the cost of the action and hence make the overall reward of getting email less. More difficult interfaces maybe? Several screens to click through to get the email? Passwords and ‘are you sure’ dialogue boxes? Punishment is the ultimate ingredient that you can add to alter the pay-off matrix. Except for those of us with immense self-disicipline (“if i check email in the next hour, no coffee at lunchtime”) self-imposed punishment isn’t an obvious strategy (and are those with self-discipline likely to have a problem with getting distracted anyway?).
[Although, on reflection, I don't think this question is as rhetorical as it sounds here. The problem is not with your ability to deliberately schedule and implement actions (including actions such as self-punishments), the problem is with automatically evoked actions (i.e. checking email) and then getting distracted by the consequences. These are two types of actions and it is entirely conceivable that they might be differently ammeanable to the deliberate control of our reflective selves]
Rewarding an alternative, incompatible behaviour
A final strategy, and one that is used in animal training to remove problem behaviours, is to reinforce an alternative, incompatible action. If you have a problem with your pet eagle landing on your head the most efficient way to stop it is to reward landing on a mat at your feet, rather than struggle with extinguishing head-landing. What this would mean in the context of email checking I will leave as an exercise to the reader.
Over to you
So, what I’m really interested in is in seeing what people think. If some of what I’ve said makes sense, what other ways are there to use it to make email less addictive? Or maybe I’ve got the solution completely wrong. Or the problem? I’d like to hear either way
Some things I read while writing this follow (thanks to Marc Baizman for the Katie Hafner article in the NYT)
‘Husband training’, New York Times article about using behaviourist principles in your marriage
February 10, 2005 New York Times, You There, at the Computer: Pay Attention by Katie Hafner (designing interfaces that work with our attention, rather than distract)