Are We Spellbound By Games – Asking Questions About Habit Forming Design

Oscar Clark   •   August 13, 2014

The Greatest Magic Trick

Once upon a time we all thought how charming it was when games had that elusive ‘addictive’ quality which delighted players. But now the term addiction has become an accusation labeled at apparently unscrupulous FreeToPlay designers who are leveraging the psychology of Operant Conditioning to allegedly extort money from players.

Games that delight us often have the magical effect of become habit forming but how as designers can we judge the balance between a game as a social good and the manipulation of an audience.

I’m not a psychologist!! I’m a marketer who has been closely involved with running games services for the last over 16 years. That being said I’m first and foremost a games fan. That means I put the enjoyment of games first but I am also deeply interested in both the application of psychology for both playing and paying. I’ve read a lot round the subject but have also have had some fantastic help from Cyberpsychologist Berni Good (but please note that any mistakes here are my own!). I gave a webinar on this topic recently, and, if you want to hear more, you’ll find a link to a recording of the session at the bottom of the page.

Design is a seductive art which is about enticing players in to your world and making them feel safe and special. This is about creating anticipation and desire for more. Key to seduction is our ability to correctly set expectations. Just as using a fake photo on an internet dating site (I’m too old/married to know personally) can damage the first date; giving misinformation to players is a fast way to get deleted. Similarly a lack of attention to the needs of new players (such as a sucky tutorial) will be an instant put off… just like abandoning your potential new partner to go talk to someone else on your first date.

But if we want our relationships to last we can’t take out players (or partners) for granted. We have to sustain that interest over the long term. We want, no NEED, our games to become habitual don’t we?

Games have always been compelling 

Games allow us to demonstrate our own competence, be faced with genuine (albeit constrained) challenges and to escape (ideally) within a social context. These qualities are alluring and done right create something which has an intrinsic reward. That matters when we think about habit forming. BJ Fogg talks about habit forming behaviour as needing motivation, ability and a trigger. Games create compelling experiences, offer challenges which we see as achievable and contain calls to action as well as instant feedback. These seem to have the qualities needed to drive habit-forming behaviour.

The argument has been that a lot of  F2P games have attempted to use habit-forming techniques to create unscrupulous designs, often relying on extrinsic rewards to reinforce player behaviour to encourage spending rather than to deliver better  inherent value for players.

However, is that actually true? And does this mean that there is an inherent flaw specific to F2P design? Game design relies on creating rule systems and patterns which, to a greater or lesser extent, railroad users to perform specific patterns of activity. The best magic trick is when we make a level or player decision feel like it was their own choice; even where it was the only real choice available. Games do this by using audio/visual/gameplay or narrative clues to trigger reaction; usually by providing instant positive feedback when the player behaves the way we want them to.

Arguably that’s a form of manipulation

Assuming you accept that manipulation is an intrinsic part of game design then we have to ask whether manipulation is inherently wrong or whether it just becomes wrong when this involves money. Is it even essentially wrong to encourage/convince people to spend money in a game? Where do we draw the ethical line?

Terms like manipulation don’t help us as their use is too broad and we end up in meaningless discussions about semantics.

Games have often been the target of moral outrage, but I fear that, as an industry, we may be contributing to our own panic, either because of an aversion to the commercialisation of games, or through a lack of focus on delivering entertainment over revenue. It seems to me entirely reasonable that we should be able to make games people are willing to pay for as long as the business model offers value and is based on informed consent.

What is Real Addiction or Conditioning?

The two principle accusations levelled at games seem to me to focus on addiction and the use of Skinner-Box-Style operant conditioning?

Addiction a very serious issue that is harder to define than you might think. However, it shouldn’t be defined as  just a habit gone bad.  According to Harvard Mental Health Letter there needs to be a genuine compulsion which overwhelms the individual’s otherwise rational behaviour. This can be driven by a physiological need where the body has a chemical dependency which causes us to crave the addictive substance. Alternatively, it can be behavioural where an unhealthy pattern of specific and inappropriate activity becomes overwhelming. Both lead to harm to the individual concerned.

Interestingly, behavioural addiction is not included in the Diagnostic and Statistical Manual of Mental Disorders because “…there is insufficient peer-reviewed evidence to establish the diagnostic criteria and course descriptions needed to identify these behaviours as mental disorders.” It has however been described as an area for further study which demonstrates the need for further research and for us to remain cautious about jumping to conclusions either way.

In The Power of Habit, Charles Duhigg, describes a series of MRI Scans conducted as part of a legal case against a casino. The scans showed that to pathological gamblers near misses looked like wins. Their brains reacted almost the same way. On the other hand, when non-problem gamblers experienced a near miss, they received a dose of apprehension that triggered a different response, “quit before it gets worse”.

Games, if we are to believe Rafe Koster’s Theory of Fun, use pattern matching to generate stimulus. Over time and given repeated success the pleasure reward diminishes as we become increasingly familiar with the patterns. Eventually we get bored with the game and move on. However as Neurologist Judy Willis described in Psychology Today “The motivation to persevere and pursue greater challenge at the next level is the brain seeking another surge of dopamine — the fuel of intrinsic reinforcement.” That does sound like something we encourage in game design.

Just A Skinner Box

In a series of experiments on Rats and pigeons, BF Skinner found using his eponymous box that an animals behaviour could be affected by adjusting their expected rewards; such as releasing food by pressing a button. By creating an escalating or random variation in the frequency of the release of that food you can even can cause them to press that button compulsively. The topic is too complex to cover in detail here and there are unanswered questions too, however it seems that a schedule of reinforcement (repeated actions with a predictable payout) plus unpredictable timing to that reward can be highly compelling.

Is that what’s happening in games? And is it unique to F2P?

The good news is that playing games a lot may not in itself be problematic for all gamers, whereas with addiction it is always detrimental for the addict (Gentile, 2009;Kuss & Griffiths, 2012): ‘‘Healthy excessive enthusiasms add to life, whereas addictions take away from it’’ (Griffiths & Meredith, 2009,p. 247).

A number of experiments have shown that humans show the same patterns of response that other animals show when exposed to the basic schedules of reinforcement; so Operant Conditioning does seem to work. However, in a game (unlike a Skinner Box), we don’t control all the factors and aren’t the only source of stimulus. In practice it’s rare that any game will have the capacity to exert much direct influence over players; the vast majority of people are just too savvy.

Even if we could really deliver the kind of influence that might affect players’ behaviour, conditioning also seems to require effort if it is to be sustained. There comes a point where that effort becomes greater than the reward. When we vary the rate of stimulus (e.g. the interval or ratio of activity needed to trigger a reward), the investment required to achieve the payoff will eventually exceed its value (eventually leading to the cessation of that behaviour). We learn from experience as an ongoing adaptive process. When conditions change, we learn new behaviours and eliminate old ones. Habits can, of course, be very hard to break – but they can also be very hard to form as (going back to BJ Fogg) they need a sustained Motivation, Action and Trigger.

It seems to me that conditioning in games is plausible; but largely impractical, not least because people are just too savvy.

Is it an Question of Ethics?

Of course being difficult doesn’t mean impossible.  But is there an opportunity? is this even an ethical question at all?

One of the biggest holes in the question of manipulation is that most (mindful) people will quickly notice any obvious attempts at manipulation. Despite this there are of course vulnerable people including children or adults with a susceptibility who might fall into the tricks of less ethical designers. There are systems in place to protect minors, legally and from a platform perspective – however this is not just a question of enforcement.  Our brand is on the line if we sully it with unethical designs which destroy trust and credibility. The anger people feel about being manipulated isn’t just about the individual concerned – they tell everyone they can. That kills any business.

Trust is what helps keep people engaged – not manipulation! It’s not just unethical to seek to form habitual behaviour in games but it is also largely counterproductive.

To be ethical our intent has to be appropriate, reasonable, and clearly communicated. Players need to have informed consent. But, for me, this isn’t just a question of ethics.

Commercially we are better off making better games. Designing game play that factors in engagement is the very best way to ensure a healthy environment that encourages players to come back to your game. Giving feedback at the right time will helps since intrinsic positive reinforcement is massively important in terms of desire to play. That feeling of mastering something and of an escape from our everyday world are more important than mechanical operant conditioning because they leave the player with a feeling that they “…cannot wait to play again”. This taps into the very essence of what it means for people to play games, rather than the work of responding to a schedule of reinforcement.

We have a lot to learn from the underlying principles behind habit forming behaviour especially from people like BJ Fogg and BF Skinner; but the value is in trying to understand players so that we can make them better games; not just to make more money.

Special thanks to Berni Good for her input and advice (any mistakes here are of course my own!)

Oscar Clark   •   August 13, 2014