Morris minors and machine slots

Natasha Schüll’s study of machine gambling, Addiction by Design, won a lot of prizes when it came out in 2013.  But it’s had nil splashback on digital games – which is curious.  Most games rely on the operation of chance – or at least unpredictability – in order to unfold.  So an understanding of how and why ‘pure’ chance entrances people would seem to be a truly handy thing to have in your back pocket.  “Know your ingredients” as Joy of Cooking would say.    (18 million copies and counting!)

There is certainly a strong parallel between the revenue structure of free to play games and the revenue structure of machine games: for both, the overwhelming bulk of their revenue comes from a small number of players.  Revenue concentration in f2p is even higher than it is in slots machines – as physical machine slots carry an entry price.

But no.  There’s no quicker way to annoy your common or garden variety digital games person (designer or player) than to suggest that gambling and ‘real’ games share important design patterns.  If you are feeling brave and want to try it, I advise you to stand well back first.   (Unless, of course,  they’re into social casino – which is definitely a story for another day.)

There are other problems with using Schüll’s ethnography as a cookbook for supplying top recipes for ‘chance entrancement.   For one thing, her study  documents the detrimental effect that machine games have on some players.

For Schüll, this arises from a fundamental structural asymmetry between gamblers and the industry that supplies them:

“The gambler not only can’t win, but isn’t playing to win, while the gambling industry is playing to win all along.”

“The relationship that exists between players and the industry is not so much a clash between two systems of value as it is an asymmetric interdependency between a system of value extraction that plays by the economic rules of the market, and a fleeting zone of non-value in which those rules are, for the player, suspended.” p. 72

Not only is the problem that the relationship is asymmetric, but, according to some researchers, the design of the games is such that

“impaired control and subsequent problem development are an understandable and ‘natural’ consequences of regular high intensity [machine] play” p 16

As Schüll explains it, the industry,  by contrast, treats addiction risk as inhering in characteristics of the player, rather than the design characteristics of the gambles they partake in.  Schüll preserves a more or less discreet silence on this issue – except, perhaps, for the hint in her book’s title.

Many of the people who are the subjects of Schüll’s case studies have life histories of abuse, anxiety and rejection, and other very serious challenges.    The close focus provided by her enthographic lens leaves us none the wiser about whether her addicted subjects are typical.  Perhaps they experience a high baseline level of suffering which is relieved by the experience of gambling,  and this makes the experience more risky for them?  Less data about more people would help untangle this, of course – but that’s not the game she plays.

It is obvious that some players of machine slots become addicted.  Whether this has to do more with an individual’s predisposition to addiction, or more to do with an experience design which is intended, from the ground up, to create addiction in any player, is not entirely clear.  Common sense would suggest that both factors are important.

Whatever the answer, there’s the issue of whether it’s good to design things that some – but not all – people may have serious problems with.    You could apply a similar ethical question to the creation of cakes, or cocktails.  Or digital games – whether f2p or not.

But even putting moral qualms aside to prove,  in a low oven with a wet cloth over them, there is very different but equally difficult problem with using Schull as a cookbook: an utter lack specificity over mechanism.

She goes into a goodly amount of detail, and also has interesting insights about:

  • the phenomenology of the machine gambler’s experience (which I summarise in another post)
  • design tricks which are used to induce that experience

What you don’t get, is any deep insight into how the designers’ tricks – some of which are very tricky indeed – work to create the powerful and nihilistic phenomenal experiences she documents.   It’s like showing separate pictures of two gears, without showing how they engage, except via a hand wave at ‘math farms’ and ‘random reinforcement schedules’.

So, like any interesting piece of work, her book creates more questions.

One thing that’s clear to me that machine slots are like Morris Minors.   The Morris Minor displays the high forehead and concentrated stereotypical ‘cute’ features of a baby.  This leads to a heightened response to take care of it.   No matter that they stopped production decades ago, and fall apart the moment you turn your back or slam the door.   They are still cute.   You still want to put money into them.

They act as a kind of super stimulus, abstracted and distilled from what you would expect to find in nature, producing a response which seems inappropriate, but may simply be the result of heuristics gone wrong, in contexts which supply powerful but misleading cues.

The plot line of machine slot inputs and outcomes probably performs a similar similitude function to real world features – although what the feature detection hierarchy is that is responsible for their power has yet to be fully explained.

Morris Minor 1000 Source: /0/06/Morris_Minor_1000_1958.jpg



The moral mysteries of chance and combat as game design elements

One fine coffee,  a games designer told me gambling  was utterly immoral.  And shooter games were not.  In the face of his complete certainty there was nowhere to go.  So nowhere was went.   But the exchange left me with a lingering sense of amusement about how differently people see the world, and a curiosity about the fundamental issues.

Clearly, if you look at preferences for skill vs chance as continuum on a single dimension, my coffee guy scored high on skill and vanishingly low on chance.  He valued a game environment that enabled and challenged the exercise of his team combat skills.  If there were bloodstains on the wrappers these skills were packaged up in, they were not meant to be taken literally, as a feeder path to actual aggression, or as an ersatz substitute for it.  Instead, the game was a thing in itself, an enjoyable recreational activity, like a sport.

But physical sports are different in important ways.  In physical sport, the action is real.   So an important function of rules in physical sport is to reduce the prospect of real bodily harm to the player.  Rules restrict and abstract behavioural possibilities away from pure violence.   Also, another function of rules in physical sport rules is to provide unnatural constraints which provide extra challenge, and serve to channel the player’s energy and creativity.   Unless you are goalie, you are only able to use your feet to move the ball, in football.   Who thought that up?  Were they trying to make it difficult?   (Yes, probably.)

Shooters do have much in common with sport, but by representing actual battle, even if ritually sanitised and exaggerated, they venture closer to human’s propensity to violence and conflict.   This is not an accidental quality of such games, but a fundamental part of their attraction.   A hugely important part of the abstraction from actual violence is that the action is not physically real, but is calculated and represented to the player.  All kinds of things are ‘possible’ – virtually – without actual physical harm. Another facet of the abstraction is that allowable actions and  consequences are highly constrained and rule-based.  In this it is similar to physical sports.

Consuming – and producing – representations of an activity is clearly related to   performing the activity itself.   But the nature of that relationship is unclear. Just as kittens chase string, and cats chase rodents,  both sports and shooters can be seen as a preparation for more deadly pursuits.   Alternatively – or additionally – they can be seen as a sublimation or an evolution of the instincts which result in physical aggression.   As far as I can see, which function they actually serve, and how this function originated,  is impossible to determine.   Though that shouldn’t stop us from thinking about it.

I’m not a shooter person.   Or at least not that I’ve yet disovered.  But it isn’t morally wrong to me, that people get satisfaction out of engaging in ritualised, rule-based conflict – either in sports or in video games.   But neither is it unflinchingly innocent.   What it is, is interesting.   And ambiguous.

Neither do I see the excitement of chance in a game as necessarily morally wrong.    Unlike coffee guy.     But neither it is inflinchingly innocent.

Many respected theorist-practitioners view chance as an important design ingredient, rather than as something nasty in the woodshed.  For example, Greg Costikyan’s 2012 GDC talk  tests the flavour of a very broad variety of types of uncertainty in games, treating the uncertainty of chance outcomes as part of a much broader palette of possibilities.  Jesse Schell – with Lens 34, in his Game Design: A Book of Lenses talks not only about chance, but also about the need to adjust the balance of skill and luck to the needs of particular audiences.

As it turns out, coffee guy isn’t alone.  Both Costikyan and Schell observe that people who adore games of skill and combat tend to dislike the use chance, if it plays anything like a leading role.  Mind you, as a bit of spice added to a loot drop it seems to be tolerated pretty well.

Although skill and luck can be construed as opposing ends of a design/experience continuum, they are not mutually exclusive.  Something which has neither skill nor chance components in it is very unlikely to be a game at all.   However it is possible for a game to have high (or at least medium high) elements of both skill and chance.  A classic example is poker.   It is more informative to say that such a game has strong elements of both qualities, than to say it  is neutral.

I think that the right carburettor mix of skill and luck in the delivery of game rewards can power a desire for repeat engagement, by combining delight, hope, and frustration.  I’m not sure exactly what the formula is, but I’m thinking about it.   Ivan Encinas from gave an interesting presentation on describing game levels using these dimensions at the data science track I curated last month. I’ll sling up a link when there’s one available.

But about the experience of pure chance, as entertainment, I’m less certain.  To be blunt, I don’t get it.

Thinking about it in the abstract, there’s potentially a case to be made that games of pure chance provide the hedonic equivalent of empty calories – a kind of basic hotwired limbic thrill, without veneer of cognition and volition (aside from the basic act of electing to engage in the game, and re-engage).

It’s not my thing.  But I know that people like all kinds of things that I don’t particularly care for.  Like pink.   Or tripe.  If it doesn’t frighten the horses or interfere with other people’s liberty I am not usually inspired to call it out on moral grounds.

But, if you look,  you will find huge vehemence against games of chance in gambling addition researchers – I recently attended a talk by very famous one said she who made all her research associates take a solemn oath never to work in the gaming industry.   If they didn’t, she wouldn’t take them on.  This struck me as melodramatic – but her sincerity and passion was unquestionable.    Like coffee guy, she believes pure chance is pure evil.

When I look at games of pure chance,  from an experiential point of view I don’t see much to enjoy,  let alone create crippling addiction.  That’s why I stuck my nose into Addiction by Design, Machine Gambling in Las Vegas, by Natasha Dow Schull (to be reviewed in my next post).   I was curious to see if she’d found and tamed the ghost in the machine.

Data science and advanced analytics in games – call for submissions!

I’m  coordinating the BRAND NEW data science in games track at the much-loved  game AI conference (now http:/,  in Vienna 20-22 July 2015.    Yipee!

We’ve already got some amazing talks and speakers already lined up from Mind Candy and Scientific Revenue.   Come and join us!   Do you have something you’d like to talk about?  Drop me a line today (or sooner)  with a few lines about what you’re interested in, and I’ll be happy to explain more about the themes we are developing for the track, and other talks that are in prep.    The programme is almost full but there are a few spaces left, so pick up your carrier pigeon and let it fly!

Continue reading