Sunday, October 25, 2015

Playing Games

Leveling up.  New items.  High scores.  New abilities.  Upgrades.  These game mechanics have a seductive allure known all to well to those of us who play video games.  

A featured article at Game Informer explores this fascinating subject.

A few paragraphs in, a minor but significant error is made, taken from no less an expert than a Ph.D. psychologist who describes the mechanics as classical conditioning - immediately bringing to mind for many Pavlov and his famous salivating dogs. 

However, game mechanics are an example of operant conditioning, not classical conditioning. Classical conditioning (also known as respondent conditioning) describes unlearned behavior (the behavior itself) elicited by the environment, i.e. salivating, eye blinking, sneezing, etc. and isn't affected by what comes after. Operant conditioning, however, is learned behavior (classes of responses) that is evoked by the environment and either punished or reinforced depending on what follows the behavior (i.e. reaching for cookies because they were delicious, or playing "just one more level" to get a special item.

This is an intermittent schedule, in particular a variable ratio schedule, the schedule that evokes the highest rate of responding across all organisms. Other schedules: fixed ratio, fixed interval, or variable interval all have lower levels of responding, whether you are a pigeon, dog or human.

What's fascinating is that operant behavior - which encompasses most everything we do, including our thoughts, does not need to be conscious. We can learn to be aware of our behavior, but we generally can't really be aware of all aspects of the schedule of reinforcement of which the behavior is a function. For example, we can only be vaguely aware of how pleasurable or painful an activity is, but a fuller account needs a high level of data collection (time, behaviors emitted, topography of stimulus, history of reinforcement, etc.). We can, however, make predictions, alter our environments and design rules for ourselves to follow. 

A tad unfortunate that the psychologist on hand muddled the subject a bit.  It may have been sloppy wording, but my suspicion is that his mistake is emblematic of a larger problem in which behaviorism is relegated to an important, yet minor "school of thought" in psychology.  Skinner is taught somewhere after Pavlov, alongside Watson, and then passed over for Bandura, Piaget, and the ever-luminous Freud.  This is an unfortunate, because in casting aside Skinner's work, including him among other "interesting" but unscientific and subjective, hypothetical (in behaviorism we call them "mentalistic") models, our power to not only define human behavior but to truly understand what creates and shapes it is sacrificed.  Far from being an outdated "perspective", behaviorism is a hard science, and underlies everything we do.

So why isn't radical behaviorism, nearly seventy years after its development, embraced more fully outside a narrow clinical application of its principles in work with special needs populations, if it is such a fundamental description of why 7 billion humans do what they do?  

Applied Behavior Analysis (the contemporary name for behaviorism), rooted in the philosophy of radical behaviorism, is complex.  It requires a relatively lengthy period of study to begin to fully appreciate.  It is also the case that  one's own preconceptions about the nature of the "mind", human nature, and free will are challenged by the evidence.  To accept radical behaviorism is to accept a materialist, deterministic, parsimonious, empirical account of human behavior, i.e. why we do what we do.  This sets up a situation in which those who have not spent considerable time studying behaviorism likely do not understand it very well, and will fail to appreciate its power.  Furthermore, they will not have come into contact with its salient features, namely hard evidence that overturns conventional wisdom held not only by the general public, but of many other areas of psychology.

Skinner described human behavior as a process of cultural selection, operating similarly to biological evolution evolution.  In the Darwinist account, certain phenotypes are created, but then selected for by the environment, which allows certain traits to pass on generationally and others to die off.  With behavior, certain behaviors are reinforced by the environment, causing them to occur again or more frequently, while others are punished, causing them to stop or occur less frequently.  Just as the environment is a constant pressure on biological evolution (food, water, availability of mates, shelter, etc.), so too is cultural selection in the form of determining which behaviors will survive and which will fade away.

The theory of evolution is now over 150 years old, and although widely embraced by scientific and reasonable people, it's detractors still object to it on the grounds that it is incompatible with an intelligent designer, or God.  Religious compatibilists such as the Catholic church, argue instead that God can still be present, in that he merely "got the whole thing going".  I'm not sure if that makes much sense when you really think about it, especially in the context of their larger Christian doctrinal exegesis, but good on them. 

The parallels with radical behaviorism are striking.  Where evolution seems to kick God out of the equation, making biological selection possible through an entirely naturalistic, deterministic process, behaviorism seems to kick Man out of the equation in a similar sense, making cultural selection possible through an entirely materialist process as well.  Where religious objections stem from a fear that their beliefs become nullified, objections to behaviorism also seem to arise out of fear: that their belief in the mind becomes nullified.

Just as religion looms old as history itself, and in this manner seemingly self-evident, so too does the notion of man's free will and self-determination.  But both are a subtle matter of framing.  A religious man looks around him and says, how could all of this come about without a masterful God?  So to the free-will romantic looks within himself and says, how can I not be in control of my actions?  

But as we began to understand the mechanics of atoms, fundamental forces and laws, so too were we able to see the universe as able to give rise to the astonishment we see around us merely through materialistic interactions.  We then began to realize that our assumptions were themselves the product of cultural histories and particular to our time and place.  As Pavlov began to discover the two-term contingency: connections between reflexive responses and environmental stimuli, and Skinner began to discover the three-term contingency: connections between our responses, environmental stimuli, and the probability of future responding, we can now see that the universe gives rise as well not only to our actions but our thoughts.

I think in many ways it has been easier for people to accept biological evolution than radical behaviorism.  For one thing, the outside world is by definition objective, and much easier to control and manipulate.  When you design an experiment to measure rainfall, continental drift, or star formation, no one worries about the ethical considerations.  But when you begin to think about testing and controlling human populations, people naturally get spooked.  And the cultural context for scientific paradigm shifts can't be understated.  Radical behaviorism was literally being developed at the same time millions of Jews were being sent to gas chambers in Europe as a "scientific solution" to social problems; the rise of communism was building entire states on the notions of mass population control and social engineering.

But Nazism no more disproves the theory of evolution than totalitarianism disproves radical behaviorism.  Sure, it raises some very urgent questions about how we structure our society.  But at the same time, from a behaviorist perspective, so does the romantic notion of free-will.  The difference is that the former is based in science, the latter on intuition and cultural tradition.  

Video game makers and casinos aren't concerned with whether we accept it or not.  They simply do what works.  The choice for us is whether we want to be aware of why we respond in the ways we do, so that we might at least have a clear-eyed perspective on the world we live in, or whether we would rather dismiss the science in favor of what we would prefer to believe in.  The stakes go far beyond video games.  The principles of behaviorism shape every behavior we engage in, and describe how our behavior will be determined by our public institutions, our economy, our schools, our courts, and most directly, our intimate relationships.  Every problem we have in life involves, by definition, a human behavior that is located within a complex environmental system of selection.  The key to knowing ourselves, and our world, lies in understanding this process.

Sunday, August 30, 2015

A Sleeping Giant

The blog Lunatic Laboratories has an great summary of an interesting paper in The Journal of Child Maltreatment titled, "Self-Criticism as a Mechanism Linking Childhood Maltreatment and Maternal Efficacy Beliefs in Low-Income Mothers With and Without Depression".

The study found that mothers who experienced more types of abuse as children–sexual abuse, physical or emotional abuse, and physical or emotional neglect–have higher levels of self-criticism, and therefore greater doubt in their ability to be effective parents....
Prior research has found that a mother’s confidence is closely linked to her motivation to use positive child-rearing strategies.
My comments on the blog are as follows:

I provide ABA services to many low-income mothers who might be described well by this article. However, (and if you're familiar with the radical behaviorist perspective, you'll know this) we don't go in for such mentalistic talk of "sense of self-efficacy".  Not to say we wouldn't agree with the basic premise.  But try this: mom's history of reinforcement hasn't taught her the requisite skills.  The difference between those who are able to implement correct strategies in stressful situations versus those who aren't won't be doing so because they "feel more self-efficacious", but rather that they have simply attained fluency.  I.e. they have practiced and learned these skills so that they can use them despite difficult setting events.  For instance, if I have a headache (or worrying about the bills, my boss, the broken car, etc.), playing a piano song will be more difficult.  But if I am fluent, I will be able to engage in the behavior.

An absolute failure of society right now is adequate delivery of treatment services to mothers in need such as these.  In California, if your child has an autism diagnosis, health insurance companies are mandated to cover behavioral services.  Based on need, these can range from 8-15 hours per week.  The amount of progress children make is nothing short of astounding.  Kids learn to speak, to communicate their desires, to label items and hold conversations.  However, much of our work is also in parent training. 

Where this becomes interesting is in multi-child households where parent-child interactions are often dysfunctional, i.e. maladaptive behaviors are reinforced and appropriate behaviors are punished.  I am not paid to provide services for the rest of the family, but a great benefit of our services is that the same strategies parents learn to use with their special-needs children are those they should be using with their other children.

We've made great progress in the U.S. when it comes to those with disabilities.  However, when the disorder is not genetic but the result of a learning history, our sense of moral responsibility, let alone socio-economic pragmatism (well-functioning families are more productive in every way - for generations!) withers.  I see two important roadblocks to overcoming this: first, it is a common belief that lower-SES people have only themselves to blame, and thus can only help themselves; and second, we lack a coherent, comprehensive understanding of what is truly causing poverty.

The science of behaviorism is deterministic and believes all thought is behavior, and that all behavior is a product of operant mechanisms.  As such, it's answer to the former roadblock is that SES is merely a function of social history and structural formations.  Just as you would not blame a town for being flooded by a river and would instead try and help them recover, so too would you the individual.  We do this with disabilities, why not learned behavior? 

That question leads to the second, greater roadblock: the philosophical, political reaction so many have to the radical behaviorist hypothesis.  The history of behaviorism's place in psychology is illustrative: it's marginalization in the fifties and sixties after Chomsky, it's quiet growth into ABA in the seventies and its refinement in the eighties, it's establishment as the pre-eminent treatment the nineties and oughts for autism and behavioral disorders.  It took decades of tireless work and empirical results to barely make a dent in public and academic acceptance.  However, it remains poorly understood, even as its principles remain as true and central to the human condition today as they were when Skinner published Verbal Behavior in 1957.  

I would argue that the reason for this is simple: a deep human bias towards the notion that we have free will, that we are not determined, that "we" are somehow special, possessive of some mysterious, supernatural "consciousness" from which we make all of our choices.  When confronted with evidence to the contrary, many experience a deep revulsion at the mere thought, a sort of existential terror.  What, many will reply, is the point of anything at all then!  

The roots of this (reinforced) notion can be found in various forms of cultural imbibement: religion, philosophy, politics, art.  To one swayed by this mythological cacophony, to accept the radical behaviorist premise is something akin to death.  And as we have seen time and time again, when empiricism and parsimony comes up against the forces of belief in legion, it often only emboldens its enemy. 

And behaviorism is not a simple concept.  I spend very little time with families explaining the deeper processes at work (the discriminative stimuli, the matching law, schedules of reinforcement).  I give them what is necessary and show them how to use the tools of behavior change.  For some, it is rather simple.  For others, it goes against everything they would consider "common sense".  But you can't argue with results.

In many areas of the political arena, change has come kicking and screaming, after results were plain as day.  Laws were passed that people were uncomfortable with at first, but then one day would question how they could ever have been otherwise.  

Unfortunately, things like racial civil rights and rights for people with disabilities don't necessarily clash with larger mythological forces anything like that of the notion of "free will".  But how about the right to a learning history of behavioral cusps, that maximizes one's ability to access to reinforcement?  Or the right to having a deficit in learning history be remedied?  In colloquial language we call these things public education, or jobs programs.  But the underlying premise if flimsy: only a basic education, only for kids, because kids don't really have free will yet (and even then, we'll basically ignore the vastly different levels of learning histories and social capital different kids have).  Jobs programs if you're lucky, because maybe it wasn't really your fault you don't have the right skills.  Maybe.

These two pitiful solutions are vague compromises made by a larger population that mostly believes that people are responsible for their own decisions.  To the extent they are provided, it is out of an intuitive sense that there isn't something quite right about massive poverty,and therefore charity is a moral duty.  

Yet the problem persists.  And so mythologies are heaped upon mythologies.  The remaining poor - even after public schools, welfare, jobs programs (and now free healthcare!) are seen by one side of the political spectrum as simply making bad choices.  Why?  Who knows!  By the other side, they are largely seen as helpless victims of racist cops and lazy schoolteachers.  Both take as a baseline notion the idea of free will and personal responsibility.  But one side sees it as all the individual's fault, with a perfectly functioning system, the other as no fault of the individual, but rather an imperfect system which does not allow even a perfectly functioning individual to succeed.  

But to the behaviorist, all can be explained by looking at individual histories.  Indeed, the system is imperfect and broken, and has been for centuries, and as a result, learning histories have been repeatedly left to become mangled and neglected over generations.  The environment and the individual cannot be separated.  Every aspect of one's life, every stimulus that one experiences in the course of the day, every schedule of reinforcement or punishment upon which one is placed, actively shapes one's behavior.  

From this perspective, blame is irrelevant, meaningless.  All that matters is intervention.  With this in mind, social policy can actually begin to apply the necessary mechanisms.  I see on a daily basis the effectiveness of ABA in the lives of children with behavioral disorders.  The treatment is a result of decades of science being accepted and applied by a social infrastructure that recognizes not only its efficacy but feels an obligation put it into action.

Humanity is behavior.  Behavior shapes humanity.  I believe one day we will live in a world in which each citizen has a right to a meaningful learning history.  And if that learning history is disrupted by accident, just like any illness, they will have a right to treatment by a qualified, trained professional.   No one will be left to "make their own choices", because the concept will no longer make sense.  Certain parameters of acceptable behavior will be allowed, and considered "freedom", as they are now.  And others won't.  But criminals won't get "justice" anymore than successful business-people will: both will be seen, correctly, as products of their learning histories.  

And women like the ones in this article won't need to exist.  Their sense of "efficacy" will not be seen as residing in them, but in all of us.

Sunday, August 9, 2015

Segregation and Responsibility

In the New York Times today a very interesting article and commentary.  I found myself largely in agreement with the thrust of the commentary: that the article places the function of its subjects' struggles in racial discrimination, when it is much more a matter of personal "choices".

Again, we see the conflict of narratives: poor minorities are helpless victims of discrimination vs. poor minorities are not making the choices that would bring them out of poverty.

Our political discourse is mired in the intransigent byproducts of this binary: one side is callous and, the other side is unwilling to deal with the truth.

To my mind, both are true.  But the key goes back to the problematic word "choice".  What does it mean?  Most people have only the vaguest notion that they are in control of their lives; an odd "sense" that they are autonomous creatures that "do what they want to do".

However, decades of experimental science tells us otherwise.  Our sense of "choice" is merely a form of labeling events and activities within ourselves and in the outer environment.  Things happen, and we label them; we "understand" them as things that have happened.  We can also label things that will happen in the future.  But in both cases, we are never doing more than describing - first to ourselves, then to others, behaviors that are beyond our control.  That is, they are subject to our history of having been reinforced for engaging in similar behaviors.  We are either following a rule of what the consequences are likely to be based on logical relationships, or on the actual contingencies involved (i.e. I've tried new foods before and they were gross so I'm not going to try them now, or that knife is sharp and I know I'll cut myself on it).

The greatest evidence of this is what is often called the "subconscious" - the behaviors we engage in without being aware of them - without labeling them.  So being "conscious" is merely being able to label one's behavior.  Most of our daily behaviors are in fact unconscious.

So why be conscious at all?  Because the behavior of labeling behaviors is itself subject to a history of reinforcement.  One can "become more aware" by receiving greater levels of reinforcement for specific labeling behaviors.  This isn't easy.  It has been said that we are awash in a "sea of reinforcement"; at each moment our mind is subject to an incalculable amount of past histories of interaction between our environment and our genetic make-up that controls our every thought and behavior.  This is the function of action.  This is the function of "choice".

So when we talk about poor minorities making problematic "choices", what are we really talking about?  History.  That history is filled with everything you might imagine.  In the article and comments, it is pointed out that a 25 year old woman is unmarried and has 3 children, and is not cleaning the mold off of her baseboard.  Why?  Sure she has "chosen" these behaviors.  Yet did "she" really "choose" them.  In the same sense, does a middle class 25 year old woman "choose" to postpone children, marriage and go to college?  But did she "choose" them?

In both cases, the behaviors were a function of the sea of reinforcement contingencies in both women's lives.  Neither woman is "free".  No one is "free".  No one is "responsible".  "Blame" does not exist.  No more so than clouds are to "blame" for rain, instead of climate.  No more so than a falling tree for a crushed house, instead of a windstorm.  There is no freedom, only causes and effects.

So, the real question is: how do we as a society design our systems so that each of us gets what we want? If there is no freedom, how is this even possible?  Just as evolution occurs in nature by natural selection, our culture evolves by selection.  We will do what is most reinforcing.  MLK talked about "the moral arc of the universe" bending towards justice.  I'm not sure if that is true or not.  But humans will select ideas, and some ideas will provide access to reinforcement, and so on.  Just as the shark millions of years ago evolved into a phenotype so perfectly adapted to its environment no offspring could evolve into anything much better, maybe human culture will finally select ("bend") towards a set of behaviors that in totum result in no more perfect system of thought being more capable of delivering what we want: safety, comfort, love, opportunity to explore our world and be fascinated.

I don't know how to deliver more opportunities for reinforcement to poor minorities.  I have ideas.  But many are merely hunches.  I do know however, that any ideas need to take into account the fundamental laws of human behavior if they are to be successful.  We do some of this already without even knowing it.  But mostly we don't, especially when we use the language of "choice", "blame", "responsibility" and "freedom".  These are backwards, antiquated notions inherited from ignorance, and in the end are responsible for vast levels of human suffering.  The sooner we dispense with them the better.