Game Theory


header notes:

A game-theoretical analysis of play & human behavior... A game-theoretical analysis of human-computer interaction... Instructional design can learn from games... Title of paper: "Implications of a biological epistemology for education" Subtitle: What Gregory Bateson, Marvin Minsky and Gerald Edelman can teach us about the mind, and what it means for teaching ((Do we add Eric Berne & Transactional Analysis? Does "Games People Play" fall under this umbrella? ...."Game theory and the games that people play?")) ------------------

Big picture time:

I need to connect game theory with learning theory, information theory... Gregory Bateson may provide the "pattern which connects"... Gerald Edelman may provide the neural substrate which elevates this from merely possible and plausible to most probable explanation... Actors read from scripts and follow plots; agents pursue strategies and obey the rules of the game. Technology lies in service of human interactions: we humans are game-players. Learning follows game-playing patterns Instructional design heeds/honors/hones game-playing strategies and instincts... ------------------ Body: In play, and in work, the desired outcome is shared focus. Shared attention, the ability to communicate, is contingent upon shared focus. And in instructional design, the outcome is controlled focus or coherent focus... guided focus. For the reader, the audience to collectively have its attention and focus manipulated in a coherent and cogent fashion, that you are instructing, you are passing along some narrative, that instruction is happening. [[(*There is an objective means to pursue things, an interface) there are empirical means for determining if an interface is achieving its didactic goals.]] Friends and families naturally pursue this kind of coherent focus, something which has been recognized from Maria Montessory through Bateson, Maslow, Rogers and Wolfberg: the importance of shared play and imagination. There is no reason why study groups, homework groups, and office-work groups cannot utilize the same organizational structures. The relevant aspect of such groups here is that they are free, dynamic: that such organizational structures allow for the freedom/dynamism that play affords. Given a group of people, diversity is not merely an inevitable attribute or consequence but a central feature and distinguishing characteristic: other people are different, simply put. Whether you categorize people according to Meyers-Briggs, IQ, emotional intelligence, multiple intelligences or even the astrological system is almost immaterial. It is enough to recognize the view articulated by Marvin Minsky in Society of Mind, namely that each one of us is necessarily composed of a great many agencies, faculties and functions, each with their relative strengths and idiosyncracies, and recognize that we respond not all equally well to the same learning interfaces or modalities. Some of us are visual thinkers, and process icons well. Some are both literal and rational, and prefer lists or text labels. Yet others are kinesthetic learners, who will not acquire new strategies for coping with the physical world unless they have actively engaged the world with their bodies in practice. It is important to develop learning spaces which enable exploitation by various kinds of minds hungry for various forms of interaction. Play requires different actors, engaged in different activities. A biological epistemology would expect and predict this diversity: diversity in sex and handedness and sexual orientation, diversity in personality types, diversity of learning styles. Bateson's differences which make a difference and Minsky's society of mind agents are both completely consonant with Edelman's explanation of consciousness as being emergent from the diversity of neuronal group selection, the diversity of perceptual and behavioral repertoires. ------------------

Structural notes:

Gregory Bateson: implications of a cybernetic epistemology. Bateson describes perception and learning as both being stochastic processes Edelman confirms that our neural mechanisms are in fact lossy and interpretive stochastic networks, assigning meaning to fuzzy constructions of motor and perceptual maps. Bateson insists that mind is an aggregate of interacting parts, a notion which is completely consonant with Marvin Minsky's Society of Mind construction ------------------

Bateson's criteria of mind:

(1) Mind is an aggregate of interacting parts or components. (2) The interaction between parts of mind is triggered by difference. (3) Mental process requires collateral energy. (4) Mental process requires circular (or more complex) chains of determination. (5) In mental process the effects of difference are to be regarded as transforms (i.e., coded versions) of the difference which preceded them. (("Air pressure differences ==> Fluid Pressure Diffs ==> Cilia root pressure diffs ==> Neuronal State differences ==> etc." -- )) (6) The description and classification of these processes of transformation discloses a hierarchy of logical types immanent in the phenomena. ------------------

notes from lunch:

Games follow a playful syntax The grammar of play involves hesitancy, ...just as... the semantics of play involves irony. A playful grammar is free, flexible, dynamic... [Maslow's Eupsychian Management] Eupsychian Dynamics --> Play Structures

  • iterative structures of dialog
  • call-and-response colloquy
  • Illiad, Odyssey
  • reading stories to children: "interruptions" are the norm, not the exception
  • random walk, co-authored dialog

It has been said that the ABCs of psychology are affect, behvavior and cognition (Gleitman, 1989). As such, it is our job as educators to help people to better think, feel and act. Towards that end, let us examine the nature of human beings and how they achieve these no-mean-feats of affect/feeling, behavior/acting and cognition/thinking. Nobel Laureate Gerald Edelman has described his work (in rather grand fashion) as "completing Darwin's program." It shouldn't be too surprising, then, that his is the only theory of consciousness which explains all pertinent aspects of mind while remaining firmly grounded in neuroscience and completely consistent with evolutionary biology. Although his work is notoriously difficult to understand, it proves quite fruitful in terms of producing conclusions useful for educational and social theory. Before getting down to the nitty gritty, let us consider the big picture. Evolution moves in stages. Natural selective forces act upon variant agents (individual genes, organisms, entire species) to produce--over huge scales of time--dramatic mutational changes. Just as ontogeny recapitulates phylogeny (the grand "as above, so below" of biology), there is a parallel between the evolution of consciousness for the species and the dawning of consciousness within a given individual. (In fact, every morning we experience a microcosmic recreation of the dawn of animal consciousness as we wake from sleep!) There are also distinct stages within the evolutionary development of a given human mind. The human being is an adaptive cybernetic organism, with both internal neurophysiological and external environmental feedback loops. Consciousness is adaptive and embodied, and it is focused/directed/attentive... and what we attend to / are pointed towards is largely other human beings. We are social creatures. Our cybernetic structures, complete with internal circuits and multiple competing strategies, coupled with our external social focus, makes us flexible and playful in our application of differential strategies for navigating our surroundings. We are game players. Establishing all of this, our inherently cybernetic and social nature, will be a long and arduous journey. But we will be richly rewarded if we undertake it, for it will reveal to us the best practices of education, both in child-rearing and in vocational development. Gerald Edelman's theory, as has been noted, is fiendishly complex. Since his is a theory of the brain, and the brain is the most complex piece of matter in the known universe, this should hardly be surprising. Edelman's natural selective view of the brain is consistent with both cybernetics and evolutionary biology. Edelman's theory is:

  • stagelike, consistent with Piaget
  • adaptive and variable, consistent with Gardner's Multiple Intelligence and Daniel Goleman's Emotional Intelligence
  • hierarchical, consistent with Kant, Russell/Whitehead, Wittgenstein, Rosch, Lakoff/Johnson
  • natural selective, consistent with both Bateson and Minsky

On this last point, some qualification is required. Marvin Minsky's theory of the society of mind may be considered "merely useful" in explicating Edelman's work; we shall assumed it can be used without much controversy. Establishing that Edelman's work seems strongly consistent with some parts of Gregory Bateson's arguments, however, becomes much more controversial a basis for extending reasoning from the whole of Bateson's thinking. This is a leap of faith I am willing to make, and a controversy I am willing to risk, based in no small part on the great reputation that Bateson's work enjoys in certain circles (Jasper Hoffmeyer, et al.) and on the extraordinary power that his theories command. It is, quite simply, easier, more fruitful and less confusing to think about biological and psychological phenomena through a Batesonian lens than not.


The fractal explosion of life

the branching of a growing tree as a metaphor for embryological development

cell growth, tissue growth continuing patternlike until a stop marker is hit

much like folic acid will trigger the sequence of genetic expression resposible for proper neural development: anencephalia if too soon, spinubifida if too late

Gregory Bateson wrote of embryological development that it more useful to think of the hand as having been formed by four branching events rather than thinking of the hand as being composed of five fingers.

limb segments and tree branchings are both the result of iterative genetic expressions.  each go round, the cells being grown by genetic expression form chainlike for a few iterations.  from major branch to lesser branch to smallest branch to leaf; from arm to hand to finger segment.  from one finger segment to the next: instead of thinking of our finger as having three segments, we should think of our finger's developmental history having two major branching events, the "scars" of which function as our joints.

and so the neural patterns are laid down.  both Eric Kandel (1989) and Edelman (1986) emphasize that the brain is the result of not just cell growth and migration but especially of cell death: we have lost two-thirds of all the neurons we have ever had by the age of four years old, and "the neurons that fire together wire together."  The brain cells that lived are the ones that learned to signal each other, typically the ones connecting mappings from the environment to useful motor output.

this randomness in the individual layout of neurons on a brain-by-brain basis is far from accidental, rather being the central feature of neural architecture. we are the emergent property of that neural complexity.

the various branchings produce our neural roadmaps, kind of like the maps of the freeways and roadsystems within a state or country.  that is what Edelman calls the "primary repertoire."  the actual traffic patterns from minute to minute and day to day are the "secondary repertoire."

(this final fragment is finished on my personal website.)