Notes: Complexity Theory, Getting the Client to Maybe
Evolutionary Psychology (Brody)-
Healing The Moral Animal: Reading List
by James Brody, 3/3/98
Notes: Complexity Theory, Getting the Client to Maybe
by James Brody, 9/7/98
A COMPLEXITY MODEL AND MENTAL HEALTH TREATMENT ... GETTING THE CLIENT TO "MAYBE."
Thursday, July 23, 1998. 7 P.M. - 9 P.M.
James Brody, Ph.D.
Host, Evolutionary Psychology, Behavior OnLine
18 attended this session, part of the 19th Cape Cod Institute seminar on "Healing the Moral Animal: Lessons from Evolution," a follow-up on "Clinical Sociobiology: Taking Charge of our Genes." Notes were completed 9/6/98.
There are several paths that merged into this seminar.
1) I was privileged to hear a talk by Jon Schull on the topic of evolution at a Cape Cod Institute in the summer of 1996. Jon has ties to the Santa Fe Institute, itself dedicated to the study of complexity and emergent order. The trail from Jon led immediately to Stu Kauffman through a book by R. Belew & M. Mitchell, Adaptive Individuals in Evolving Populations: Models and Algorithms, 1996, Addison Wesley, Reading: MA.
2) Clinical observation reveals that some clients are particularly sensitive to small changes in medication dose and type as well as to slight alterations in their environment, whether a spoken comment or the triggering of an old memory. Such clients can be our toughest to help. Complexity theory nicely describes these phenomena as well as implying remediative steps.
3) Kauffman (1995, At Home in the Universe, NY: Oxford; 1991, Origins of Order, NY: Oxford) discusses chaos, stasis, and phase transitions in order to account for the rococo biological organizations that surround us. "Complexity" is a puzzle in a universe assumed to be predicated on disorganization and heat loss. The model describes life as a relatively unsurprising event and evolution as a directed beyond the steering it gets from natural selection. Kauffman's models are mathematical/logical derivations and appear valid regardless of the arenas to which they are applied. Thus, I'm intrigued by applying them to models of CNS organization and to models of human organization. there are powerful implications for understanding our cognitive assets, language, social organizations, and even morality.
BASICS
"Complexity" refers to the development of organized structures (including living ones) from simpler ones and Kauffman moves from the interaction of simple molecules to the development of more complicated ones, from interactions of 2 molecules to networks of 100,000 or more.
Networks are a collection of decision units; a "decision" by the entire net is a function of the units that compose it. Each unit (gene, neuron, human being?) can be inhibited or excited by the activity of other units in the system. Decision units follow Boolean rules such as AND, OR, NOR, NOT OR. An "AND" unit will only fire if two specified inputs occur. An "OR" unit will fire if either input or both of them occur. Certain rules, OR and AND, seem more likely to produce coherent output from the net.
Once you have a collection of units -- molecules, genes, neurons, or people -- there are 3 states that the network can occupy -- chaotic, orderly, and phase transition.
"Chaos" refers to there being too much interconnection between units. The action of each one affects activity of all the other units. A network of only 200 units would require billions of times the age of our universe to repeat a pattern. Prediction depends on repetition of events. Given that there is basically no repetition, a change in one bulb eventually changes the activity of every other bulb in an unpredictable manner. If you systematically increase the number of connections between units, CHAOTIC BEHAVIOR OCCURS WHEN THERE IS MORE THAN 3 CONNECTIONS AFFECTING EACH UNIT. Chaotic behavior is a more a function of the number of interconnections and is influenced only in a trivial manner by the number of units. If every gene talked to every other gene, if every neuron talked to every other neuron, there would be immobility just as there is for a mother who attempts to do everything personally when she has more than 2 children.
"Order" refers to there being one or two interconnections between units. In a network of 100,000 units (bulbs, genes) with 2 connections between each unit, repetitive activity for the entire network occurs in only 317 cycles. "Order" occurs very quickly in comparison with networks with more than 3 interconnections.
A "Phase Transition" is the zone between chaos and stasis, wherein there is SOME orderly output from the network and some constraints on that action, constraint that is a consequence of the cross talk between decision units. Three interconnections underlies a narrow zone between chaos and rigid order, wherein natural selection can operate. That is, adaptive patterns will survive; maladaptive ones are eliminated.
In the logic circuits that Kauffman describes, the phase transition occurs in the range of 2.5-3.5 connections (with some fudging upward or down as a result of the type of connection that is made) between units (again, neurons or people, light bulbs or genes -- it doesn't matter).
Neither chaos nor stasis is affected by the number of decision units. A mob can have 5 members or 10,000. Steam can have a million H2O molecules or a million times more.
The essential element for switching from confusion to cement is that of the interconnections between decision units. A committee with 5, equally active, members will generally accomplish little unless a hierarchy is formed or subcommittees (the same thing!) are created. More than 4 members means that some of them are being silent, uninvolved, or forming subgroups that act apart from the main body.
Most of the significant genetic affects on behavior appear to involve networks of between 2 and 10 genes (Bailey, 1998).
Phase transitions are common and often narrow. For example, there is a very large range of temperatures below 0 degrees C. There is a very large range of temperatures above 100 degrees C when water becomes steam. The phase transition between 0 and 100 degrees C is very narrow but allows life to exist.
"Phase transition" can also be used to describe:
1) The meeting of oil and water
2) Some bacteria live in dust deposits in Arctic ice. The darker color melts the surrounding ice and the bacteria thrive. The darker liquid is also heavier than the surrounding ice and sinks gradually until insufficient light reaches the dust. At that point, the liquid freezes and the bacteria die. However, each layer of bacteria seems to produce the one immediately higher so there is a continuous rotation upward of the living colony.
3) The number of children in a family cause it to move from a sense of freedom to cement. Two adults can form a union and retain great personal autonomy for travel, housing, eating, and social activity. Adding one child significantly slows things down. Adding a second child moves the unit to 4 elements and "cement" applies. The constraints on parents include physician appointments, finding sitters, satisfying relatives' inquiries, church demands, and shifts in sleep (both the arrangements and the amount that is available). Families (or any group) with more than 3 members choose between immobility or they form hierarchies. A mother with more than 2 children will have to create a hierarchy if she wishes to remain effective. The older help with the younger. Mothers who cannot form a hierarchy are immobilized (perhaps a precursor to panic, elicited in older days by restraint and by a sense of confinement).
4) It may be that we are a hierarchic species not because of random choices or because of some adaptive "need" to control aggression but because of the numbers of complexity theory. WE CANNOT FUNCTION IN A COHERENT GROUP AND RETAIN MOBILITY UNLESS A HIERARCHY IS FORMED. A hierarchy reduces the number of simultaneous demands exerted on every member of the group. Polling everyone's opinion preempts ALL organized group activity. Chaos (cement) prevails because every member affects every other member on every issue and we orally challenge, "Who's in charge here?"
5) Phase transitions might usefully be compared to the notion of "set points" in human moods. Love, anger, laughter, bonding often have a quality of all or none. Our primate minds seem to have an "Us/Them Chip" in that we quickly label other people as for or as against us.
6) Some psychiatric disorders have characteristics of extreme order or of chaos.
A. Impulsiveness. Rigidity and internal order are seen in impulsive behavior whether reflected in changes of mood, shifts in goals, making sudden purchases, experiencing panic or social phobia, head injury, bipolar disorder, borderline personality disorder, or antisocial personality disorder. All of these disorders could reflect insufficient cross talk (perhaps via lateral inhibition) between decision elements within the person. Although external behavior can appear erratic, the internal processing can be highly orderly and not very complex as the person reacts in an extreme fashion even to small shifts in their settings.
Impulsiveness is a negative sign, regardless of other characteristics of a disorder. This observation is based on the correlation of impulsiveness with suicide risk, assault, bankruptcy, traffic accidents, early parenthood, higher frequency of divorce, and premature death. Impulsiveness will be related to the clients' failing to complete homework drills, collect accurate information, or to internalize directions that are given.
B. Immobility. An internal chaotic state is perhaps seen in schizophrenia and some anxiety disorders in which immobility occurs when the client cannot "choose" between competing options. There is perhaps too much cross talk, too much weighing of options and every small past experience produces a shifting of goals, plans, and memories.
A GOAL OF THERAPY: THE "MAYBE RANGE"
Less impaired people have more selective fears and can reason about them or take steps to collect information and direct avoidant behavior more selectively. They are like "dimmer switches" in which there is a graded response to changes in input. The light goes a little brighter or a little darker with changes in the switch; likewise, in the person there is a little fear or a little anger that serve to mobilize planning and adaptive sequences without eliciting a full scale escape or attack.
Flexibility without immobility, and a lack of impulsiveness are positive signs and generally synonymous with "maturity." The client has the ability to say "maybe" and check memories, examples from parents and friends, or seek advice on the internet (still mildly impulsive) or library (not impulsive at all!)
A range of interventions increase the "width of the Maybe Range" and can stabilize the client.
A) Things that Slow the Client Down by Increasing Cross Talk
Consultation with friends -- ordinarily a good sign, especially if the client is willing to consult with friends who disagree with him. The alliance is intact despite disagreement over a particular issue.
A therapist increases the number of inputs on the client; the therapeutic contact in itself and independently of therapeutic persuasion, should decrease client impulsiveness.
Marriages, social clubs, church affiliations, pets, magazine subscriptions, and even golf increase the number of simultaneous demands on a client and reduce the odds of impulsive activity. (Nathan Azrin once treated alcoholics by getting them 3 jobs, 3 clubs, and a raft of magazine subscriptions. He left them no free time for impulsive behavior.)
Cognitive therapy drills the client in what am I feeling, what actually happened, what are other interpretations, and how do I confirm my suspicions? The result is an increase of "checking with' other experiences to make feelings and actions less impulsive, less extreme.
Moral teachings slow down impulsive behavior by making some actions (marriage and procreation) socially irrevocable. Moral teachings also slow down actions that are biologically irrevocable such as murder. Moral lessons give even the isolated a base of examples and instructions that are internalized checks on social conduct.
B) Things that Speed Up Decisions
Some clients need to move out of chaos.
"Enablers" are constrained too much by real and expected consequences from a partner; therapy may consist in getting them to act with less consultation with their domineering partners. Dependent personalities can have a similar tactical condition even outside of marriage demands.
A spouse can be "trapped" in a marriage despite feelings of hopelessness and panic. Financial obligations, children, demands of kin or religious contacts, pets, and magazine subscriptions cement the relationship. (One of the most certain ways to produce ulcers in a white rat is NOT electric shock but immobilization. Simply wrap them in chicken wire.)
Such people are like the telephone pole a mile from my house. The pole was sheared in a bad storm but remained upright, held in position by the 15 or so wires that crossed its top. Sometimes we need to help people cut some wires, gaining some freedom and contentment with each severed obligation. Such things as a cleaning or lawn service, relaxing rules about daily tasks, or reducing the number of relatives and friends who have to be called daily ... all reduce chaos.
Form a hierarchy ... instruct an anxious mother that her older child is designed by nature to produce children of their own at 13 yrs. and can take many responsibilities for helping younger siblings. Likewise, our children have always been reared by groups of adults; relatives (and day care staff with the right attributes) and neighbors can be substantial help.
Find a model for the anxious to imitate.
TREATMENT:
Assessment and diagnosis should give explicit consideration to impulsiveness and immobility, regardless of whatever complaints the client may have.
Impulsiveness will vary with areas of talent and aptitude. For example, some impulsive ADHD children are not at all impulsive when caring for children. Some impulsive drivers are quite methodical architects. Some financially impulsive people are extremely methodical when managing their social contacts. Identifying islands -- of either impulsiveness or of good self-regulation -- could be an important step.
Procedures for impulsive people should increase the amount of cross checking that they do before acting, either with other people or within their own minds. Such procedures might include:
Physiological tricks of proper diet, exercise, and sleep pattern. (Reports are beginning to appear about the possible effectiveness of sleep regulation for controlling rapid cycling.)
Adjustment of social demands. One young lady commented to me that she could not get in trouble or spend excessively so long as she was busy working. Three jobs seemed optimum for her. Enrolling someone in several list_serves (there are 67,000 from which to choose) may have a comparable effect.
Form hierarchies or simplify them, depending on the nature of the problem.
Adopt pets, acquire hobbies (plants demand recurring attention providing the client has some minimal interest in them), practice a sport (marathon training, for example, makes tremendous demands for time management and impulse control -- impulsive marathoners often collapse against the curb late in the race), take up a religion.
Cognitive therapy, particularly along the lines given above and additionally, "check with someone who asks good questions." "Can you check with someone who disagrees with you and still listen to them?"
Eliciting competing psychological adaptations. Former loves, teddy bears, and childhood scenes usually elicit tranquillity in a frightened person.
Morality is likely genetic to the same extent that our grammar is. (It regulates cross talk between people, a variable far too important to be left to social learning methods alone.) Moral sensitivity will vary between clients just as every other trait does. It has an important function, of internalizing the amount of cross checking that must be done, even by an isolated person, before he or she takes action. Morality represents an accumulation of rules about situations, rules that usually work in the direction of long term individual happiness and group survival. Morality can be an important tool for the client's self regulation. Despite the "value neutral" stance of many therapists, the client's existing rule system might be reinforced or challenged so that it allows him freedom within the bounds of good sense.
NOTE:
There are other people saying highly similar things but sometimes without referencing Kauffman. His writing can be difficult but, nonetheless, highly rewarding for people interested in evolution and genetics. He cannot be scanned even if you are familiar with the material; some modern writers may find it easier and more exciting to "discover" the same things for themselves.
He appears to establish a ladder that moves from simple molecular interactions through genetic actions. His model is startling in that it predicts and explains genetic functions that have stable outputs despite a high degree of organizational variability in the genes! Earlier notions of natural selection's enforcing genetic uniformity seem dispensable and perhaps fundamentally wrong. It seems possible that genetic variability allows each of us to be "different" from one another in some ways and very much "alike" each other in still other ways.
Genetic "tuning" is addressed ... tuning by each other and plausibly by other events that occur to the organism. Genes as "conductors" continually learn from the orchestra and appear to provide a gamut of strategies that are evoked as a function of age, setting, and changes in social demands. (Goodness, teenager infatuations represent a shift in genetic tuning! No surprise to most parents.)
Despite all the "complexity" of genetic activity, it appears that the numbers of genes (like the numbers of neurons and the numbers of people) taking part in any one decision is on the order of 3 and that involving more units will slow things down, involving fewer usually speeds them up.