Replying to LO20246 and LO20345 of Leo Minnigh
also touching on
LO20363, Entropy production and emergence in behavior (John Gunkler)
LO20632, Democracy and the learning organization (Richard Goodale)
LO20616, Is learning our earnest (At de Lange)
Greetings to all,
In LO20246, Leo Minnigh wrote:
>If these questions are of some valid[ity], we have some clues for the
>behavioural sciences.
>1. We could increase the flow speed, by creating a difference. Either by
>pushing, or pulling. Should we make (in reality or in the mind)
>'something' attractive, or rejective? Should we fire employees, or should
>we offer them an attractive alternative, or do we stress the rejective
>elements?
>2. We could increase the flow speed too, by combining the push and pull. How
>Could we keep the effects under control? Does immergencies, or emergencies
>occur?
>3. We could create oscillations by creating two attractors, or two
>rejectors (firing employees in an overloaded marketplace). What will be
>the behavior of humans in such environment? Real life looks often like
>this. ...snip...
and in LO20345, Leo wrote:
>This is probably the right moment to mention something which struggle for
>a long time: addiction, enslavement, like drugs, alcohol, over-eating
>cigarettes, gambling.
>All these things are great attractors, although the victom usually knows
>very well that it is unhealthy. The mind says NO, the heart says YES. The
>short term consequence is desired, the long-term not. With all our
>cooperate thinking, there are several possibilities to handle this
>world-wide problem.
>1. Push the victom in another direction
>2. Punishment after use
>3. Make the attractor unreachable (legal and physical barriers). (BTW, one
>can see that if the attractor is strong, this method is of limited value)
>4. distract the immediate attractivity
>5. create another strong but less harmful attractor.
Greetings Leo!
As John Gunkler observed in LO20363, you are struggling very successfully
with behavioral principles. Your thoughts above are evidence that you are
beginning to see the systems of behavioral energy in the world around you
through the behavioral lens, and it is helping you identify the "levers
for change".
All of the solutions that you mentioned above are possible strategies that
may be taken to change behavior. Your list is not exhaustive (and I'm
sure it was not an attempt to generate an exhaustive list), but it is
certainly representative. John mentioned another approach that might be
taken, that of decreasing the attractiveness of the consequence that is
maintaining the "undesired" behavior. Sometimes this is relatively easy
to accomplish, for example, in situations where "missing ignorance" (to
borrow At's term for ignorance due to no prior learning) is at work. In
these situations, simply bringing the "downside" to the attention of the
individual will be enough to get them to change/reorganize their behavior
(an antecedent intervention is sufficient).
Sometimes, however, it is much more difficult, as in the specific cases of
prominent addictions like drugs, alcohol, over-eating, cigarettes and
gambling. Gambling is its own beast, and requires a fairly extensive
level of consideration to explain (John mentioned schedules of
reinforcement as playing a large role in this. It does. At some point in
the not-too-distant future, I hope to post something on this, but I won't
get in to it this time). The others that you mentioned, drugs, alcohol,
over-eating, and cigarettes, all have something in common, that being,
that they are all primary reinforcers (they elicit a
biological/physiological response that is immediate, certain to occur, and
generally perceived as positive by the participant - the most powerful
type of reinforcer). Most laboratory research that has been done with
animals has used a primary reinforcer to reinforce behavior. For example,
rats are often kept in a state of hunger (80% of free feeding weight) for
purposes of enhancing the reinforcing effect that the food pellets will
have.
These addictive activities also tend to be powerfully reinforced by other
sources as well, for example, the social aspects of cigarette smoking and
alcohol consumption are also present. In the United States, where smoking
in public places is generally prohibited, a cigarette break often means
that you get to take a break from whatever it is that you are doing and go
hang out for a while with the friends that you have made while smoking,
understandably perceived as positive in the workplace. The water cooler
has been replaced in many organizations by the designated smoking areas.
Alcohol and drugs also have the added effect of allowing one to escape the
pressures/stress that they may feel in other areas of their lives.
Obviously, over-eating has the pull of immediate gratification provided by
the taste of the food (I have never heard of someone who over-ate
something that they didn't like), which will pull us well beyond the
amount of food necessary to sustain our health, or even reduce the hunger
pangs that we feel (an antecedent condition that prompts us when to eat).
Changing these behaviors is often very difficult, even for the facilitator
that understands behavioral principles. As you observed, often times our
behavior will oscillate between alternative consequences as we struggle
with choosing a path (witness the many rehabs and relapses that
individuals will go through as they struggle with various addictions).
Some characteristics of the contexts in which these behaviors are engaged
may also become powerful antecedents for the occurrence of the behavior.
For example, drinking a cup of coffee may cause a person to "yearn" for a
cigarette decades after they have quit the habit (if they often had a
cigarette on their coffee break). Recovering alcoholics and drug addicts
often have to completely change their hangouts and friends in order to
keep themselves out of situations where they may be tempted to have a
drink or take drugs.
These social reinforcers and the social context have a great impact on our
behavior in many cases. One of the preeminent examples given to
demonstrate the importance of social reinforcement in the work place is
the Hawthorne Effect. In this study, the original intent was to examine
the effect that physical characteristics of the work environment, such as
light levels, would have on productivity. They found that their
experimental results were profoundly influenced by a confounding variable
which they interpreted to be the attention paid to the experimental group
by management. The identification of this "psychological" factor to which
the experimental outcome was attributed is commonly given as the seminal
event in the organizational sciences fields. Interestingly, reexamining
the results of the Hawthorne study from a behaviorist perspective suggests
that the common understanding of the Hawthorne effect is actually not
quite correct, and the paths that have been taken following may have done
much more to cloud our understanding of motivation than help it (see H. M.
Parsons (1974). What happened at Hawthorne? Science, volume 183).
In LO20363, John Gunkler talked about the "balance of consequences". We
may change the balance of consequences by taking any one of the steps
which Leo mentions, and we may be able to combine several interventions
targeted at different areas where this balance may be altered with some
success (this is what we will do to "build our bridge").
For example, we may try to make the sustaining consequence of the
undesired behavior less attractive (for example, having an overdose may be
enough to scare some drug addicts straight - usually for a limited period
of time - in general, anything done in excess will satiate someone's
desire to engage in the activity for a period of time). Many of the aids
available to facilitate the person who desires to quit cigarette smoking
take an approach along these lines. For example, nicorette (the nicotine
gum) and the nicotine patches that are on the market attempt to reduce the
desire to smoke by reducing the physiological craving for nicotine. This
may allow someone to change some of the other social habits and give them
the opportunity to connect with some other positive consequences that
result when not smoking (reduced coughing, better breath, white teeth,
increased aerobic capacity). They may be able to establish a new routine
to replace the one that they followed which involved smoking. The new
inhaler on the market even allows them to mimic the smoking motion. [My
opinion is that all of these are of some value when paired with other
things. The largest potential downfall with all of them that I see is
they don't address the physiological response to nicotine, and this must
be addressed at some point. Reducing the amount of nicotine begins down
this path, but with some people, this may serve to keep them in touch with
the dependence and "whet their appetite" for the vice.]
Attaching punishment to the behavior is another way to discourage the
occurrence of a behavior - and again, it will usually depress the rate of
the behavior for a while. Restricting access to the reinforcer (such as
having an inpatient drug rehab program where the environment is controlled
is an example) is another possible approach. However, if this type of
activity is the only approach to "changing" behavior (getting one behavior
to stop, but not taking any action to replace it with another behavior),
the effects are rarely permanent. Often, the performer will be working
diligently to find ways around the barriers that we are putting up,
behavior that will ultimately be reinforced by the consequence that have
always reinforced the behavior (obviously considered as positive by the
performer). The general rule (actually a hyperbole) is that for every
barrier that we erect, there are 1000 ways to behaviorally bypass it. If
you want to observe the extent to which any person can be creative, simply
try to use these techniques and then observe the solutions that will be
found to get around them.
When concocting our behavioral "cocktail", there are some possible
complications and pitfalls that need to be considered. For example, you
mentioned combining the pushing force of negative reinforcement with the
pulling/attracting force of positive reinforcement as one way to possible
speed up the flow of behavior. In my experience, this is rarely the case.
Usually, the general characteristics of one paradigm manifest themselves
over the other. Specifically, several studies seem to indicate that the
general ratio of 4:1 (positive reinforcers:punishers) or greater before
the positive reinforcers begin to "win out" over the negative reinforcers
(compliance in order to avoid the punishment). In other words, if both
negative reinforcers and positive reinforcers simultaneously exist for the
same behavior, the characteristics of the negative reinforcement dominate
(doing just enough of the behavior to avoid/escape the negative outcome)
until the positive reinforcer is 4 times more probable to result. As the
general area of the 4:1 ratio begins to be exceeded, discretionary
behavior begins to be exhibited (rates of performance higher than what is
required to avoid/escape the negative outcome). Having said this, there
are some times when we may need to use negative reinforcement to get a
behavior to occur. However, if we choose to do this, we should have our
positive reinforcement plan ready to pick up/take over the reinforcement
of the behavior immediately should it occur.
Using positive reinforcement to influence behavior requires a much higher
degree of rigor to implement effectively than most popular management
books would indicate. Using positive reinforcement is much more than the
Management by Walking Around (MBWA) suggested by Tom Peters, and goes far
beyond the Theory Y concepts proposed by MacGregor. It requires much more
than being a nice guy, trying to take some time to "catch people in the
act" of doing something good/right on the occasion that we step outside of
our offices, or a "way to go!" on a job well done. Even the vast majority
of our reward programs or incentive programs (including gainsharing and
profit sharing plans) fall well short of meeting the criteria for
effective positive reinforcement. Not that these things are necessarily
bad things. They may fit very well with effective use of positive
reinforcement. However, as I'm sure many of you will testify, they just
fall far short of having the effect that we expect/hope that they will.
Our common sense and experiences rarely provide us with enough of an
understanding to enable us to do this effectively.
[A side note: Theory X, which MacGregor proposes as Theory Y's dialectical
dual, corresponds closely with negative reinforcement. I believe that
MacGregor was picking up the vastly different characteristics of behavior
"pushed" by negative reinforcement and behavior "pulled" by positive
reinforcement when he articulated Theory X and Theory Y, but he didn't
quite get his fingers on the pulse. Given that this is the case, I would
agree with the assertion of Richard Goodale in LO20632 that Theory X and
Theory Y approaches work equally well, and the Theory X is the ascendent
approach used in the business world today. The point that I would come
back to, the very point that I have been trying to make all along, is
essentially that Theory Y, MBWA, attaboys/girls, high fives, etc... do NOT
meet the criteria of positive reinforcement. It is simply nice behavior,
and it may, or may not (much more likely) have the effects that are so
often promised. I often get on a soapbox for this reason when people
begin talking about how reinforcement doesn't work, or claiming that using
positive reinforcement will damage a person's "intrinsic motivation".
Alfie Kohn, for example, has made a name for himself in identifying that
rewards are often grossly misapplied, and as a result have the effect
exactly opposite of that which was intended. However, he draws
conclusions and makes suggestions that I am staunchly opposed to in light
of what behavioral psychology has been able to demonstrate. For more on
this, please see LO18449 and LO19896.]
Aubrey Daniels, in his book Bringing Out the Best in People (an excellent
starting place for those interested in understanding more about the
application of behavioral principles - he also talks about the ABC's of
behavior, as I have. John Gunkler identified a source for the ABC model
off-list that goes as far back as the '70's), identifies 4 common errors,
or misapplications of positive reinforcement. They are the following:
1)The perception error - we attempt to reinforce someone else by using an
incentive that they don't find reinforcing at all (this is often something
that WE would find reinforcing, so we assume that it will be for them
too). It is important to note that we can usually know almost immediately
if our attempt to reinforce was successful - if it was, we should be
seeing more of the behavior that we reinforced. Many times, we will think
that we reinforced someone, but, if we watched their behavior, we will see
that nothing changes. In this case, we may have REWARDED them, but we
didn't reinforce them. In other cases, the rate of their behavior may
actually go down, indicating that we actually punished them (example might
be publically recognizing the efforts of a shy person. They may find this
to be mortifying and extremely punishing, resulting in their efforts to
avoid further public acknowledgement). We must continue to watch what
happens afterwards to know how our attempts to reinforce are received.
2)The contingency error - in order for our efforts to reinforce to be
successful, the outcome which we provide must be contingent upon behavior.
In other words, the reinforcer must only be available when the behavior
occurs - it only results if and only if the behavior occurs. In many
cases, the "reinforcer" is available regardless of whether or not the
behavior happens. This frequently is the case when we provide a reward
for achieving an outcome (rather than the behaviors that produced the
outcome). The outcome may have resulted because of dumb luck. In other
situations, the outcome was realized by the efforts of some, but in spite
of others. If we provide a reward to the whole group involved in the
effort, we will end up punishing the efforts of those who carried the load
of the work and reinforcing those who were actually barriers to success.
This is one of the major downfalls of efforts to reinforce teams and
efforts to reinforce through the use of year end bonuses and success
gainsharing.
3)The immediacy error - essentially, in order for a reinforcer to be
effective, it should be provided as close in time to the occurrence of the
behavior that it is intended to reinforce as possible. Immediately is
best. Otherwise, the reinforcer falls prey to the decay of delay that I
have mentioned in previous posts (see LO19894 and LO20081). Worse yet, we
may end up "accidentally" reinforcing some other behavior. This may
result in what behaviorist have defined as "superstitious" behavior. This
explains why baseball players may engage in behaviors that may even be
considered bizarre in order to avoid jeopardizing a streak of good
performance. And it also forms the basis for what I like to refer to as
"the corporate raindance" - we don't know exactly what we did that
produced the outcome that we wanted, so, rather than take the chance of
leaving out the one piece that may have actually produced the outcome, we
do everything over again indiscriminately.
4)The frequency error - the bottom line here is that one reinforcer won't
result in the acquisition of any behavior. This discussion gets back to
the schedules of reinforcement that John Gunkler eluded to in LO20363. As
a general rule, it takes frequent reinforcement (ideally continuous
reinforcement) to facilitate the acquisition of a behavior, but the
behavior may be maintained with far less frequent reinforcement. For
example, the reinforcement schedules of various slot machines generally
follow these rules. Rumor has it that the machines at the end of the row
have a higher pay off rate than those in the middle in order to get you
"hooked", and the quarter slots have a higher pay of rate than the dollar
slots... in order to shape your slot machine playing behavior. BTW, no
matter what slot machine you play, if you play long enough, you will
eventually lose all of your money. Another example that may be of
interest, most video games provide positive reinforcers at the rate of 100
per minute!
At has encouraged (for example in LO20330), as have I on numerous
occasions, that we are best off if we use positive reinforcement to guide
behavior. Without positive reinforcement, spontaneous and irreversible
self-organization is not possible. Methods for doing this have become
fairly well developed for frequent and routine behaviors (such as can be
found in many manufacturing environments). In the late 60's, the 70's,
and the early 80's, there were numerous behaviorists who made an impact on
organizations (they were known as "performance technologists"). If you
are interested in some of their work I would recommend digging in to Tom
Gilbert, Dale Brethower, and Robert Mager and Peter Pipe, who were some of
the first to really try and apply this perspective to the work place. More
recently, Geary Rummler and Alan Brache, and Aubrey Daniels have written
books that can be located fairly easily. The use of positive
reinforcement to change behaviors related to safety has also been a fairly
hot topic in industry for the past several years, spawning some writing
that might be of interest (Tom Krause and Terry McSween are two authors
who might be worth checking out). In the area of compensation, Bill
Abernathy has written an easy-reading book called the Sin of Wages that is
worth a look.
Much of this work has been pigeon-holed into fairly small compartments of
application, such as the frequent and routine tasks of manufacturing (much
as Deming's ideas have largely been pigeon-holed into a fairly small area
of application). There has been some debate over how useful the
behaviorist perspective is in more complex environments. I think the
discussion on the list over the past couple of months around the
behaviorist perspective, and the ties that have been made to entropy
production and emergence, show the utility of adopting this perspective
even in much more complex environments. I am not asserting that adopting
a behaviorist perspective will make changing a complex system simple. I
am saying, however, that adopting a behaviorist perspective will make
complex systems, especially of behavior, a lot less complicated to
understand. It provides a framework (or theory in the sense that Deming
used) for learning as we observe our behavioral environments. As Deming
has said, experience without theory teaches us nothing, so this provides a
critical piece to our learning puzzle.
I really appreciated At's clarification of the difference that he
perceives between complicated and complex (LO20616). He wrote the
following:
>Complexity (for me) is the consistent and coherent organisation of
>innumerous different entities. When this consistency and coherence
>of complexity gets destroyed, the result is a complication. We cannot
>avoid the complexity of reality, but we can certainly avoid to add
>to the complications of reality. Reality is complex, but it should not
>be complicated also. ...snip...
The behaviorist perspective allows one to see and understand the complex
interrelationships between multiple entities and identify levers for
change. It provides a lens for recognizing the effects of our
interactions on others. It enables us to understand the dynamics of our
organization's culture and psychological climate, and provides us with a
method for changing it.
Recently, I had an off-list dialogue with Kent Quisel (another member of
the learning-org list) in which we discussed the application of the ABC
model as an "unbiased", or culturally paradigm-free, lens for
understanding organizational responses to dynamics of larger systems in
which the organization operates (such as their market place, and in
interactions with their customers). Depending on the structure of their
environment, organizations may encounter differing challenges that may
influence how they organize themselves and the dynamics of the
interactions within the organization. The ABC model will work in harmony
with many of the system-dynamics models and aid in understanding these
dynamics.
For example, the time horizons of many organizations are driven by the
competitive pressures of their markets (a negative reinforcement feedback
loop), the structure and nature of their relationships with their
customers, and the demands placed on them by other stakeholders in the
organization. Given that this is the case, the behaviorist
perspective/ABC model is extremely effective when used within a larger
systems perspective. By using the ABC lens, we can examine the feedback
loops that result from the interactions with each of these areas,
understanding which are positive reinforcement/amplifying loops, which are
negative reinforcement loops, and which are balancing loops. These loops
form the basis for determining the criteria by which organizational
effectiveness can be assessed.
The behaviorist term for these criteria is "metacontingencies". These
organizational level metacontingencies have many of the same
characteristics that individual behavioral contingencies have, but they
operate at a higher, more complex, level of organization. The extent to
which these feedback loops are considered and connected to more micro
level behaviors will determine the extent to which they can influence all
kinds of organizational behaviors. Just as the outcomes to an individual
behavior have certain characteristics that determine the effectiveness of
the outcome as a reinforcer, these feedback loops for metacontingencies
have similar characteristics. The more immediate the organizational
outcome, the more greatly it will influence the self-organization of the
organization. Whether it is perceived as positive or negative will also
play a large role in what kind of organizational behavior will be
"motivated" by the feedback (how we respond to impending organizational
doom may look very different than responses to impending organizational
success). Finally, the certainty of the outcome to result given an
organization's behavior, the more greatly it will factor as a
"meta-reinforcer".
To illustrate, an organization that is responding to a negative
reinforcement paradigm that is set up by urgent customer demands (such as
working to avoid missing an urgent deadline) will organize itself very
differently and have very different internal dynamics than a company that
has longer time horizons, or is insulated from their environments (like
governmental bureaucracies in many instances). A publically traded
company, "driven" by the stock market to post consistently quarterly
earnings and pay dividends, will often choose a very different path of
capital investment than a privately held companies that have owners
interested in building the company over a lifetime (one reason that Warren
Buffett has said that he invests so heavily in companies that aren't
traded on any stock exchange).
A company in a mature, or declining industry is much more likely to become
dominated by negative-reinforcement feedback loops tied to loss of market
share, erosion of margins and other accompanying outcomes. In such
situations, the environment provides an abundance of negative
reinforcement pressures and the organization does not need to further
aggravate the situation by creating their own negative reinforcement
pressures in response. In contrast, a company in a market that is
exploding will often work itself to the point of exhaustion and feel
energized (those in the organization find the work extremely positively
reinforcing).
The political behaviors that dominate many an organization are most likely
to escalate in situations where the organizational effectiveness
metacontingencies are either not fed back to those within the
organization, or they are, but the individual feels disconnected from
them. When the external consequences are not strong enough or immediate
enough to pull/push behaviors linked to satisfying these contingencies,
the social reinforcers that are administered within the organization will
begin to pull or push behaviors (usually push) instead. The organization
may deteriorate into one where CYA is the rule and behaviors tied to
building and maintaining fiefdoms, rather than on activities related to
organizational effectiveness, dominate the organizational landscape.
Without any understanding, the mental models that will result will become
progressively more pessimistic and caustic (see the power and influence
discussion in LO19894).
I say all of this in the hopes that I can demonstrate that the ABC
model/behaviorist perspective provides a means for relating patterns of
organizational behavior and individual interactions to the structure of
the environment in which the organization "lives", and a method for
changing behaviors so that the alignment of individual behavior with these
organizational metacontingencies becomes stronger, even in complex
environments that have typically been considered outside the reach of
behaviorist methods. Drawing these linkages is the basis on which shared
vision can be built within the organization, and can move (pull, not push)
an organization along the path to changing the paradigm motivating
behavior from one of negative reinforcement to one of positive
reinforcement. I believe that the behaviorist perspective is scalar in
nature, and can be used to understand dynamics regardless of the level of
consideration.
Leo, I hope that this helps you to understand how the behaviorist
perspective can be applied. I apologize for the delay in the response,
and I hope that your interest hasn't decayed too much over time. :-)
I hope that you are doing well.
Jon Krispin
--"Jon Krispin" <jkrispin@prestolitewire.com>
Host's Note: In association with Amazon.com, these links...
Bringing Out the Best in People : How to Apply the Astonishing Power of Positive Reinforcement by Aubrey C. Daniels http://www.amazon.com/exec/obidos/ASIN/0070153582/learningorg The Sin of Wages : Where the Conventional Pay System Has Led Us & How to Find a Way Out by William B. Abernathy http://www.amazon.com/exec/obidos/ASIN/0965527603/learningorg ...Rick]
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>