Changing another person LO19894

Jon Krispin (
Tue, 17 Nov 1998 14:14:47 -0500

Replying to LO19869 --


Your point that we are ourselves changed in our interactions and attempts
to influence and change others is a crucial one. This is an area of which
we must constantly be cognizant. It is also an area that has been the
subject of research in the field of social and organizational psychology,
and I thought some of the conclusions might be worth sharing.

One of the professors with whom I worked in graduate school was David
Kipnis (Temple University - he officially retired several years ago,
although he stuck around long enough to be on my dissertation committee).
He spent much of his career studying the effects of attempts to exercise
power and influence on the "powerholder", or influencer, as opposed to the
target of the influence attempt. He conducted many studies which
suggested that there were indeed "reliable" changes that occurred in the
attitudes, attributions, and even in the subsequent behaviors (influence
attempts) of the influencer as a result of the type of influence tactic
used. He termed them metamorphic effects of power.

In Kipnis' model, the selection of the influence tactic used was assumed
to be the result of cost-benefit analysis (more or less formal, depending
on the circumstances) on the part of the influencer. The tactic selected,
it is assumed, is believed to be sufficient to have the desired effect
(behavioral change on the part of the target) while carrying the least
cost (in terms of building resistance that will make subsequent influence
attempts - if there are to be any - difficult and costly). As the
expectancy of successful influence declines, the influence tactic selected
becomes stronger (usually more aversive to the target, but not always -
more on this below). This is especially true if the target is assumed to
be able to perform but hasn't (usually attributed to lack of motivation as
opposed to lack of ability). Now for the real kicker, as the strength of
the influence tactic increased, the influencer became more inclined to
devalue any good performance, more likely to rate the target less
favorably, and even express an increased desire to distance themselves
from the target. The change in behavior was attributed to the tactic
selected, with the influencer as the causal agent, not any personal
characteristic of the target.

Kipnis originally defined the exercise of power to be an attempt to get
the target to do what the influencer wanted when they otherwise wouldn't.
Obviously, this type of influence is wrought with problems (although
certainly not out of the bounds of what happens in many organizations on a
daily basis), and clearly falls outside of the influence attempts that you
have defined as acceptable (at least for yourself - I, by the way,
subscribe to a similar guideline), that being that the person or group
with whom we are working has granted permission to participate in the
change effort.

However, we are not off the hook yet. Kipnis extended his research to
situations where the target was aware and permissive of the influence
attempts and even amenable to them (one example of this would be in
counseling situations where the client has approached the counselor
seeking assistance), although Kipnis spent most of his time focused on
organizational settings). He did additional work to show that the use of
more effective methods of behavioral change (such as behavioral
technologies - other researchers have examined SPC monitoring and process
improvement techniques in this light and drawn similar conclusions) may
result in similar negative attributions - although not necessarily.
Kipnis' conclusion was that influence tactics should be limited to those
which leave choice as to how to respond in the control of the target if
the relationship with the target is to be even maintained, let alone

Rick, I don't know whether or not you are aware of this, or another, line
of research that has lead to your caution when it comes to working with
others as a change agent, but obviously you have encountered something
along the way which has raised your level of sensitivity to this issue
(maybe your own experience, or vicariously as you have observed others).
You have good reason to be cautious.

I am going to shift gears for a second, but I promise that I will come
back to the discussion above. I would also like to talk about the
application of the behavioral principles uncovered by Skinner and others
within an organizational context (originally brought up in this thread by
John Gunkler. As you may surmise from my comments above, I do not count
myself as a radical Skinnerian. I will admit, however, that I am a firm
believer that these principles are absolutely fundamental to understanding
how and why people do what they do, and what can be done to change things.
Behavioral psychology is much more than an amoral technology for changing
behavior (although it is certainly a technology for changing behavior, and
a very powerful one at that. I also believe that it can be used
immorally.). It has actually played a very large piece in my own thinking
and life practices about the role that I play as a change agent, as a
husband and father, and as a friend. Behavioral psychology has actually
been widely applied in organizational settings and with great success.
Typical attempts to construct interventions in organizations are pleased
with effect sizes in the .3-.5 range. Behavioral interventions more
typically see effect sizes in the range of .85-.9 and can have an greater
impact of an order of magnitude. These effects have been demonstrated to
last over time (some longitudinal studies have documented sustained
improvement from baseline for 10 years and more). This might be classified
as a "reliable" impact on behavior, for better or worse. There are a
number of academic journals that document this type of intervention if you
would like to research some of the history for yourself (see the Journal
of Organizational Behavior Management and the Journal of Applied Behavior

First, a very brief primer on the behaviorist perspective. It can be
summarized using what is now commonly know as the ABC model of behavior:

The "A" stands for antecedents to behavior. These are any and all
conditions and events that occur prior to a behavior (may be as simple as
an environmental condition such as cold weather preceding the donning of
warm clothes or a dark room preceding the flipping of the light switch, or
as complex as organizational training and seminars providing guidance and
direction for a large organizational improvement intervention). These
prior events and conditions provide information that can influence the
choice of behavior in a given context. Antecedents may get someone to try
a behavior, but, on their own, they do very little to "motivate" the
behavior to continue once it has been tried.

The "B" stands for the behavior itself. The specific behavior as a topic
of discussion is usually defined and clarified again by a particular group
or context. For example, leadership may the topic of discussion.
Obviously, there is quite a bit of debate as to what constitutes
"leadership", and it is made up of a multitude of individual behaviors, so
some dialogue and discussion is usually required to come to a level of
consensus as to what it means to the particular group of people involved
in the discussion.

The "C" stands for the consequences that follow the behavior. These may
be results that are produced as by the action (intentionally or
unintentionally), but they may also include events that followed the
behavior that may have had nothing to do with the behavior itself (an
example of this would be superstitious routines followed by baseball
players in an effort to not screw up some streak that they are on. The
fact that they were wearing a given pair of socks may have had nothing to
do with their 5 for 5 performance on a given night, but they won't take
the chance of messing with anything, so the socks become "lucky" socks).

The consequences following the behavior are what will ultimately determine
whether or not the behavior will continue as before, strengthen in terms
of the rate or likelihood that it will occur again in the future, or
weaken. In point of fact, there are consequences for every behavior
(whether or not they happen by design is another question), there are
always immediate consequences that result from the behavior, and there are
often future consequences that result from the behavior. As you might
guess, the immediate consequences have a far greater degree of influence
on a behavior than the future consequences. In fact, the rate of decay
for the degree of influence is not just linear in nature, it is

In the behaviorist perspective, the most time is spent considering the
consequences that are in place for a given behavior, and then designing
interventions that can have the effect of changing behavior by focusing on
feedback systems to make particular consequences more (or less) salient to
the performer, and providing reinforcement to strengthen desired

Without getting too lengthy (maybe I already have), consequences can be
classified into 2 mutually exclusive categories - those that increase the
likelihood that the behavior will occur again, and those that decrease the
likelihood. Consequences that increase the likelihood that a behavior
will occur again are know as reinforcers. Those that decrease the
likelihood that a behavior will occur again are known officially as "those
consequences that decrease the likelihood that a behavior will occur
again" ;-).

For the purposes of this discussion (and, believe it or not, for the sake
of brevity) I am going to ignore the 2nd category and only talk about the
first category - reinforcers. There are 2 general ways that consequences
can increase the likelihood that a behavior will occur. The first,
positive reinforcement, results when a behavior is followed by a desired
outcome (from the perspective of the actor). When this occurs, the actor
is more likely to try the behavior again in a similar context. The
second, known as negative reinforcement (this is most definitely not the
same as punishment), occurs when a behavior is followed by the *removal or
reduction* of a negatively perceived condition. Negative reinforcement is
also commonly known as escape or avoidance behavior. While punishment
reduces the likelihood that the behavior will recur, negative
reinforcement increases this likelihood. To give an illustration, we take
Tylenol (or some other pain killer) when we have a headache. The Tylenol
*removes or lessens* the headache (the negatively perceived condition),
so, the next time we have a headache (an antecedent condition), we are
MORE likely to take Tylenol again.

In most organizational contexts, negative reinforcement paradigms are
prevalent. Examples of these include customer deadlines (we will work
like crazy to avoid missing the deadline) and many management techniques
(for example, the work of David Kipnis which I mentioned above is
essentially the study of attributions made by influencers following the
use of negative reinforcement techniques). Certainly, most management
techniques are not intended as negative reinforcers, but, given the
feedback systems that are in place (such as the financial performance
information) and the complexity inherent in most organizations, many
managers end up, by default, acting on the negative exceptions and ignore
the positive exceptions. This can influence the tactics that they select
to "change" the performer (see also above) and a vicious downward spiral
can begin.

As if that isn't enough, attempts to "crank down the screws" usually do
get more of the behavior desired by the influencer (and it gets it NOW)
and this ends up positively reinforcing the influencer (they get the
outcome that they want, so they are MORE likely to use the same approach
in the future).

You may ask (and I have been asked) - if negative reinforcement gets more
of the behavior that I want, what's the problem with using it? Now we are
finally to the topic initially at hand - changing a person and the ethical
implications. While negative reinforcers will get more of a given
behavior, there are several major problems. The first is the cycle of
attributions that is set up and the adversarial nature of the
relationships that get established when this type of reinforcement is
predominant. As I mentioned above, the behavior that is influenced by
negative reinforcement is known as an escape and avoidance behavior. This
has several implications. One is that, if the aversive condition is
strong enough, the actor may choose to exit the situation rather than take
action within the context to escape the condition or avoid it in the
future. Second, those that choose to stay usually become more and more
resistant to this type of method of influence, and, they may even reach
the point of open aggression against (witness the violent nature of some
labor-management disputes, even in recent history). The last one that I
will mention is that, while more of the desired behavior will result, the
ceiling for performance is fundamentally limited. The most performance
that can be obtained is the least required to avoid the negative outcome.
Obviously, if continual improvement is to be sustained, this isn't enough.

Only positive reinforcement can truly maximize performance. It is also
the method that "produces" ownership, pride in work, and empowerment while
driving fear out of an organization (how's that for promising the world?).
However, this doesn't happen without a plan. Often, the very behaviors
that are necessary to ultimately sustain continual improvement and
maximize performance are met with immediate punishers (for example, the
learning disciplines take much more effort to implement than not and more
effort is punishing) and the positive outcomes that they may
ultimatelyrealized are far removed in space and time from the uptake of
the new behaviors (see system dynamics for more discussion around this).
The end result is that many of these behaviors which we may prompt in our
training (an antecedent) will starve to death when they try and stand on
their own (they are met with a lack of positive reinforcers and will
eventually die out. This process is known as extinction).

Our job? We need to do our best to create feedback systems and an
infrastructure that will provide positive reinforcement immediately to
encourage the uptake of the behaviors necessary for sustaining continual
improvement and learning. Simply providing instruction, while absolutely
necessary, is not sufficient for the transition. This may be as simple as
creating a group of people that will encourage each other, and it may be
as complex as completely redesigning the compensation, appraisal, and
knowledge management systems in a multibillion dollar, multinational
conglomerate. This must be done at least for a period long enough that
the performer is ultimately connected with the delayed but real
improvement to the new system of interacting. Once these consequences are
experienced, they may be enough to reinforce the behavior and sustain it
over time.

It is possible that all of this can be used for the wrong ends. A person
who is fluent in these concepts may be a master con artist who can take
you for everything that you have and leave you feeling pretty good about
it (this was my father's assessment of the Disney organization when he
visited the Orlando complex for the first time), but it is also the only
way that any organization is going to attain and fulfill its shared
vision. If the end is the benifit of an individual, the use of these
techniques might be classified as manipulation. However, if this approach
is used to attain a shared vision, there is nothing in the approach that
you wouldn't want everyone in the organization to be aware of and even use
on you. The end never justifies the means, but it is also true that the
end will never be attained if the means aren't correct.

Any attempt to create an organization is ultimately limited by the extent
to which these concepts are considered and addressed. Actually an attempt
at organization is trying to do this anyway, we might as well do it with
some intention and understanding.


Jon Krispin


"Jon Krispin" <>

Learning-org -- Hosted by Rick Karash <> Public Dialog on Learning Organizations -- <>