APPLIED BEHAVIOUR ANAYLYSIS: SHAPING
Shaping is a strategy used to develop
certain target behaviour that an individual do not exhibit occasionally and
encourage a behaviour to exist. Shaping
is defined as the differential reinforcement of successive approximation of a target
behaviour until the person exhibits the target behaviour (Miltenberger,
2008). Shaping is the process of steps being introduced in each level, starting
from a starting behaviour that the subject exhibit sometimes, and reinforcing
the behaviour, and stop reinforcing that behaviour, leading to extinction of a behaviour,
and development of a new behaviour that leads on towards the target behaviour.
The process is repeated until the individual exhibits the desired target
behaviour. However, this strategy is also susceptible to some problems and
produces undesirable results if it is not carried out precisely or if the
target behaviour is not best suited to this method. To best describe the
application and misapplication, two cases will be presented to show contrast
between application and misapplication of shaping.
Application
We are presented with a scenario; imagine
you live in a house with a backyard. The door to the backyard is in your family
room. You let your dog Felix out into the backyard a few times a day. You
decide you would like to teach Felix bump the back door with his nose before
you let him outside. Currently, whenever Felix wants to go outside, he walks
around the family room and often walks past the back door. In this scenario,
shaping strategy is suitable to be carried out. The guideline to execute a
shaping strategy requires us to, define the target behaviour, determine whether
shaping is the most appropriate procedure, identifying the starting behaviour,
choosing the shaping steps, choosing the reinforcer, differently reinforce each
successive approximations, and moving through the shaping steps at a proper
pace. Now, we determine that the target behaviour is for Felix to bump his nose
on the back door before being allowed out into the backyard. One of the most
important factors is determining the starting behaviour of Felix. In this
scenario, Felix is said to walk around the family room and often walks past the
back door. That can be determined as the starting behaviour of Felix prior to letting
him out. Despite seeing Felix pacing about the family room and walking past the
back door, we would like Felix to be specific about getting out into the yard,
rather than just assuming that it wants to go out. Now that we have the
important element of starting behaviour and target behaviour identified, we
must now select a reinforcer to act as a reward each time a successive
approximation is achieved by Felix. The use of a reinforcer is important as we
want to encourage a desirable behaviour to increase in frequency. For this
situation, the reinforcer could be a dog treat that Felix favours and likes.
The reinforcer will be immediately applied when Felix behaves according to cue.
The reinforcer of conditioned respond item must be administered a second
interval of each respond. For example, if a clicker is used in this shaping
process, a successive approximation followed by a click, and immediately fed
with a treat. The conditioned response could also be in the form of verbal cue,
such as “Felix, yard?” Depending on the wanted target behaviour, any
conditioned response is open to the trainer. In this scenario, verbal cue would
be the suitable natural reinforcer for little Felix. Now, I shall determine the
steps for the successive behaviour to occur. Although, the steps may be altered
through the shaping process, it does not necessarily have to be the exact
step-by-step process. What matters most is that each successive approximation
must be closed gap and the extinction of the previous behaviour must take place
because my goal is to have Felix display the target behaviour, which is bumping
its nose on the back door. The starting point for me should be, every time
Felix walks about in the family room, I shall say “Felix, yard?” and give him a
treat. The successive approximation would be Felix would be given a treat only
when he approaches the back door. I would want the behaviour of walking about
the room to be extinct and anyway other behaviour that follows the cue “Felix,
yard?” would not be rewarded. Perhaps, now I would stand near the back door
that leads to the yard. The process may take a while and continuously repeated
till Felix comes straight to the back door every time “Felix, yard?” is called
out. Treats are given immediately after a successful action is performed, right
after natural reinforcer is spoken. Now that Felix knows to approach the back
door, the goal of getting to its target behaviour is closer. The next
successive approximation would be “Felix, yard?” on encourage bumping his nose
to the door. This could be done by putting my hand with the treat at the door.
This step is repeated several times till Felix is used to it. Occasionally, I
put my hand on the door but no treat in it and immediately reward the behaviour
after it bumped its nose on the door. The next successful approximation would
be when hand gesture is removed and Felix is rewarded every time it bumps its
nose on the door. The final successful approximation would be every bump on the
door, the door is opened by me and Felix is rewarded. Throughout the whole
process, the criteria for each successful approximation may change to suit
Felix immediate behaviour as it may not be as straight forward and easy but,
what is important is, that the target behaviour is achieved and behaviours that
leads to each successful approximation be extinct. By the end, of the shaping
process, Felix will now respond to the natural reinforcer, “Felix, yard?” by
going straight to the door and bumps its nose on it.
Misapplication
We saw how effective shaping strategies can
be successfully implemented in the previous scenario. However, shaping may also
be maladaptive if it is implemented incorrectly to and may lead to negative
feedback. Here’s a scenario to see how shaping can be misapplied:
Dr. Williams, a school psychologist, was
working with an extremely socially withdrawn adolescent, Jenny. Dr. Williams decided
to use shaping to help Jenny develop social skills. He identified the target behaviour
as making eye contact, smiling, standing up straight, talking at a normal voice
volume, and nodding and paraphrasing when the other person said something. Dr.
Williams was going to reinforce successive approximations of this target behaviour
in therapy sessions, in which he played the role of a classmate and engaged in
conversation with Jenny. In each session, Dr. Williams and Jenny role-played
four or five short conversations. Before each role-play session, Dr. Williams
reminded Jenny which behaviours she should work on. As a reinforcer for
exhibiting the correct behaviour in the role play sessions, Dr. Williams bought
Jenny an ice cream cone in the school cafeteria once a week.
In this scenario, the shaping technique
cannot be used successfully. Why? It is because under the shaping guidelines,
is the starting behaviour. Jenny is naturally withdrawn and do not make any
social gestures that suggest her effort to develop social skills. Shaping
requires that the starting behaviour must already be occurring at least occasionally
but if no particular target behaviour is exhibited at all, other strategies is
required to generate such behaviour. In addition, Dr. William seeks too many target
behaviours at one go and this will create too many criterion which would make
it difficult for Jenny and confuses her in the process. The approximation
successions may also have too long intervals in between the reinforcement which
is the ice cream which is once a week. The long duration gap between reinforcer
and successive approximation would cause the behaviour to extinct in that
period. Another reason why shaping would be a problem is, whenever Dr. Williams
reminded Jenny on behaviours that she needs to work on, it might create a
negative reinforcement as shown in Skinner’s operant conditioning. A positive
reinforcement would be getting ice cream and negative reinforcement would be
acting out target behaviour to avoid social discomfort. Shaping strategy requires
the person to willingly execute behaviours that leads to successful
approximation and finally, to the target behaviour. In this particular
scenario, strategies like prompting or chaining would be better suited.
In conclusion, shaping techniques can be
beneficial and improve the lives of people. When used correctly, it will yield
good benefits. Such as, in the case where Homer (1971) worked with Dennis, a 5
year old child with intellectual disability and had a spinal condition called spina bifida, in which the spinal cord
is damaged before birth. Unable to walk, Dennis could only crawl. With two
shaping procedures, Dennis learned how to use crutches successfully and made
him independent and his quality of life was improved. We must also be wary that
misapplication will create an undesired behaviour or even worse still, shaping
problematic behaviours.