intro
There are four types of behavior: reflexes, habits, skills and voluntary behavior. Reflexes are elicited by environmental stimuli. The best technique for changing them is systematic desensitization. Habits and skills require automating behaviors until they occur without thought. Voluntary behavior is the focus of behavioral change.
Here is how you change behavior.
Select a Target
1. Yours, not others
It is extremely difficult to change the behavior of others. It is almost always best to change your behavior. These principles will work to change other people or other animals, but when learning the steps (and in general) start with yourself. You don’t understand or control their reward system. What you think is rewarding might be punishing for them.
2. Positive, not negative
Change negative goals into positive ones. “I don’t want to be so out of shape” can be converted to “I want to fit into smaller pants.” If you can’t find something positive to do, you haven’t sufficiently defined the problem.
3. Increase, not decrease
This is a good test of positive-ness. Don’t try to lose weight or stop smoking. Find an incompatible behavior to increase. If you can’t find something to increase, you haven’t sufficiently defined the problem.
4. Pick one behavior, not everything
Instead of “I’ve going to stop smoking, eat less junk food and be healthier” go for “I going to go to the gym more.”
5. Clear, not vague
Instead of “I want to build strength and endurance,” state it more clearly: “I want use the rowing machine more.”
6. Continuous scale, not discrete
Instead of “I rowed today,” which is categorical (yes or no), use a continuous measure of your activity. “I rowed 5000 meters” would be better.
Tracking
Now that you have defined WHAT, it’s time to figure out where you are. The next cluster of steps focuses on the situation within which the behavior will occur. You will want to use the encoding specificity principle to match training conditions to goal conditions.
7. Identify Antecedents
The importance of what happens before a behavior is underestimated. Consider your client’s beliefs and prior experience. Has the dog you rescued from the pound been abused in the past or grown up playing tag? Is learned helplessness or learned optimism present? Even if their mental functions are limited, your client is not coming as a blank slate.
8. Analyze the Context (setting)
Behavior doesn’t occur in a vacuum. There is always context. What triggers, cues and distractions are present? Is the current behavior a link in a behavioral chain? Where does the current behavior fall on that chain? What are the natural consequences provided in the environment?
The environmental stimuli of a setting have a strong impact of on emotion and behavior. These stimuli are classically conditioned and rarely extinguished. They can last a long time. When you return to your family home, it is not uncommon to be overwhelmed with memories which have been triggered by the setting. Similarly, drug addicts often can’t return to their previous haunts or friends because their cravings are triggered by the setting.
Combinations of stimuli can trigger responses regardless of the consequences. In sign tracking (sometimes called auto-shaping), a pigeon is given food with no requirement for doing anything. The food is just periodically given. Just before the food arrives, a lighted key is illuminated. At first the pigeon just orients to the key; the lighting up gets it attention. Soon it starts pecking at the lighted key. Once theory is the pigeon thinks it is food. Another theory is that the lighted key triggers the eating sequence (like Pavlov’s dogs beginning to salivate).
Interestingly, if the food stops coming (a condition called negative auto-maintenance) the pigeon continues to peck at the lighted key. This phenomena might be limited to birds but it indicates that classical conditioning can potentially impact operant conditioning settings. Regardless of the mechanism, it is clear that the contextual setting is very important.
9. Observe current behavior
What does the current behavior look like; what are its physical features (topography)? Doesit occur cyclically or at a particular time of day or year (temporally)? What exactly is going on?
10. Consequences
How does the current behavior function? Why does it occur (functionally)? Does it provide auto-stimulation, auto-reinforcement, escape, obligations, learned helplessness, punishment or reinforcement?
Auto-stimulation is any behavior that a person generates to initiate internal sensations or processes. In infants, REM sleep auto-stimulates brain growth. The stimulation can be soothing, erotic or neutral. Auto-stimulation includes rocking, hair twirling, finger tapping, etc.
Auto-reinforcement is any behavior that you do because it is fun to you. More mindful than auto-stimulation, auto-reinforcement includes starting fights because you like the internal sense of power it brings, practicing singing even though everyone tells you to be quiet, and making your bed when no one tells you to do so. People can’t be completely controlled because there are internal rewards no one can access.
Escape is a release from chores, undesired behaviors or adverse circumstances. It includes getting out of a burning building, calling your therapist to get out of rehab chores, and hitting the teacher (being sent to the principal’s office) to get out of taking a test. Escape behavior is rewarded with negative reinforcement.
Obligation is the cost of performing a behavior. Does doing the behavior sign you up for magazine subscriptions? Does asking for help or attending a wedding incur an additional family obligation? What is the cost of doing business?
Learned helplessness is a consequence of random punishment. The individual learns that there is no solution to the problem so it doesn’t try to solve it on subsequent trials or on similar tasks.
Punishment decreases the likelihood a behavior will occur again. It impacts the entire class of behaviors. If the behavior is less likely to occur, the consequences are punishing.
Reinforcement increases the likelihood a behavior will occur again. It impacts the entire class of behaviors. If the behavior is more likely to occur, the consequences are reinforcing.
11. Is intervention necessary
Not every problem needs to be solved; not every behavior needs to be modified. Self-injurious behavior needs to be stopped but it does not need to be punished. Decide whether outside help is needed.
12. Task analysis
Break it down into bits. Is this a chain, a cluster or a single behavior? What exactly is included?
Set a Goal (where do I want to be)
13. Goal specific, not general
Operationally define your activity. What are the criteria? How will you know you have succeeded? If the goal is to have a horse stand still (general), use a target stick (follow stick). It is easier to see the outcome when the horse touches its nose to the target. Look for subtle signs that your dog isn’t stressed (calming signals).
14. Frequency
How many times to you want the behavior to occur? Is it an everyday behavior or once a year? I want to row every day.
15. Duration
What is the duration? How much time or distance? I want to row for a minimum of 20 minutes or a distance of 3000 meters, whichever comes first.
16. Intensity
How intense? I want an easy workout, without a lot of strain.
17. Pace
What is the pace? I want a steady pace, with occasional bursts just to break it up a bit.
18. Precision
Define the goal with precision. Make the goal easy to measure. I want to drink so many ounces, take so many strokes, run so many feet, etc.
Establish a Behavior (Getting from here to there)
19. Try without reinforcement
As part of a Track, Reward & Try strategy, Start tracking the behavior and see what happens. Is tracking enough to maintain the behavior. Add trying. Is the combination of tracking and trying enough?
20. Reward, not punish
It is possible for punishment to be helpful in some limited circumstances but in general, avoid it. Rewarding is much better. Start with continuous reinforcement. For every instance, give a reward. Look for environmental reinforcers (activities they want to do) that are already present. Are there leaves to go jump in or mud puddles to splash? Rewards are not limited to snacks.
21. Proportional reward
Make the reward proportional to the effort. Instead of giving the same amount of reward for walking 10 feet or 100 yards, build a correlation. For every foot I walk, I get one minute of TV time. The more peas you eat, the bigger the piece of pie. Similarly, after a great effort, give a jackpot (an unexpected abundance of reward).
22. Contingent
Make the contingent on doing the behavior. The reward is given only for completion. This is Grandma’s Law: if you eat your peas, you can have pie.
23. Click or cluck
Use a signal to indicate success. When Grandma says “Well alright then,” you know you have eaten enough of your peas to get some pie. Use a toy clicker as a signal if people will be far away from you while performing (gymnastics, volley ball, running, etc. The clicker works well underwater too (for signally swimmers).
Non-verbal children and animals must be taught that the clicker means reward. The clicker sound is classically conditioned to a favorite bit of candy or snack. The click is not a reward; it provides information. It says “correct.” It is an event marker.
24. Reward Successive Approximations
Start with small steps, not a gigantic task. To teach your dog to shake hands, reward “paw in the air” behavior. Use whichever paw tends to occur most often. Interestingly, female dogs (and cats) are generally more right-pawed; males are left-pawed. Hold up your palm and reward the imitating behavior.
After “left paw in the air” is established, for example, reward the paws movements that are closer to your upheld palm. Then reward those that are even closer. Eventually, you will reward only when the paw hits your palm.
Once the behavior is strongly on cue (“high five” produces a paw-to-palm hit), replace the clicker with a release word which indicates the task is complete.
25. Prompt, not lure
Don’t trick your learning partner into performing a behavior. Don’t lure them with food. Reward them after the success (or partial success). Dogs, for example, which have been lured to go through a tunnel, continue to look at the hand which held the food. The focus is on the food, not the behavior.
In operant conditioning, give the prompt after the event occurs. When the does sits, say “Sit” and give the dog a treat. The prompt will become associated with the behavior and the reward (backward conditioning).
26. Vary reinforcers
Although a continuous schedule of reinforcement is used to acquire behaviors quickly, add variety to the other components. Vary length of training sessions; not everything has to be 10 minutes long. Use a variety of rewards.
The Premack principle indicates that behaviors less likely to occur in free time can be reinforced by more favored behaviors. Running around can be used as a reward for learning high-5.
Latency is a good indicator of reward value. If your elephant is slow to eat the carrots you are offering, try the potato scraps.
Systematically introduce non-food reinforcers during learning. Food is great for starting out; small bits please, you don’t want to ruin their dinner. But verbal praise, smiling and jumping up and down can also be quite rewarding.
Novel stimuli in the setting can also be used as reinforcers. Chasing birds can be a nice change from dog biscuits.
Vary the size of reward. When there has been a lot of effort given, reward it. Give an unexpectedly large jackpot reward. A rare effort deserves a rare reward.
Maintaining a Behavior
27. Change to variable ratio
Once a behavior is well established, change from continuous reinforcement to intermittent reinforcement. Partially reinforced behavior tends to persist. On continuous reward, stopping the reward tends to stop the behavior. With intermittent rewards, the behavior tends to persist.
If you got a fish every time you went fishing, you would accept that as the norm. When you stopped getting fish, you would soon stop fishing. But if you only catch a fish once every two or three trips, you’ll continue to fish despite not catching any fish. To maintain a behavior, make the rewards more like fishing or gambling and less like a guarantee.
28. Thinning
After a behavior is well established and on an intermittent schedule of reinforcement, thin the ratio of behavior to reward. Change the sales commission schedule.
One way to thin reinforcement is a two-fer. In its simplest form, instead of rewarding one high-5, require two high-5s. In technical terms, you are changings the ratio of the fixed ratio reinforcement schedule. More work is required for the same amount of reward.
Another way to thin reinforcement is to give less reward for the same work. Instead of a dog biscuit, you only get a half a biscuit for the same job.
29. Fading prompt
After thinning the reinforcement, fade the prompt. Instead of needing to say “Clean your room,” say “room” or say the prompt quieter. Change from prompting to reminding. Similarly, say “sit” quieter. Make it part of the normal chain of events that occurs when you stop walking.
30. Generalizing
Like other animals, we are context specific. We learn to solve math problems at the dining room table but can’t remember the formula in the living room. Practice your behaviors in a variety of settings.
Teaching your dog to perform a behavior in the presence of distractions requires exposure to different settings. Teach in the left side of the back yard, the right side of the back yard, in the front yard, at the local park, at a distant park, etc. To generalize, behaviors must be practiced in a wide variety of settings.
31. Chaining
A great way to maintain a behavior is to connect it to other behaviors. If you want to remember to floss your teeth, put it in the middle of your morning routine. After you wash your face, and before you brush your teeth, sneak flossing in between them.
Forward chaining is the way we learn a song. We start at the beginning and work our way to the end, adding phrase after phrase. This is a naïve mnemonic device. We do this naturally.
But backward is often more effective. Begin with the end of the chain, and add links to the beginning. Brush your teeth. Then learn to floss and brush your teeth. Then learn to wash your face, floss and brush your teeth.
Backward chaining is a great way to learn poems, creeds, speeches, the Constitution, the Gettysburg Address and the Declaration of Independence. Start at the end and add links to the front.
32. Superstitious behavior
Reinforcement impacts the entire class of behavior (operant). It is not as specific as we like to think it is. Many behaviors are done at the same time. A puppy may high-5, bark and wag its tail. You intend to reward the high-5 but all the behaviors are equally rewarded. Animals tend to develop patterns of behavior that have nothing to do with getting the reward they are seeking. Puppies whine, children pout, and adults act aggressively (or timidly). Although superstitious behavior can’t be avoided completely, over time the rewards should reinforce the desired behaviors more.
33. Extinction
Two things happen when a reward is withheld (extinction). There will be an extinction burst and spontaneous recovery.
A extinction burst is an increase in the behavior that used to get a reward. If you no longer give money for the ice cream truck, your child will not say “okay” and move on to something else. If crying and nagging used to bring forth ice cream money, this behavior is suddenly increase. We try to increase the intensity before we change behaviors. When people argue, they get louder and louder. They are increasing their intensity, not sharing a new piece of information.
Spontaneous recovery is our mind’s way of keeping a rule just is case we need it again. After successfully ridding yourself of the nagging associated with an ice cream truck’s tune, nagging behavior will suddenly reappear. It is similar to your body’s reaction when you start exercising again. “Are you sure you want to do this?” it seems to say, but then gets with the program. Once we have learned a rule, we are reluctant to let it go.
