The Importance of Feedback

John Stein

After a career in youth services, teaching and the police, John travels with his family and and among other things regularly contributes articles to The goodenoughcaring Journal.


I learned about feedback when studying systems theory in one of my undergraduate courses in sociology or psychology.  The thing I remember is that systems that do not receive feedback cannot adjust themselves.  Systems that cannot adjust themselves cannot survive.  The concept had its origins in engineering.

It reminded me of when I was building a short wave radio from a kit as a child of about ten. Like putting together a model.  You didn’t have to know anything about radios.  Just follow the instructions, screw in terminals and sockets and solder the appropriate wires.  The thing I remember is the radio had a ‘feedback circuit,’ an automatic volume control. Although not important for regular radios and TVs, where you tune to a station to receive its signal and adjust the volume to suit you, this circuit is important on short wave radios, where you tune to a frequency and receive signals from several other radios transmitting on that frequency.  When the volume is adjusted for a weak signal, as from a radio far away, then it suddenly receives a strong signal from a radio nearby, it can be most unpleasantly loud, even painful, especially if you are using headphones.  If it is adjusted to a strong signal, you will miss weaker signals. So there is a feedback circuit that measures the current going to the audio output (to headphones or speaker) and feeds the information back to the amplifier circuit, which then instantly decreases or increases the output current to the level that has been set with the volume control.

There are many systems in the human body that adjust themselves based on feedback—digestive, circulatory, immune, respiratory, just to name a few.  The system I am concerned about here is the behavioural system.  Most behaviour, both voluntary and involuntary (e.g., reflexes), has a purpose. Behaviour that achieves its purpose in a given setting is likely to repeat itself in similar and possibly other settings when people (and other animals) perceive the feedback.  Behaviour that does not achieve its purpose, or achieves something undesirable, is likely to be eliminated, provided people are aware—i.e., they perceive the feedback.  

The mission of people working with children is to help them learn to perceive the outcomes of their behaviour in various settings, which serves as feedback to help them learn to adjust their behaviours.

‘Good job!’—Is it feedback?

I first heard ‘Good job!’ used in a strict behavioural program on a psychiatric unit for children with dual diagnoses of developmental disabilities and behavioural problems. Children earned tokens (plastic poker chips) for such things as getting up, showering, and being on time for school and appointments.  Then there were a list of rules that exceeded short term memory.  Each of these 35 behaviours had a fine of a few tokens.  An additional 12 behaviours had  room restrictions of from 1 to 3 hours in addition to fines. Staff were not to talk to children who were misbehaving other than to state the behaviour and the consequence—the fine.  “That’s cursing.  That’s two chips.”  Children were then to go to their token jar, take out two chips, and give them to the staff.  Staff didn’t talk with the children much, and when they were doing something staff wanted them to do, the comment was, “Good job making your bed!”  Usually in a singsong kind of exaggerated voice tone.  

The time I still remember after leaving the program over 15 years ago occurred when a boy was serving a room restriction.  I do not like room restrictions, which amounted to children being restricted to their room for 1 to 3 hours.  They could accumulate up to 3 room restrictions—a possible 9 hours.  While serving the room restriction, children could sit on a chair on the ward, provided it was within 10 feet of their room and away from activity areas.  They could not talk to anyone, and staff could only talk with children if it was relevant to how they were serving their restriction.  In Louisiana, it is customary to greet people with ‘Good morning” the first time you see them each day.  It is known as ‘speaking.’  How often have I heard parents in my neighborhood correcting young children—“Jason, aren’t your going to speak?”  I am originally from Pennsylvania and had to learn to ‘speak’ when I came to Louisiana as an adult.  So, when I would go onto one of the units, I would say “Good morning,” to the children, even those serving a restriction.  I received a bit of criticism for this, but my greetings were eventually tolerated.  One morning when I was on a unit where a boy was serving a restriction, a lady in charge of managing the behvioural program walked onto the unit, looked at the boy and said “Good job serving your restriction!” in that singsong voice tone.  Then turned away, completely ignoring the boy from then on. 

The next time I heard ‘Good job’ was when a relative who was an elementary school teacher in a nearby state was visiting our daughter’s family for the holidays. In the exact same voice tone she ‘complimented’ our 4-year-old granddaughter with “Good job” for whatever she was doing, and then went back to her reading.  I recently learned from our granddaughter, now 10, that the teachers in her school use the term with some frequency.  

Has it become ubiquitous? I sincerely hope not.

Is it feedback?  I don’t think so, unless it has to do with something that can be considered a job, such as cleaning the windows or cutting grass.  Serving a restriction is not a ‘job.’  Nor are many other things for which I have heard the phrase used.

(The day after I finished writing this section, my wife and I each received mail from our insurance company with a headline in bold print:  “Good Job Getting Your Flue Shot!”   Really?  Getting a flu shot is a job?)

Are Rewards and Punishments Feedback?

Again, I think not, at least, not effective feedback.  If the purpose of the behaviour is to get the reward, yes, then the reward is feedback.  However, rewards can only be effective when they are operating.  If no one is available to deliver the reward, the behaviour is much less likely to occur.  If we want a behaviour to be repeated, then there must be something else to reinforce (rather than ‘reward) the behaviour.  

Skinner and other behaviourists wrote about reinforcing behaviour, rather than rewarding the subjects of their experiments.  Reinforcement is something that makes something stronger, as in rebar that reinforces a concrete driveway.  It is the behaviour that receives the renforcement, i.e., it is strengthened in some way—it is more likely to occur, to occur more frequently, or to become more resistant to extinction.  Rewards, on the other hand, target the subject—person, lab rat, what ever. It is they who receive the reward, whether or not the behaviour is reinforced.  Although I have heard people talk about reinforcing children for their behaviour, I cannot think of a reinforcement that makes children stronger in any way. 

Further, when reinforcements are no longer operating, as when no adults are present to provide the reward, the behaviour undergoes extinction.

Unfortunately, behaviourists did not give us two terms for punishment.  In behavioural psychology, punishment is something that weakens a behaviour in some way, usually making it less likely to occur.  In the ‘real’ world, punishment is something people do to others to make them suffer in some way to ‘pay’ for their undesirable behaviour.  We hope that punishing people will weaken the behaviour.  All we have to do is look at the criminal justice system and crime rate in America to see how effective this strategy is.  

So while beahviourists make a distinction between rewarding people and reinforcing behaviour, we are left with the same term of punishment for punishing people and punishing behaviour. 

More, it is the behaviour that immediately precedes the punishment that is punished.  Consider the shoplifter.  A shoplifter takes precautions to avoid being caught.  No matter how much we want to punish the shoplifter for shoplifting, we are punishing the shoplifter for getting caught rather than for the shoplifting itself.  Or the speeder.  I sometimes drive too fast, but I know how to watch for police and avoid getting caught.  The only times I have gotten a speeding ticket was when I was not watching for police.  The behaviour that was punished was my inattention, not my driving too fast… 

Rewards and punishments do indicate the approval and disapproval of adults, so when pleasing adults is a purpose of children’s behaviour, rewards and punishments may have some effect.  But consider when children are angry with adults and want to retaliate by making adults angry.  What effect does punishment have then?  The punishment imposed by adults communicates to children just how successful they are at making adults angry. The punishment often doesn’t stop the behaviour until adults impose the harshest of punishments.  Adults think the punishment finally stopped the behaviour. The behavior stops because children’s desire to retaliate has been satiated—‘I’ve finally succeeded in making me mum as mad as I am!’ 

Consequently, rather than rewarding children for desirable behaviours and punishing them for undesirable behaviours, we need to do something else.

Natural Consequences

Natural consequences are things that occur naturally as a result of behaviour.  Natural consequences occur whether or not adults are present to impose them.  When children learn to perceive, understand, and anticipate the natural consequences of their behaviour, these consequences begin to influence their behaviour, reinforcing some behaviours and punishing others.  Unlike rewards and punishments, which can only operate when adults are in a position to provide them, natural consequences always occur.   

Providing rewards and imposing punishments distracts children (and other people such s criminal offenders) from the natural consequences of their behaviour.  That’s what makes Restorative Justice effective in many case—it puts offenders in touch with their victims and their community and the real consequences their behaviour has for others—consequences that occur even if they had not been caught.  Restorative justice involves having offenders meet with their victims and others from the community to talk about the offense and how it affected their victims.  It seems that in many cases, offenders give no thought to their victims, either before or after committing their offense.  When they meet with their victims and others sometime after the offense, they see their victims as people and learn exactly how their offenses affected their victims. This often results in some remorse.  Then, the victims and the others from the community and the offender discuss possible consequences, which involve how to make amends to the victim and the community, to repair the harm that has been done as best as possible.  Punishment, on the other hand, is designed to make offenders suffer in some way in proportion to their offense without repairing the harm.  Once they ‘pay’ for their offense through punishment, there’s no need to do anything else, to repair the harm.

So what are these natural consequences?  Well, did the behaviour please someone, or help someone, or make someone happy?  Did it make them like you or think well of you? Did the behaviour harm someone, physically, emotionally, or economically?  Did it make someone angry?  Did it make someone dislike you or hate you?  Did it make someone want to retaliate?  How did the behaviour make you feel?  Did it make you feel proud?  Happy?  Clever?  Stupid?  

Each behaviour has many consequences.  Talking with children about these consequences when they are calm helps them to learn to think.  Asking “Why did you do that?” most often gets a response of “I don’t know,” and makes it a bit more difficult to continue the conversation.  More specific questions such as “What did you hope to accomplish when you said (or did) that are more likely to get some response.  And if the answer is still “I don’t know,” then more specific questions can follow.  “Did you want to make her angry?”  “Did you want to embarrass him?”  “Did you want to pay him back for something he did?”  “How do you think she felt?” “Did you see his face when you said that?”  “How would you feel if someone said (or did) that to you?”  “How do you think he feels about you now?”  “What could you have done differently?”   “What do you think would have happened if you had done…instead?”  And finally, “What are you going to do now?” to make things better.

Logical Consequences

I find the concept of logical consequences helpful.  Logical consequences are consequences designed to target the behaviour rather then the person.  Logical consequences may be necessary when natural consequences are not effective in changing behaviour.  

Consider a teenager who has trouble coming home on time on Friday and Saturday nights. He has no trouble coming home when he goes out after school or during the day on weekends, no trouble coming home on school nights, just on Friday and Saturday nights.  His mother has patiently explained to him several times how his coming home late affects her.  She gets worried, anxious, has trouble sleeping even after he comes home.  But these consequences have had little effect, so mum feels a need to do something else.  Rather than ‘grounding’ him for a week (or a month…), she targets the behaviour.  She grounds him only for Friday and Saturday nights for a week or two.  “It’s important to me to know you are safely home by midnight, and you haven’t been able to do that, so next week, I am not going to allow you to go out after dinner on Friday or Saturday.  You can have friends over if you want, but you may not got out.  That’s the only way I know to be sure you will be home by midnight.  Then we’ll try again.”

There is no desire to punish the child, only to prevent the misbehaviour.  Is it unpleasant to be grounded on Friday and Saturday night?  Of course.  Does the child feel punished?  Not so much as being grounded for a whole week.  Clearly, the consequence—the ‘punishment’—is targeting the behaviour rather than the child.  I.e., it is the behaviour that is punished rather than the child.

Just a Bit More on Punishment and Feedback

Sometimes, punishment does work.  Especially when it happens to someone else. When punishment happens to someone else in a social setting, it is feedback that reinforces the positive behaviours of others.  I’ve mentioned that I occasionally drive over the speed limit.  I only do this when there is no traffic, no other cars around, like late at night. When traffic is heavy, I drive most cautiously.  And when I see someone driving recklessly fast with no regard for others and I see them pulled over by a police car, it reinforces my diving appropriately, obeying the speed limit.

One other example.  When I was about 10 years old and an only child, I had a lot of trouble coming home for dinner at 5pm.  (My father was out of town on a construction job.)  My mother was most patient, for weeks.  She patiently explained the natural consequences—how my coming home late when dinner was ready at 5pm meant that her dinner was spoiled—dried out or cold.  She told me to come home when the street lights came on.  And other things.  Nothing worked.  I was not deliberately coming home late.  I was just unaware of the time.  My friends didn’t have to be home until six, and I was focused on our games, making plays in baseball or football.  She bought me a watch.  It stopped working within a week, apparently damaged while catching baseballs.  She bought me a more expensive watch that was shock proof.  Still didn’t help.  The day I remember, I looked at my watch and it was 5:05pm.  I ran all the way home, getting there by 5:10 and feeling I had done ok.  Without a word when I came in breathless, she took me over her knee, grabbed a blue plastic ruler that belonged to my best friend (we were working on a school assignment at my house), and wailed me on my butt as hard as she could, breaking his ruler. I cried and went to my room when she was finished—after the ruler broke.  Thing is, I had on brand new blue jeans.  In those days, they were quite stiff until they were laundered a few times.  With the heavy denim jeans with rear pockets, I felt no pain.  None.  But it worked.  I was never late again. It didn’t work because of the pain she intended. It worked because of the feedback—I finally realized how my being late affected my mother.  It made her so angry she tried to hurt me.  And because I had to explain to my best friend how his ruler got broken.  It was the natural consequence, not the pain of the spanking that changed my behaviour.  From that day forward, I learned to be VERY aware of the time. To this day, if anyone asks me what time it is, I can tell them within 5 minutes without looking at my watch or a clock.  In fact, when I awake in the middle of the night, as old men so often do, I usually know the time before looking at my bedside clock. I am almost always right  within five or ten minutes.  My mother’s anger was the feedback that resulted in a change that has lasted some 60 years.

Some Conclusions

While we can certainly influence children’s behaviour with rewards and punishments (although not always as we intend), our ultimate goal should be to teach children to control their own behaviour.  The natural consequences of their behaviour are the only things that can do that consistently.  Helping them to perceive and understand those natural consequences provides our best strategy.  When natural consequences are ineffective, and they sometimes are, we can then use logical consequences to target the behaviour with our consequences rather than punishing our children.  When they learn to control their own behaviour, we can relax and be confident that they will do well. And when they don’t, that they will learn from their mistakes, even when we are not there to teach them.  Or punish them. 


To comment on this article or to contact the author email

Return to Journal Index