Back in my undergraduate years I took a required course for all psychology majors entitled Experimental Psychology. Early in that course in a lab we were each instructed to train a white rat to turn right in a T-maze. (A successful trial was defined as the rat not going past a line on the left side of the maze and the rat had to proceed all the way to the right within five seconds—to eat a small piece of compressed grain.) When the rat made five consecutive “correct” responses, it was determined that the rat had “learned” the appropriate response. I was so proud that my rodent took the fewest number of trials of the group to reach the criterion.

Once all the rats in the class had been “taught” to turn right, the issue became how many trials would it take for all the rats to “learn” to turn left (meeting the same criterion for a correct response as before). Three different randomly assigned conditions were used under which the rats would learn the new response: Under condition 1 the rat would receive a mild shock if they turned right (punishment). Under condition number 2 the rat would receive a mild shock for turning right but receive food for turning left (punishment and reinforcement). Under condition 3 (the one I and my rat happened to be in) the rat would simply receive food for turning left (reinforcement, only).

We students were asked to vote ahead of time as to what condition we thought the rats would learn the new task the fastest. The general consensus among the 40 or so undergraduate psychology students was condition 2. We were wrong. It was condition 3—by a landslide. The rats in condition 3 required statistically significantly fewer trials to all reach criterion than the rats in condition 2 and the rats in condition 2 similarly out-performed the rats in condition 1.

Upon doing the subsequent research for the write-up of this experiment it became evident that for years it had been well-documented that condition 3 (reinforcement, only) was, by far, the best way to change a behavior—with animals or children.

The research is clear: If you want to change a behavior in your child, it is far more effective to reward the alternative desired behavior than to punish the undesirable behavior. Nevertheless, despite a half-century of conclusive research, most parents punish the undesirable behavior—which the data indicates is the least effective way to change that behavior.

The next time your child misbehaves, try to ignore that behavior and attempt to arrange the environment such that the child subsequently exhibits the appropriate alternative behavior. When that occurs, be sure to reinforce it. In short time the child will be exhibiting the desired behavior consistently. (This process also works for spouses, too.)

Author's Bio: 

Larry F. Waldman, Ph.D., ABPP is a licensed psychologist who has practiced in the Paradise Valley area of Phoenix for 32 years. He works with children, adolescents, parents, adults, and couples. He also provides forensic consultations in the areas of family law, personal injury, and estate planning. He speaks professionally to laypersons, educators, corporations, and fellow mental health professionals. He teaches graduate courses for the Educational Psychology Department for Northern Arizona University. He is the author of “Who’s Raising Whom? A Parent’s Guide to Effective Child Discipline,” “Coping with Your Adolescent,” “How Come I Love Him But Can’t Live With Him? Making Your Marriage Work Better,” “The Graduate Course You Never Had: How to Develop, Manage, Market a Flourishing Private Practice—With and Without Managed Care,” and “Too Busy Earning a Living to Make Your Fortune? Discover the Psychology of Achieving Your Life Goals.” His contact information is: 602-996-8619; 11020 N. Tatum Blvd., Bldg E, Suite 100, Phoenix, AZ 85028;;