Children's memory for experienced events can be every bit as reliable as that of adults and adolescents, given that 'accuracy' is precisely defined.  A long line of psychological research has illuminated several important facets of child event memory and testimony that must be understood by those who wish to use these memories in legal or clinical settings.  When we understand how a child's event recall can be affected in such ways to improve or decrease veracity, we can adequately prevent problems.  Additionally, we can more effectively identify when problematic interviewing styles have introduced the potential for contamination or bias.

We must be careful when using the term 'accurate'.  After all, we never expect 100% accuracy when interviewing adult witnesses about their memories.  We know that they may make some detail errors and forget to mention a thing or two.  Thus, expecting children to be 100% accurate is unreasonable.  If some error for adults is acceptable even in legal cases, so then must it be acceptable for children.  However, we often make the assumption that recall for important details is more resistant to error.

When we recall experienced events, we can do so in four ways. We can produce a detail that is accurate (i.e., an accuracy), we can produce a detail that is inaccurate (i.e., a commission error), we can fail to produce an important detail that happened (i.e., an omission error), and we can fail to produce a detail that did not happen (i.e., a correct rejection).

When we say that children can produce event recall that is as accurate as that for adults, the equation used is very specific.  This refers to the accuracy of the details provided.  That is, how true are the things they actually say?  Furthermore, this also refers to recall produced in 'free recall' tasks.  Free recall tasks do not contain specific questions, but rather may ask questions such as, 'Tell me about your day."

In free recall, children and adults make utterances which are approximately 75% true and 25% false.  This is a 3:1 ratio and has been supported in many independent studies.  However, there is one important facet of recall that is not included in the equation and ratio; omission errors.  When we say that children can be as accurate as adults in certain, controlled circumstances, we are measuring what they say - but not what they forgot to say.

In fact, when we measure errors of omission across age groups, we can determine that children engage in far more errors of omission than do adolescents or adults.  What this means is that in free recall, children are more likely to fail to say things which may be important to a legal case.

Investigators know this full well. But this sets the major problem and is the reason that free recall tasks are not the primary method for interviewing children about experienced events.  If it were, the result would probably be either very quiet children who never report things that actually did happen or children who become tangential and lose the thread of the question.

Thus, the most used task is the guided question or even the specific question.  These are standard methods for use when dealing with adults. In the absence of appropriate methods for interviewing children, these have also become standard for interviewing kids.  These guided or direct questions are seen as necessary to help the child produce the details important for the case.

However, research shows that children are more susceptible (than adults) to contamination by the extra information that is often included in specific questions. In addition, a recent study also shows that these questions may begin to affect the ratio discussed above.

In summary, this means that in free recall, kids are as accurate as adults in what they say, but they say less.  Police need the details, so are forced into asking specific questions.  Problematic specific questions may introduce contamination. 

However, there is a bright side.   The research base has illuminated several major sources of bias.  We know that several common interviewing techniques can introduce bias.  Knowing this, we can easily minimize bias by ensuring that these 'biasing factors' are not present in interviews.  Also, research has illustrated some useful interviewing techniques which improve overall testimony by reducing omission errors, without increasing commission errors.

Even so, most child forensic interviews are done without the benefit of this information and usually in an unstructured, free-form and problematic manner, even though this is now easy to correct.  The result is that the product of these interviews is easy to bring into doubt.  When an interview has no clear examples of problematic questions, it is usually a powerful piece of evidence.  When there are numerous problematic questions, it is usually a weak piece of evidence - if those questions can be identified accurately.

I have evaluated numerous interviews conducted by police departments and have found that most are still problems, though a few departments have done great work in improving their interviewing techniques with children. 

If you are an attorney (prosecuting or defense) dealing with a case of child sexual abuse, I am usually willing to consult regarding these cases.  I can help to evaluate interviews for the presence or absence of problematic interviewing styles.  Also, I can help programs to learn to control these problems before they happen.

A close colleague, William O'Donohue, Ph.D., is also usually willing to consult in these cases.

If you would like more information,

Contact        Matthew Fanetti, Ph.D. at MFanetti@MissouriState.edu

                                            or

                      William O'Donohue, Ph.D. at wto@unr.edu