Welcome to the first in an occasional series of blog posts on recent scientific research, bringing important findings to the notice of our defender community. Should you have any questions or information to share, please contact Andrew Davies.

Take home message
This new research shows public defenders in a South-Eastern state have better knowledge of the science on child witness reliability than jury-eligible laypeople (a sample of college students) but not as good as forensic psychologists. Defenders should consider improving training in this area, retention of experts when needed, and the limited knowledge of the average juror.

• Full report: Julie Buck, Amye Warren, Maggie Bruck & Kathryn Kuehle (2014), “How Common is ‘Common Knowledge’ About Child Witnesses Among Legal Professionals? Comparing Intervieweds, Public Defenders, and Foresnic Psychologists with Laypeople.” Behavioral Sciences and the Law.
• Link: http://www.onlinelibrary.wiley.com/doi/10.1002/bsl.2150/abstract
• Sample: 137 public defenders from a south-eastern state in the U.S. Also surveyed were 39 forensic psychologists, 44 investigative interviewers, and 192 jury-eligible college students.

The Details
The researchers asked 12 questions of public defenders, forensic psychologists, ‘investigative interviewers’ (employees of child protective services, child advocacy centers, and sheriff’s departments) and jury-eligible college students. The questions concerned the reliability of child witness testimony and addressed issues of almost unanimous scientific consensus.

The researchers found:

• Public defenders were only 56% accurate in their knowledge.
• Public defenders with more experience of child abuse cases were significantly more knowledgeable, but far from flawless (67% accurate).
• Forensic psychologists were significantly more accurate than other groups (overall accuracy rate of 80%).
• Public defenders were generally more skeptical of child witness accuracy than other professionals, though this skepticism did not extend to the use of anatomical dolls, which have been shown to increase false reports of abuse in children.

The results for public defenders are below.

Item	Correct answer	% Public defenders correct
1. When asked an open-ended question, such as “What happened next?”, young children's answers are usually as accurate as those of adults.	Yes	16%
2. When asked a strongly leading question, such as “He told you to keep it a secret, didn't he?”, young children's answers are usually as accurate as those of adults.	No	84%
3. It is necessary to ask leading questions to get young children to disclose abuse, because they are too afraid or embarrassed to report it.	No	55%
4. If an interviewer believes that abuse occurred, then it is okay for the interviewer to ask the child suggestive questions (e.g., “Did he touch your private parts?”).	No	65%
5. Suggestive questioning (e.g., “Did he touch your private parts with his hand?”) by an interviewer is okay once the child has already disclosed sexual abuse.	No	56%
6. If a young child does not report or does not seem to remember much information, it is okay for an interviewer to ask what might have happened or what they think happened.	No	90%
7. If an interviewer believes that a child has been abused even though the child denies it, the interviewer should keep interviewing the child until he or she discloses abuse.	No	85%
8. When questioning a child about sexual abuse, an interviewer should say things such as, “Don't be afraid to tell me what she did to you” or “You will feel better once you tell me.”	No	57%
9. If a young child provides inconsistent information over time (adds some details and drops others), it means an abuse allegation is false.	No	57%
10. If children play in a sexual way (demonstrate sexual touching) with anatomically detailed dolls, it means they have been sexually abused.	No	75%
11. Using anatomically detailed dolls (dolls with genitals) increases the amount of accurate information that young children report.	No	29%
12. Using anatomically detailed dolls increases the amount of false information that young children report.	Yes	26%
Overall accuracy rate		56%