BOTTOM LINE

“Confirmation Bias”

“the tendency to process information by looking for, or interpreting, information that is consistent with one’s existing beliefs. This biased approach to decision making is largely unintentional and often results in ignoring inconsistent information. Existing beliefs can include one’s expectations in a given situation and predictions about a particular outcome. People are especially likely to process information to support their own beliefs when the issue is highly important or self-relevant.”

(The Encyclopedia Britannica, britannica.com/science/confirmation-bias )

“When the only tool you have is a hammer, every problem looks starts to look like a nail.”

ST vs. Ramon Mendez Escobar, October 6, 1997

Page 70-72

Q. But, in fact, you’re a child advocate, aren’t you?

A. Yes.

Q. And right on you curriculum vitae it indicates that part of your job over at St. Joe’s is to provide expert testimony, child advocacy and community outreach, correct?

A: Yes.

Q. And you consider that part of your job to be providing testimony on behalf of children?

A. No. The child advocacy part – the majority of my time is spent working with community agencies to make child abuse, investigations easier for children, less traumatic.

Q. But you have built your professional career, and really a lot of your identity, on being a child advocate, right?

A. Both for – well, that’s not my primary objective, no.

Q. But that’s something that you pride yourself on being, and advocate for children?

A. Yes.

Q. And in child molestation criminal cases you have testified on behalf of the State on numerous occasions, haven’t you?

A. Yes.

Q. And you haven’t testified for a defendant, have you?

A. That’s correct.

Page 80 of Dutton’s dissertation

As a forensic interviewer, I was generally considered a neutral party in the investigation process. However, I was also considered an advocate for abused children in other areas of my professional life. For example, I have served on the Governor’s Office Children’s Justice Task Force, a task force that sought to improve how child abuse cases are investigated and prosecuted. I have also advocated for individual children in the legal system as the need arose. Therefore, I had to be vigilant in my research process to identify, ameliorate and report my potential bias. I was also open to conclusions driven by data that had the potential to be contrary to my prior expert testimony in child abuse litigation.

Gender differences in children’s disclosures and legal narratives of sexual abuse.

Wendy A. Dutton. A dissertation presented in partial fulfillment of the requirements for the degree doctor of philosophy.  Approved April 2011, Madelaine Adelman, Chair. Arizona State University.  Page 80.

Q: Are you still a biased child abuse advocate when you are not conducting a forensic interview?  For instance, now in your role as a cold expert?

A: No.

DEEP DIVE

Psychological science has found that experts tend to find what they “expect” to find.  This goes by different names: “Confirmatory Bias”, the “Rosenthal Effect” (“Pygmalion Effect”), the “Rosenhan Experiment (Sane in Insane Places)”, “Mental Set”, “Primacy Effect”, or “First Impressions” dynamic.  The CSA expert needs to demonstrate they are cognizant of this phenomenon that adversely affects objective decision-making and show reasonable steps have been taken to avoid the various pitfalls.  For example – the tendency for information gathering to be influenced by first impressions or a bias.  There is a natural (albeit it unintentional) tendency to accept the essential accuracy of the self-reports of subjects in a study.

•Confirmation biases can occur through a researcher engaging in hypothesis-confirming information seeking and interpretation, overweighting confirmatory signs and underweighting dis-confirmatory signs, especially in unstructured interviews.

•Interviewers’ characteristics and biases can influence how interviewers interpret information and determine subsequent queries (Tanaka-Matsumi, 2004).

•A lack of training and experience on the part of the research interviewers can result in erroneous information and weaken the validity of their findings (Bootzin & Ruggill, 1988).

•Data gathered from interviews can be biased due to social desirability and participants’ own discomfort with disclosing personal information (Guinn et al., 2010). Self-report data can be accurate, but also can be an exaggeration or a minimization (Haynes, Mumma, & Pinson, 2009; Haynes, O’Brien, & Kaholokula, 2011).

•Because the interview method requires retrospective recall, information gathered is prone to memory errors and recall biases (Graham, Catania, Brand, Duong, & Canchola, 2003).”

Psychological research frequently uses techniques for mitigating the effects of confirmation biases, called “blind” or “double-blind” experimental protocols (Gluud, 2006). This is where the interviewer and/or the interviewee do not know the underlying purpose of the study.  This helps to eliminate confirmatory bias.  This writer doesn’t know of a single CSA study that has used “blind” interviewers or interviewees.  This is an important failure that undermines research into CSA.

Problematic Behaviors Should Not Be Used to Infer That Sexual Assault Has or Has Not Occurred

The court should never allow behaviors to be an implied “dispositive tool” (truth-telling instrument).  As already described, the data reported in anecdotal studies do not distinguish between true, unfounded, or false reports of sexual abuse. In addition, because no determination was made to gain additional assurance that sexual assault indeed had occurred, few if any quantitative relations of symptoms or stages have been developed.  There can be many other pathways to the symptoms found (other trauma, previous childhood abuse, etc.). These symptoms were not shown in CSA studies to be unique to sexual assault. Thus, it would be improper to reason backward in an attempt to conclude that because features of the CSA were or were not found in an individual case that the person was or was not sexually assaulted.

A Logical Error

CSA experts can sometimes insinuate a flawed logic in cause-and-effect.  Here’s how it works.

Situation A

•Bob went snow skiing,

•While doing so, he broke his foot.  (Forward logic – “a” led to “b”)

•I see a number of people day-to-day who have a broken foot.

•THEREFORE, they must have broken it while skiing. (Backwards logic – a broken foot means the person was skiing)

Situation B

•A person is sexually assaulted.

•That person goes on to display a number of maladaptive behaviors in the future, like staying in the abusive relationship and delaying disclosure of the abuse.  (Forward logic – “a” leads to “b”)

•The complaining witness in this case stayed in the relationship with the accused and delayed disclosing until after they broke up.

•This is behavioral proof that the victim was sexually assaulted. (Backwards logic)

This is a logical fallacy termed “affirming the consequent.”  In the first situation backwards logic would lead the jury to conclude all individuals with a broken foot did so while skiing.  In the second situation a complaining witness staying with her assailant is proof that the abuse happened.  When examined rationally the conclusion is nonsensical.  However, jurors might accept this line of reasoning because it sounds logical, when in fact it is completely illogical.  Counterintuitive Behaviors are not dispositive as to whether sexual abuse actually occurred.

Problematic behaviors are never evidence that the sexual abuse actually occurred.  The proper use of CSA testimony is to educate the court so that the court does not instinctively conclude that the sexual abuse did not occur based solely on a victim acting in a manner that the court believes is inconsistent with how a “real” victim of sexual assault would behave.