In 2013, my colleague and friend, Dr. Robert Simon, and I wrote in our book3 that we believed cognitive biases to be the greatest risk to forensic neutrality and objectivity in child custody and other forensic work.  Many parents often feel that the custody evaluator is biased when a report comes in against that parent’s wishes.  It is common for litigants and their attorneys to believe that the evaluator did not like the client, did not utilize a neutral process, or reached conclusions that are not supported by the data.  Because the outcome is unfavorable, and because the process may have been suspect, the belief is that the evaluator must have reached this unfavorable conclusion because of bias.  Evaluators also reach conclusions that are displeasing to a litigant based upon a solid and well integrated piece of work.  However, bias is perhaps the greatest threat to the integrity and probative usefulness of forensic work products.  Understanding what bias is and is not, understanding various types of bias and understanding how bias can be detected in child custody evaluations is fundamentally important.

The numerous practice guidelines and Rules of Court that I know of each have an admonition that tells evaluators to avoid the impact of biases in child custody work4.  Custody evaluators are not advocates for one party, nor are they advocates for a particular outcome.  Evaluators are advocates for a thorough and scientifically supported process that gathers comprehensive data of diverse nature, tests various hypotheses, and reaches conclusions that are supported by the data gathered.  In order to accomplish this task, evaluators must avoid letting biases of different kinds enter their  reasoning.  In this article, I will briefly discuss the concept of bias in child custody work, how it surfaces, and the potential impact in different types of cases.  Part 2 will focus on ways to spot this bias in child custody evaluations.

The Risk of Heuristics and Cognitive Bias in Child Custody Work

Although there are many different forms of bias, such as personal biases, e.g., gender bias, or professional bias, e.g., relocation or overnight bias, I’m going to focus on cognitive bias for this article.  In recent years, there has been ongoing research related to judgment suggesting that clinicians and others are prone to distortions based on various cognitive biases, attribution effects, and similar heuristics that lead to speeding up the process of reaching conclusions.  These types of bias are more subtle, and I believe they are the type most likely to influence the work of child custody evaluators, especially those who often are seen as doing good work and having proper procedures.

Heuristics are defined as simple, efficient rules that describe how people make decisions or reach conclusions when faced with complex problems.  Certainly the nature of problems that occur in child custody disputes are complex and the factors that must be considered and weighed in making decisions about the custody of children is highly complex.  Kahneman5 identified that people use a variety of heuristics to solve complex problems, often creating a shortcut in logic and reasoning, observing that people tend to use heuristics that are overly simple because of the difficulty of complex heuristics.  Some heuristic shortcuts lead us to solve complex problems by focusing on simple issues, or only part of the problem, and others lead us to ignore some of the information we have to reach our solutions.

In addition to heuristics, Chabris and Simons6 observed that humans tend not to see what we aren’t looking for.  For example, is communities where pedestrians are more common, there are fewer accidents because drivers are looking out for pedestrians, whereas in communities where pedestrians are less common there are actually more accidents since drivers are not expecting to see them.  In the same way, child custody evaluators who are focused on one element of the family dynamics, e.g., domestic violence or high conflict, might not even look at  dynamics associated with other issues, e.g., quality and history of parenting, Gatekeeping, etc.

There are many examples of how these heuristics and cognitive biases, as well as blind spots, might operate in a child custody case.  Potential heuristics or cognitive biases that are most common in child custody work can include:

Anchoring Heuristic

With anchoring, the evaluator will overly rely on certain information during the evaluation process at the expense of other information.  Once  the anchor is “set”, there is a risk that other information is interpreted in a way that is consistent with that anchored information.  The “Primacy Effect” is an example of a bias that includes the anchoring heuristic.  The primacy effect is observed in situations where the data that we gather first affects the way we interpret and gather later data.  The early data anchors our understanding of subsequent data.

Availability Heuristic 

The Availability heuristic refers to the tendency to focus on what is most available in memory.  Things that may increase this availability include data that is more vivid or unusual, or perhaps more emotionally charged.  When something is repeated frequently, we tend to remember it more and even believe it more.

Confirmatory Bias

Confirmatory bias is the tendency for a custody evaluator, having formed an opinion or strong impression before completing all of the data gathering, to start looking for certain data or evidence that supports the opinion or impression that has formed.  Then, data that is collected is seen through the evaluator’s pre-­conceived beliefs and used to support the preconceived opinion rather than the data being fully and neutrally evaluated.  Data should be gathered in a systematic manner.  When data is gathered in a selective manner or is perceived through a pre-fashioned lens, there is greater risk of being influenced by confirmatory bias.  Confirmation bias leads to increased confidence in one’s findings, largely because the process of gathering data and data analysis was not scientifically grounded and undertaken in a forensically neutral manner.

Recency Bias  

Recency bias is the cognitive bias that exists by focusing on the most recent data one has heard and reaching conclusions based on that data.  The opposite of Primacy bias noted above, there is a tendency with Recency bias to de-­emphasize data gathered earlier in the evaluation process and emphasize the data gathered towards the end of the evaluation process.

Stereotyping

With stereotyping, the evaluator is affected by characteristics of the individual being evaluated rather than by the data being collected.  For example, when Parent A appears to be histrionic and over-­reactive, the evaluator simply makes an assumption of Parent A’s claims and allegations are the result of their histrionics and gives little weight to data that appear to support the allegations being made by the parent.  This is often consistent with confirmatory bias as well.

Data Gathering Bias

I frequently notice that some evaluators will believe more in some types of data than others and will have greater faith in one part of the process, giving greater weight to information obtained via that procedure.  For example, if the evaluator highly believes in his capacity to observe healthy parenting behavior, he may give greater weight to his observations than other collected data.  Similarly, some evaluators have great faith in the value of psychological testing to support conclusions about parenting and custody.  Finally, other evaluators might believe that they can determine credibility on the basis of their interviews with the parents.

Research Bias  

There is a risk that custody evaluators will use research to support a pre-­conceived opinion.  I often see report narratives in which evaluators generically describe that “research suggests” a particular thing when formulating opinions and recommendations at the conclusion of an evaluation.  They do so without providing citations to the research being  mentioned or without describing research that might support a different outcome.  This regularly occurs in relocation-­related evaluations.

“Truth Lies Somewhere in the Middle” Bias

Many evaluators and judges, in particular those who are at risk for burnout because they have worked in the system for so long, have a greater tendency to exhibit this bias.  There is a tendency to perceive that parents in conflict over the custody of their children make an equal contribution to that conflict.  While that situation is seen in some high conflict situations, there are other instances in which one parent drives most of the conflict and the other parent tends to be more reactive to that conflict.  This “truth lies somewhere in the middle” bias prevents evaluators and judges from recognizing the unique contributions of each parent to the conflict.  However, these unique contributions to the conflict are likely to be an important and relevant factor to consider in a given case, particularly when determining the specifics of a child-­sharing plan.  Assigning equal blame to both parents is a mistake when the responsibility for different components of the conflict are more likely caused by one parent rather than the other parent.

“For the Move” or “Against the Move” Bias7 

From my perspective, many child custody evaluators appear to have one of these two points of view; they either see relocation as something to be avoided at all costs or they tend to be in favor or relocation by a primary custodial parent.  Those who tend to be pro-­relocation take the position that a custodial parent who wishes to move should generally be allowed to move as long as the custodial parent has a legitimate reason for moving and is not attempting to interfere with the access rights of the other parent.  Evaluators might bring a unitary approach and conclude that this parent can move with the child if they determine that one or the other parent is “the psychological parent” or primary custodial parent.  They may also conclude that the move should be permissible after determining that there is a legitimate reason for moving or that there is no evidence of interference with the other parent’s access.  While the laws relating to parental relocation vary jurisdictionally (there are many states in which case law or statutory law supports such a presumption in favor of moving), there is no evidence in the psychological literature to suggest that it is helpful or appropriate for psychologists to have such a presumptive belief in relocation cases.  There is no research suggesting  that because a parent is happy following a relocation that the children will automatically be happy and adjust to the move.

Conversely, there are many custody evaluators who perceive that it is a parent’s responsibility to stay near the other parent in order to preserve the child’s access to the other parent and the involvement of both parents in her life.  While there is research data to support the belief that children derive a benefit by having both parents’ active involvement in their lives8, extrapolating that data to support a presumption against moves confounds the issue.   There are many circumstances in which a move is both legitimate and justified, whether for academic, economic, or other personal/family reasons.  In those cases, a parent is going to move, with or without the child.  In those circumstances, it is incumbent on evaluators and judges not to confuse the preference and value for shared co-­parenting that exists in some of the research and some statutory laws with a presumption that moves will automatically harm children.

Conclusions

This brief article has addressed the many heuristics that potentially impact the work of child custody evaluators.  Part 2 will focus on how to review a custody evaluation report and consider whether or not the evaluator might have been blind to critical issues or were affected by one or more heuristics, leading to missing data or an over-­simplified analysis of the issues.


1.  First of two parts, this article focuses on the nature of cognitive biases and Part 2, which will focus on how to observe if bias has potentially interfered with an evaluation will follow in a future newsletter.

2.  Dr. Stahl is a Board Certified Forensic Psychologist, licensed in California, Arizona, Hawaii, and Michigan.  He is a former Board member of AFCC, as well as the CA and AZ chapters and is a regular speaker at AFCC meetings across the country.  When not speaking and writing, Dr. Stahl serves as a consultant and expert witness, as well as a court-appointed child custody evaluator throughout North America.

3. Excerpted and modified from Chapter 4, Stahl, PM and Simon, RA (2013).  Forensic Psychology Consultation in Child Custody Litigation: A Handbook for Work Product Review, Case Preparation, and Expert Testimony, Chicago, IL: Section of Family Law of the American Bar Association.  For more information on this topic, please see the entire chapter.

4.  See e.g., CA Rule of Court 5.220, APA Guidelines, AFCC Model Standards,  AACAP Guidelines

5.  See e.g., Kahneman, D., (2011).  Thinking Fast and Slow. New York: Farrar, Straus and Giroux, and Tversky, A. & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases, Science, 185, 1124–1130]

6.  See e.g., Chabris, C and Simons, D, The Invisible Gorilla: How our Intuitions Deceive Us, Harmony, 2011

7.  See Stahl, PM, “Avoiding Bias in Relocation Cases”, Journal of Child Custody, 3, 3/4, 109-­124

8.  See e.g., Kelly and Emery, 2003