Should Forensic Psychologists Make Custody Recommendations?

Unfortunately, few attorneys and judges are aware of the somewhat intense controversy occurring among psychological professionals about whether we should be making custody recommendations at all. Within the last few months the National Register for Health Service Providers published an article for psychologists in which it is explicitly stated that clinicians should avoid making such recommendations (Morris 1997, 2003), a stand congruent with that previously taken by eminent forensic psychologists in treatises on custody assessment (eg, Melton, Petrilia, Poithress, and Slobogin, 1997; Heilbrun, 2001). Some forensic ethicists have gone so far as to call for a general moratorium on the participation of psychologists in this area of clinical work (O'Donohue & Bradley, 1999). The American Psychological Association (APA), in its guidelines for practitioners (1994), also notes that ” … the profession has not reached consensus about whether psychologists ought to make recommendations about the final custody determination to the court (p. 679).”

Following is a brief outline of the most salient arguments against the practice of making specific custody recommendations, offered in order to foster more dialogue about this very important issue.

Arguments Against Making Specific Recommendations

Inadequate specialized knowledge. A defining characteristic of the expert witness is the notion that the expert's discipline has specialized knowledge that goes beyond information possessed by the general public and that goes beyond mere common sense or logic. However, with regard to the capacity to make reliable and valid predictions about specific custodial schedules for a child, it is easy to argue that psychology as a discipline falls far short of what should be required of the expert witness.

Psychology has a long tradition of illuminating and comprehensive theories about human functioning. We have now accumulated many decades of empirical research on child development and personality functioning. A literature is slowly developing about the benefits and liabilities associated with various custody arrangements for aggregate groups of children who are studied via research paradigms. Unfortunately, our capacity to translate this theoretical and empirical tradition into highly specific custody recommendations that can be shown to be reliably associated with positive child functioning is wholly unproven. That is, it has not been established that specific custody recommendations, if followed, directly cause more positive child functioning than other plans that might be chosen.

While our growing knowledge about the effects of various custody arrangements on children may eventually provide “broad brush” guidelines for finders of fact, it is very unlikely that our field will ever be able to produce actuarial data or empirically reliable decision rules that would be able to take into account the host of unique, person-specific and family-specific variables presented by real, living and breathing families (eg, how can we reliably assess whether a specific plan fits for a circumstance with a mother possessing certain traits, a father possessing certain traits, children with certain traits, sibling constellations involving different developmental levels, unique environmental stressors, etc.?). Although many clinicians are willing to offer highly specific recommendations that imply they are able to mentally juggle such complex factors, most careful researchers would assert that our capacity to make such multivariate predictions is, at least at this point, unproven. In addition, what we do know via research is entirely correlational in nature because we are unable to randomly assign children to different custody arrangements in order to establish which plans cause which child which outcomes under which conditions, in order to develop reliable decision rules.

Value-laden nature of the ultimate “choice.” It has been asserted that even “hard” scientists cannot be value-free in their work. As a social science, psychological assessment and reasoning is even more infused with a valuing process that involves implicit assertions about things that are “good” for people, “bad” for people, and so forth. However, within the broad dimension of best interests, the chance for a clinician's personal values, biases, social judgments, and subjective assumptions to be mistaken as objective and compelling science is quite real. While it may be possible for a psychologist to make a relatively objective assessment of certain custody-relevant dimensions (is a parent depressed, is a child adaptively attached to a parent, etc.), it is the next level of assumptions often present in custody reports (the “shoulds”) that can become infused with the clinician's values and subjective assumptions under the guise of “science”: children should live with their primary attachment figure; babies should sleep in one primary home; shared custody is bad for children if the parents are arguing frequently; occasional alcohol abuse should eliminate a parent as a custodial option.; physical discipline is bad for children. While a clinician might be able to garner theory-based or empirical research support for such assertions, reasonable people can disagree about whether such assumptive guidelines should be controlling decisions about real families. Given their deeply subjective and value-laden nature, such assertions and the recommendations they are used to support are more appropriate for the finder of fact than for a psychologist operating within a forensic framework because their presentation within a psychological report, unless explicitly labeled as such, conveys a deceptive air of scientific objectivity about value-laden matters.

It is the psychologist's job to assess the “whats” of a family (What indications of impaired parenting capacity are present? What indications of pathologic parent-child relations are present? etc.). It is the role of the judge to enter into the realm of the “shoulds” (Now that it has been established that there is a pathological relationship between this parent and this child, where should the child live?).

Lack of Professional Agreement About Method and Theory

There is no shortage of books and articles about preferred methods for assessing custody issues. A review of this literature, however, while yielding certain consistencies in the prescribed evaluation protocols, is most striking in the number of differences between authors about important issues. Some say all parents should complete an intelligence test while others view this as a waste of clinical time. Some authors suggest exploring a child's preferences while others view this as a dubious and highly risky clinical practice. Some suggest always having a joint session between parents while others view such sessions as unnecessary. While certain forensic writers encourage clinicians to interview extended family members, others downplay or omit such a procedure. Some clinicians are enamored with extensive batteries of projective and objective personality inventories whereas others view such protocols as only relevant for a small percentage of families. Our discipline even has heated and profound debate about the meaning and relevance of certain developmental theories that are often the central underpinnings of custody recommendations (eg, while some use attachment theory to argue for the identification of the “psychological parent” and for schedule continuity in custody evaluations, others assert that such an approach is outdated or, at least, overly simplistic).

Given that divorce psychology remains in its infancy, and given the extreme complexity of the psychological issues at hand in custody matters, it is no surprise that our discipline is evolving in its understanding of optimal assessment procedures and in how to apply its own theories. However, this does not vitiate the reality that a discipline should not be viewed as competent to render valid and reliable custody recommendations when it is unable to achieve significant consensus about appropriate behavioral-scientific technique or about the application of its central developmental concepts.

Forensic psychological assessments are often pivotal documents that can have a dramatic effect on the trajectory of a contested custody dispute and, ultimately, on the path a particular child's life will take post disposition. These documents are often eagerly awaited because of their potential value in providing leverage for one side over the other and for their capacity to settle cases – qualities derived from the frequent presence of a specific custodial recommendation in the report. Clients arrive at the clinician's office often feeling as if their lives are in the hands of the court-appointed expert. Forensic reports arrive in court as documents that represent the application of a behavioral “science” and there is therefore a common expectation that the recommendations will be weighted heavily because they will go beyond common public knowledge or subjective value choices. It is the recommendations section that is often read first by the consumers of the report because of its perceived impact.

Those who argue that it is appropriate for psychologists to make specific custody recommendations offer, among others, the following rationales: An important component of a best interests decision is the achievement of a family plan most likely to support a child's emotional health and happiness – and who better to make such an assessment than a doctor skilled at understanding the factors that promote healthy emotional functioning in children? What professional, among those involved with a divorcing family, is better equipped to evaluate a child's emotional and developmental needs? Others state that it is simply unrealistic to assume that judges – professionals primarily trained in matters of law – can competently analyze and weight the various issues related to the best interests of a child, issues that require a capacity to apply psychological knowledge. Judges need our recommendations because psychologists know how to understand children and families in a way that they do not.

Intense Controversy

Unfortunately, few attorneys and judges are aware of the somewhat intense controversy occurring among psychological professionals about whether we should be making custody recommendations at all. Within the last few months the National Register for Health Service Providers published an article for psychologists in which it is explicitly stated that clinicians should avoid making such recommendations (Morris 1997, 2003), a stand congruent with that previously taken by eminent forensic psychologists in treatises on custody assessment (eg, Melton, Petrilia, Poithress, and Slobogin, 1997; Heilbrun, 2001). Some forensic ethicists have gone so far as to call for a general moratorium on the participation of psychologists in this area of clinical work (O'Donohue & Bradley, 1999). The American Psychological Association (APA), in its guidelines for practitioners (1994), also notes that ” … the profession has not reached consensus about whether psychologists ought to make recommendations about the final custody determination to the court (p. 679).”

Following is a brief outline of the most salient arguments against the practice of making specific custody recommendations, offered in order to foster more dialogue about this very important issue.

Arguments Against Making Specific Recommendations

Inadequate specialized knowledge. A defining characteristic of the expert witness is the notion that the expert's discipline has specialized knowledge that goes beyond information possessed by the general public and that goes beyond mere common sense or logic. However, with regard to the capacity to make reliable and valid predictions about specific custodial schedules for a child, it is easy to argue that psychology as a discipline falls far short of what should be required of the expert witness.

Psychology has a long tradition of illuminating and comprehensive theories about human functioning. We have now accumulated many decades of empirical research on child development and personality functioning. A literature is slowly developing about the benefits and liabilities associated with various custody arrangements for aggregate groups of children who are studied via research paradigms. Unfortunately, our capacity to translate this theoretical and empirical tradition into highly specific custody recommendations that can be shown to be reliably associated with positive child functioning is wholly unproven. That is, it has not been established that specific custody recommendations, if followed, directly cause more positive child functioning than other plans that might be chosen.

While our growing knowledge about the effects of various custody arrangements on children may eventually provide “broad brush” guidelines for finders of fact, it is very unlikely that our field will ever be able to produce actuarial data or empirically reliable decision rules that would be able to take into account the host of unique, person-specific and family-specific variables presented by real, living and breathing families (eg, how can we reliably assess whether a specific plan fits for a circumstance with a mother possessing certain traits, a father possessing certain traits, children with certain traits, sibling constellations involving different developmental levels, unique environmental stressors, etc.?). Although many clinicians are willing to offer highly specific recommendations that imply they are able to mentally juggle such complex factors, most careful researchers would assert that our capacity to make such multivariate predictions is, at least at this point, unproven. In addition, what we do know via research is entirely correlational in nature because we are unable to randomly assign children to different custody arrangements in order to establish which plans cause which child which outcomes under which conditions, in order to develop reliable decision rules.

Value-laden nature of the ultimate “choice.” It has been asserted that even “hard” scientists cannot be value-free in their work. As a social science, psychological assessment and reasoning is even more infused with a valuing process that involves implicit assertions about things that are “good” for people, “bad” for people, and so forth. However, within the broad dimension of best interests, the chance for a clinician's personal values, biases, social judgments, and subjective assumptions to be mistaken as objective and compelling science is quite real. While it may be possible for a psychologist to make a relatively objective assessment of certain custody-relevant dimensions (is a parent depressed, is a child adaptively attached to a parent, etc.), it is the next level of assumptions often present in custody reports (the “shoulds”) that can become infused with the clinician's values and subjective assumptions under the guise of “science”: children should live with their primary attachment figure; babies should sleep in one primary home; shared custody is bad for children if the parents are arguing frequently; occasional alcohol abuse should eliminate a parent as a custodial option.; physical discipline is bad for children. While a clinician might be able to garner theory-based or empirical research support for such assertions, reasonable people can disagree about whether such assumptive guidelines should be controlling decisions about real families. Given their deeply subjective and value-laden nature, such assertions and the recommendations they are used to support are more appropriate for the finder of fact than for a psychologist operating within a forensic framework because their presentation within a psychological report, unless explicitly labeled as such, conveys a deceptive air of scientific objectivity about value-laden matters.

It is the psychologist's job to assess the “whats” of a family (What indications of impaired parenting capacity are present? What indications of pathologic parent-child relations are present? etc.). It is the role of the judge to enter into the realm of the “shoulds” (Now that it has been established that there is a pathological relationship between this parent and this child, where should the child live?).

Lack of Professional Agreement About Method and Theory

There is no shortage of books and articles about preferred methods for assessing custody issues. A review of this literature, however, while yielding certain consistencies in the prescribed evaluation protocols, is most striking in the number of differences between authors about important issues. Some say all parents should complete an intelligence test while others view this as a waste of clinical time. Some authors suggest exploring a child's preferences while others view this as a dubious and highly risky clinical practice. Some suggest always having a joint session between parents while others view such sessions as unnecessary. While certain forensic writers encourage clinicians to interview extended family members, others downplay or omit such a procedure. Some clinicians are enamored with extensive batteries of projective and objective personality inventories whereas others view such protocols as only relevant for a small percentage of families. Our discipline even has heated and profound debate about the meaning and relevance of certain developmental theories that are often the central underpinnings of custody recommendations (eg, while some use attachment theory to argue for the identification of the “psychological parent” and for schedule continuity in custody evaluations, others assert that such an approach is outdated or, at least, overly simplistic).

Given that divorce psychology remains in its infancy, and given the extreme complexity of the psychological issues at hand in custody matters, it is no surprise that our discipline is evolving in its understanding of optimal assessment procedures and in how to apply its own theories. However, this does not vitiate the reality that a discipline should not be viewed as competent to render valid and reliable custody recommendations when it is unable to achieve significant consensus about appropriate behavioral-scientific technique or about the application of its central developmental concepts.