Sbraga, T. P., & O'Donohue, W. (2003). Post Hoc Reasoning in Possible Cases of Child Sexual Abuse: Symptoms of Inconclusive Origins. Clinical Psychology: Science & Practice, 10(3), 320-334.
This article addresses the role of psychological expert testimony in possible cases of child sexual abuse (CSA). It describes the logical conditions that must be met in order to engage validly in post hoc reasoning from present symptoms experienced by a child to the conclusion that the child has or has not been sexually abused. The use of various stress and sexual abuse accommodation theories as support for these conclusions is examined. In addition, the practical problems of using Bayes' Theorem to inform jurors in suspected cases of CSA are described, and an actuarial model of post hoc reasoning is proposed. Given the gaps in knowledge concerning the relevant probabilities and thus the differential probabilities of symptom/abuse versus symptom/no abuse, it is argued that it is not possible currently to reason from symptom observation at Time 2 to abuse status at Time 1. The implications of this thesis are explored. Finally, the valuable forensic contributions that may be made by mental health experts in cases of alleged CSA are suggested. (PsycINFO Database Record (c) 2003 APA, all rights reserved)Author's email address