Do We Remember Our Childhoods Accurately?

Childhood bipolar disorder as recollected by adults

Posted Sep 23, 2012

In the absence of evidence to document the existence of bipolar disorder in children, those who espouse it  often use studies of the childhood memories of adults with bipolar disorder in an effort to document the presence of the disorder in childhood.  As recently as June, 2012, an editorial in the prestigious American Journal of Psychiatry referred to a study using this dubious methodology as its sole evidence for the existence of bipolar disorder in childhood.

Asking adult bipolar disordered patients about their childhood memories of their mental illness is an error laden approach to gathering data about the childhood precursors of adult bipolar illness.  To illustrate the contrast between the results of asking adult bipolar disordered patients about their childhood bipolar disorder and more careful approaches, one such study will be contrasted with the results of two longitudinal studies.

In a study of the recollections of adult bipolar disordered patients of their bipolar disorder during childhood, 983 adult patients with a diagnosis of bipolar disorder were interviewed about their recollections of their bipolar disorder during childhood: 272 of the 983 adult patients (27.6 percent) reported the onset of their bipolar disorder to have been before the age of 13 years; 370 of the 983 adult patients (37.6 percent) reported the onset of their illness to have been between the ages of 13 and 18 years. The average age of patients who remembered the onset of their bipolar disorder as having been below the age of 13 was 38 years old.  This is a long time for the accurate recollection of the date of onset of a psychiatric illness. Contrasting with the above study of the childhood memories of adult bipolar disordered patients, is the highly regarded and carefully done Dunedin Multidisciplinary Health and Development study. (Kim-Cohen, J. et al. Prior Juvenile Diagnoses in Adults with Mental Disorder: Developmental Follow-back of a Prospective-Longitudinal Cohort. Arch Gen Psychiatry.  Vol 60, July 2003). The psychological development of 1037 representative three year olds in New Zealand followed to age 26 years and beyond was carefully scrutinized.  The participants were interviewed approximately every two years.  At age 26, 48.2 percent of the group were found to have a psychiatric diagnosis. These adult psychiatric diagnoses were compared to the psychiatric diagnoses of the participants beginning at age 11; 73.9 percent of those diagnosed at age 26 had a psychiatric diagnosis at age 18 or younger.  Specific to bipolar disorder, diagnoses of mania (a necessary criterion for bipolar disorder) were not found below the ages of 18.  At age 26 years, 3 percent had had a prior manic episode and of this 3 percent of adult bipolar-disordered patients, 93.1 percent had had a psychiatric diagnosis prior to the age of 18, but the prior diagnosis was not mania or bipolar disorder. And 22 percent had had conduct disorder (delinquency) or oppositional defiant disorder during childhood or adolescence and 13 percent had had a diagnosis of depression before the age of 18 years. Delinquency, oppositionality, and depression were significant childhood precursors to bipolar disorder in adults.  No cases of bipolar disorder in childhood were found. These findings of this carefully done study contrast greatly with the Perlis et al study of adult bipolar disordered patients’ recollections of bipolar disorder in childhood.

A second report from the Dunedin study illustrates the obstacles to accurate memory of psychological events during childhood.  (Henry, B., On Remembrance of Things Past: A Longitudinal Evaluation of the Retrospective Method. Psychological Assessment, 1994 Vol. 6 pp 92-101). Eighteen-year-olds from the Dunedin study described above were asked to remember a variety of events during their life. The information had already been collected over time. The purpose of the study was to gain a measure of how well 18-year-olds remembered these events. The events included objective occurrences such as the number of moves made during childhood and subjective occurrences of interest to mental health personnel such as episodes of depression and anxiety.    Objective events such as moves were remembered relatively well but subjective events such as anxiety were poorly recollected.  Subjective events had no significant relationship to what had actually occurred during childhood. There were no recollections of mania in childhood, but the lack of correspondence between the emotions of childhood reported by adults and the childhood emotions actually observed suggest that reports of mania during childhood from adult bipolar disordered patients should be received with skepticism.

The unreliability of memory in reports of adults about their childhood is a well known phenomenon.  Retrospective reports of adult bipolar disordered patients about their childhoods do not merit the scientific status that has been given to them.

Copyright Stuart L. Kaplan, M.D., 2012.

Stuart L. Kaplan, M.D., is author of Your Child Does Not Have Bipolar Disorder: How Bad Science and Good Public Relations Created The Diagnosis, available on

Be sure to read the following responses to this post by our bloggers: