Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Adolesc Health. Author manuscript; available in PMC 2012 July 1.
Published in final edited form as:
PMCID: PMC3286875

Restrictive Anorexia Nervosa And Set-Shifting in Adolescents: A Biobehavioral Interface

Elizabeth R. McAnarney, M.D., Jennifer Zarcone, Ph.D., Pamela Singh, B.A., Jennifer Michels, M.A., Sarah Welsh, M.D., Theresa Litteer, M.S., P.N.P., Hongyue Wang, Ph.D., and Jonathan D. Klein, M.D., M.P.H.



Set-shifting is a neurocognitive concept defined as the ability to switch tasks flexibly. Set-shifting scores are worse in adults with Restrictive Anorexia Nervosa (AN-R) than in controls. Adolescence is a developmental period when young people must respond flexibly to new situations. The purpose of this study is to compare the set-shifting scores of 24 adolescent females with AN-R and 37 matched normal adolescent controls (ages 14–20).


Methods included sociodemographic, psychological and biologic data, and neurocognitive testing using the Behavior Rating of Executive Function – Self- and Parent-Reports, the Cambridge Neuropsychological Automated Battery, and the Wisconsin Card-Sorting Test. Statistical analyses included t-tests, multiple analysis of variance, and correlations.


Study and control subjects were similar on sociodemographic data and intelligence quotient. There were differences in Body Mass Index and the Eating Disorder-3 evaluation. Significant differences in the composite score of set-shifting between the study and control groups were found using multiple analysis of variance.


Adolescent females with AN-R had significantly worse set-shifting scores than did control subjects. Future studies of adolescent AN-R subjects should include biologic (fMRI) and neurocognitive measures to determine the mechanisms at the brain-behavioral interface so that treatment can be directed specifically to set-shifting deficits.


Individuals who have restrictive anorexia nervosa (AN-R) are underweight, fearful of gaining weight, and have perceptual distortions of their body image. Set-shifting is a neurocognitive construct defined as the ability to switch tasks flexibly. Set-shifting scores are worse in adults with AN-R than in controls [1]. Examples of set-shifting statements from the Behavior Rating Inventory of Executive Function-Student (BRIEF-S) [2] are: “I have trouble getting used to new settings; I get upset with a change in plans.”

Studying set-shifting in adolescents with AN-R is important because: (1) Adolescence is a developmental period when young people must respond flexibly to multiple, new demands; inability to do so, may result in developmental challenges for the adolescent; (2) set-shifting capacity has not been examined in adolescents with AN-R; and (3) understanding the biobehavioral construct of set-shifting may provide a useful framework for future studies of the etiology and treatment of AN-R utilizing modern neurocognitive and neuroimaging methods. The effects of malnutrition on set-shifting appear to be minimal.

The hypothesis for this study is that adolescents with AN-R have worse set-shifting scores on formal neurocognitive testing than do normal adolescent controls.


The AN-R group consisted of 24 adolescent females, ages 14–20, who met the DSM-IV criteria for AN-R. Exclusions included the diagnoses of bulimia nervosa, eating disorder not otherwise specified, gastrointestinal conditions, major psychiatric conditions (schizophrenia, bipolar disorder, and severe depression), developmental conditions, and use of psychotropic medications. We chose to study only AN-R subjects as the neurobiologic bases of AN-R and bulimia nervosa may differ.

The control group consisted of 37 healthy females ages 14–20 with no history of eating disorders or other chronic illnesses. AN-R and control subjects were matched by group for age, ethnicity, socioeconomic status (health insurance status and parental education level), and intelligence quotient (IQ). Informed consent was obtained by a trained team member from both the adolescent and the accompanying parent. The Institutional Review Board at the University of Rochester Medical Center (URMC) approved this study.

Set-shifting was determined using the Behavior Rating Inventory of Executive Function, Self and Parent-Reports (BRIEF-SR and PR) [2], by the adolescent’s performance on both the computerized versions of the Wisconsin Card Sorting Test (WCST- Revised and Expanded) [3], and the Intra/Extradimensional Set Shift EDS subtest of the Cambridge Neuropsychological Test Automated Battery (CANTAB) [4]. In both the WCST and CANTAB, subjects identified rules and responded to stimuli based on those rules. Ability to respond appropriately to change in rules on these tests is related to their ability to switch tasks flexibly (set-shifting). Psychological evaluation included the Kaufman Brief Intelligence Test-2 [5], the Eating Disorder Inventory-3 (EDI-3) [6] and measures of depression and obsession-compulsion. Biological and sociodemographic data were also collected.

Multivariate analyses of variance (MANOVA) were conducted to identify significant differences between the two groups (AN-R and controls) in set-shifting outcomes: BRIEF-SR Shift Composite (composed of the Behavioral and Cognitive Shift subscales), BRIEF-PR Shift, number of Total and Perseverative Errors on the WCST, and the standardized number of Extra-Dimensional Stage Errors (EDS) on the CANTAB IED. Post-hoc analyses were carried out to examine individual measures using univariate, two-sample t-tests or Wilcoxon AN-Rk Sum test, where appropriate, using SAS 9.2 (SAS Institute, Cary, North Carolina).


Our study and control subjects’ sociodemographic and IQ scores were similar, whereas Body Mass Index (BMI), EDI-3, depression, and compulsion and obsession scores were worse for those with AN-R compared to control subjects (Table 1). Significant group differences on set-shifting between the AN-R and control groups were found using a MANOVA on the composite set-shifting score (Wilk’s λ = 0.77, F(1, 25.5) = 4.04, p < .006). Post-hoc analyses on individual scores are summarized in Table 2 with p-values from univariate, two-sample t-tests provided. Both BRIEF-SR composite and BRIEF-PR set-shifting scores were significantly different between groups (60.8 ± 11.0 vs. 48.4 ± 11.0, p<0.001 and 54.5 ± 9.6 vs. 47.1 ± 10.6, p< 0.008 respectively) (higher scores are worse scores) with both AN-R subjects and their parents reporting that AN-R subjects had more set-shifting problems than controls. On the WCST, the AN-R group had significantly fewer Total Errors (57.7 ± 10.6 vs. 63.0 ± 5.0, p=0.011, respectively) and nearly significantly fewer Perseverative Errors (60.5 ± 11.0 vs. 64.8 ± 6.4, p=0.057, respectively) on the WCST (lower scores are worse). There were no differences in EDS errors on the CANTAB. There was no relationship between nutritional status (BMI) and any of the individual set-shifting items.

Demographic and Psychological Characteristics of Study Population
Set-Shifting Assessments


These data support our hypothesis that adolescent females who have AN-R have worse set-shifting scores than do normal controls. Set-shifting deficits in adults with AN-R have been demonstrated with multiple neuropsychological tests including the WCST which we used. [1] Set-shifting deficits may be a marker for AN-R as it has been shown that women who have anorexia nervosa and their unaffected female siblings have worse set-shifting scores than do healthy controls, raising the question of whether set-shifting is an endophenotype for anorexia nervosa. [7]

Further understanding the cognitive and neural mechanisms of set-shifting deficits in adolescent AN-R subjects may provide insights into the etiology and treatment of this condition. In a recent study, 15 adult women with anorexia nervosa and 15 control women underwent event-related functional magnetic resonance imaging (fMRI) while performing a task that involved both cognitive (rigid approaches to changing rules) and behavioral (stereotyped or perseverative behaviors) components of set-shifting [8]. AN-R participants made significantly more errors in the behavioral component, but not in the cognitive component. Behavioral shifting was associated with reduced activation of the ventral anterior cingulate striato-thalamic loop and activation in the frontal and parietal brain regions on fMRI. Our results from the BRIEF-SR indicate that AN-R subjects had significantly worse scores on both the behavioral and the cognitive components of the BRIEF-SR Shift than did the controls (Table 2). We question whether the degree of an adolescent’s set-shifting deficits predicts poorer long- and short-term outcome. That is, is a higher score on set-shifting on initial presentation (more set-shifting deficits) associated with poorer outcome?

The strengths of this study include an exclusive focus on adolescent females who have AN-R, a matched healthy control group, and the use of behavioral and self-report evaluations, and parental measures of set-shifting deficits. Limitations of the study include the small study sample and minimal biological data.


Future set-shifting studies of AN-R subjects should use both behavioral and cognitive set-shifting evaluations simultaneously with functional resonance imaging (fMRI). Use of neuroimaging and computerized cognitive evaluations may clarify the mechanisms at the brain-behavioral interface that underlie this complex condition. Such technologies might also be used to test potential interventions to modify set-shifting deficits [9]. Additionally, we question whether scores on set-shifting screening predict outcomes of AN-R.


We acknowledge the members of the staff and the adolescents from the Adolescent Eating Disorder Center of the University of Rochester Medical Center (URMC) and from the Panorama Pediatric Group in Penfield, New York. In addition, we thank Drs. Constance Baldwin, Richard E. Kreipe, Thomas K. McInerny, Laura Jean Shipley and colleagues, Peter Szilagyi, and Jill Halterman, and Suzanne Bumpus, MS, FNP-C, Carole Berger, and Patricia Pincus and the staff of the Clinical Research Center of the University of Rochester Medical Center.


No reprints

Presented in part at the Pediatric Academic Societies’ meeting May 2, 2009 in Baltimore, Maryland

There is no conflict of interest, real or perceived, and the study was supported by a grant from the Strong Children’s Research Center of the Department of Pediatrics, University of Rochester Medical Center. This publication was made possible by Grant Number UL1 RR 024160 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Information on NCRR is available at

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Roberts ME, Tchanturia K, Stahl D, et al. A systematic review and meta-analysis of set-shifting ability in eating disorders. Psychol Med. 2007;37:1075–1084. [PubMed]
2. Guy SC, Isquith PK, Gioia GA. Behavior Rating Inventory of Executive Function: Self-Report Version: Professional Manual. Lutz, Florida: Psychological Assessment Resources, Inc; 2004.
3. Heaton RK, Chelune GJ, Talley JL, et al. Wisconsin Card Sorting Test manual: Revised and expanded. Lutz, Florida: Psychological Assessment Resources, Inc; 1993.
4. Luciana M. Practitioner review: Computerized assessment of neuropsychological function in children: Clinical and research applications of the Cambridge Neuropsychological Testing Automated Battery (CANTAB) J Child Psychol Psychiatry. 2003;44:649–663. [PubMed]
5. Kaufman AS, Kaufman NL. Kaufman Brief Intelligence Test. 2. Minneapolis, MN: Pearson Assessments; 2004.
6. Garner DM. Eating Disorder Inventory-3: Professional Manual. 3. Lutz, Florida: Psychological Assessment Resources, Inc; 2004.
7. Holliday J, Tchanturia K, Landau S, et al. Is impaired set-shifting an endophenotype of anorexia nervosa? Am J Psychiatry. 2005;162:2269–2275. [PubMed]
8. Zastrow A, Kaiser S, Stippich C, et al. Neural correlates of impaired cognitive-behavioral flexibility in anorexia nervosa. Am J Psychiatry. 2009;166:608–616. [PubMed]
9. Tchanturia K, Davies H, Campbell IC. Cognitive remediation therapy for patients with anorexia nervosa: Preliminary findings. Ann Gen Psychiatry. 2007;6:14–20. [PMC free article] [PubMed]