Base Rates of Negative Response Bias and Malingered Neurocognitive Dysfunction among Criminal Defendants Referred for Neuropsychological Evaluation

Abstract

Several studies reveal the prevalence of negative response bias (NRB) in civil forensic settings, but little NRB base rate information is available for criminal forensic neuropsychological settings. We reviewed the published literature on neuropsychological NRB in the civil setting. We then present data from 105 criminal defendants serially referred for neuropsychological assessment to determine the prevalence of NRB. The rate of NRB using one positive indicator was 89.5%. The rate was 70.5% when using two or more positive indicators and 53.3% for three or more indicators. Based on the Slick, Sherman, and Iverson (1999) classification for malingered neurocognitive dysfunction (MND), 19% were Valid, 26.7% were Possible MND, 32.4% were Probable MND, and 21.9% were Definite MND. The combined rate of probable and definite MND was 54.3%. Results suggest rates of neuropsychological NRB and malingering in criminal forensic settings are higher than in civil forensic settings.

ACKNOWLEDGEMENTS

Opinions expressed in this paper are those of the authors and do not necessarily represent the position of the Federal Bureau of Prisons or the US Department of Justice. Portions of this study were presented at the annual meetings of the National Academy of Neuropsychology in Seattle, WA, November, 2004, and Tampa, FL, October, 2005. Barry R. Ardolf is now at the Henry Ford Health System in Detroit, Michigan. Christi Houston is now at the Cox Health System in Springfield, Missouri. The authors would like to thank Glenn Larrabee, Ph.D., Kevin Greve, Ph.D., and Scott Millis, Ph.D., for their review and contributions to this work. Address correspondence to Robert Denney, Psy.D., US Medical Center for Federal Prisoners, P.O. Box 4000, Springfield, MO, 65801–4000; [email protected].

Notes

Note. PCCS–Persistent Post-Concussion Syndrome; PM–Probable Malingerers.

^a The study included one individual that was not a litigant but was included only because of exaggerated appearing test scores; removing this case would result in an actual consecutive series of N = 41 and NRB rate of 63.4%.

^b This series actually included 9% criminal defendants.

Note. Adapted from Slick et al. (1999). Diagnostic criteria for malingered neurocognitive dysfunction: Proposed standards for clinical practice and research. Clinical Neuropsychologist, 13(4), 545–561.

^a Defendant diagnoses included four severe gunshot wounds to the brain, one brain tumor, one neurocysticercosis, one severe learning disability, two chronic alcohol dependence with psychosis, one Capgras psychosis, one cerebrovascular accident with secondary mood disorder with psychosis, one mild traumatic brain injury combined with acute schizophrenia, one moderate traumatic brain injury, and one presumed Alzheimer's disease.

Note.

^a Valid × Definite, p = 0.001;

^b Possible × Definite, p = 0.001.

Note. Defendants with these diagnoses were also actively psychotic: Capgras Syndrome; one cerebrovascular disease; two chronic alcohol dependence; and one mild traumatic brain injury.

Note.

^a χ ²(6, n = 105) = 3.99, p = .68;

^b χ ²(6, n = 105) = 3.06, p = .80. Four cells (33.3%) had expected counts of less than five. Having cells with such low expected counts can spuriously increase sensitivity of Chi-Square; results were nonetheless non-significant.

Note. FBS – Fake Bad Scale; WAIS DF – WAIS-R or WAIS-III discriminant function; FOT – Finger Oscillation Test; HRNB – Halstead-Reitan Neuropsychological Battery discriminant function; VIP-NV – Validity Indicator Profile, nonverbal subtest; FCT-NV – Forced Choice Test of Nonverbal Ability; VIP-V – Validity Indicator Profile, verbal subtest; WMS-R DF – Wechsler Memory Scale – Revised discriminant function; CARB – Computerized Assessment of Response Bias; TOMM – Test of Memory Malingering; Abbreviated DMT – Abbreviated Hiscock Digit Memory Test; RCFT – Rey Complex Figure Test Atypical Recognition and Recognition Failure Errors; VSVT – Victoria Symptom Validity Test.

^a Number of positive test results over total number of defendants administered that measure.

^b Percentage of 105 defendants who were administered that measure.

^c Percentage of defendants scoring positive on that measure out of only those who were administered that measure. This score does not represent test sensitivity as not all defendants were necessarily considered malingering.

^d Indices include: Total Errors > 87; Errors on subtests I & II > 1; errors on subtest VII > 5; Bolter Items missed > 3; and Easy Items missed > 2.

^e Sum of left and right hand means.

^f Mean ungrouped time + Mean grouped time + Errors (i.e., no. of cards on which there was a miscount).

^g Accuracy score is the simple sum of words correctly recalled out of 15 possible.

^h We recalculated the rates using only Irrelevant and Suppressed criteria for invalid classification, considering Careless as a valid finding, as these patterns more clearly indicate motivation to perform poorly.

ⁱ  < 82.5% correct on Immediate Recognition, Delayed Recognition, or Consistency.

^j Total score = recall correct + (correct recognition – recognition errors).

^k  > 2 of any combination of Atypical Recognition Errors or Recognition Failure Errors.

^l Any score in the Questionable (random) or Invalid (below random) range based on the scoring program.

Note.

^a Includes those with below random performances (p < .05, one-tail) on forced choice procedures other than traditional NRB tests (e.g., SVT for remote memory).