Malingering of Psychiatric Disorders and Cognitive Impairment In Criminal Court Settings
Published in The Prosecutor, 38(5), Sept.-Oct. issue, 2004

Dr. Steve Rubenzer is a clinical and forensic psychologist in Houston, Texas. He has had eight years experience as a court-appointed examiner in the criminal justice system. He can be reached at (281) 481-5715 or srubenzer@earthlink.net.

Defendants who successfully feign mental impairment take advantage of society's compassion for the mentally ill and create skepticism towards those that are truly incompetent or insane. They create substantial additional costs for the court system for transportation, psychiatric care, and attorneys' and court fees. Witnesses may become unavailable and the chances for a successful feigned insanity defense or other favorable outcome may improve after confinement in a mental hospital. But malingering may constitute obstruction of justice and has been upheld as a basis for a sentencing enhancement by the Court of Appeals for the Fifth Circuit (U.S. v. Greer, 158 F.3d 228 (1998)). Lastly, malingerers may create management problems for psychiatric staff at the facilities where competency restoration treatment is provided. They may be career criminals who, when placed in a supportive but dull psychiatric environment, disrupt the ward routine and take advantage of legitimate psychiatric patients. Good public policy demands that they not be allowed to disrupt justice administration and impose the above-mentioned costs on society when competent assessment at the pretrial stage can prevent it.

The insanity defense has occasionally been a tempting defense for criminals seeking to avoid criminal responsibility. Defendants may also feign incompetence to stand trial, or, most commonly, fake mental illness in order to receive a mitigated punishment at sentencing. With the recent Supreme Court decision ¹ prohibiting the execution of the mentally retarded, there is fresh incentive for criminals to feign a mental disease or defect. But how can prosecutors ensure that defendants do not get away, literally, with murder, by faking it?

Surveys of psychiatrists and psychologists who work in forensic settings report that malingering (faking or exaggeration to avoid negative consequences) occurs in 16-18 percent of persons who present as significantly impaired.² This is almost certainly an underestimation, since those who are successful in their performance would not be counted, and there is little evidence that clinicians-unaided by specialized tests-can reliably distinguish malingerers from persons actually suffering from a mental disease or defect. One recent study found that psychiatrists working in a state forensic facility, relying on interviews and file data, failed to identify 50 percent of malingerers detected through specialized testing.³ The misidentification rate among clinicians in non-forensic settings who provide treatment (as opposed to assessment) is likely to be much higher.

The American Psychiatric Association's Diagnostic and Statistical Manual clearly states, "Malingering should be ruled out in those situations in which financial remuneration, benefit eligibility, and forensic determinations play a role."⁴ However, clinicians may not know that the patient even has such motivations, often do not suspect the possibility of malingering, and typically lack the training or tools to assess for malingering even if they suspected it. Not surprisingly, they rarely find it. In contrast, forensic psychologists consider malingering assessment a crucial element of their craft and routinely test for it. Because this potentially places the examiner in opposition to the examinee's interests, evaluation in forensic settings is viewed as a professional specialty that is incompatible with providing treatment.⁵ Other differences between forensic evaluators and the treating clinicians are summarized in Table 1. ⁶

Table 1

	Therapists	Forensic Examiners
Who is the client?	Patient	Attorney or the court
Goals	Provide treatment and support	Objectively evaluate a defendant or claimant
Data	Accept what the client says	Corroborate examinee's statements with collateral information
Emphasis	Treatment, "helping"	Assessment of psycholegal issue at stake
Trust	Assume basic honesty	Assess for malingering or attempts to create a positive impression
Accountability	Anticipate little challenge to conclusions, diagnoses	Anticipate cross examination, consider alternative hypotheses, explanations
Privilege	Governed by therapist-client privilege	Governed by attorney-client privilege, if any
Knowledge of legal issues	May be unaware of legal standards, rules of evidence	Familiar with case law governing the issue to be addressed, Daubert and FRE standards for evidence
Attitude	Dislike, fear Avoid court appearances	Accept legal proceedings as part of their work; develop testimony skills

There are several types of malingering. Less sophisticated persons may present as globally impaired, with both severe cognitive deficits and psychiatric symptoms. More sophisticated feigners may be more specific. For example, they may present as psychotic but cognitively intact, or vice versa. Some may present with only a very specific deficit, such as amnesia, incompetency to stand trial or hallucinations. Often, the more general the malingering, the easier it is to identify and substantiate, both through testing and through collateral sources. However, techniques to assess malingering of specific disorders and symptoms are rapidly being developed.

Assessment
Skepticism is an essential quality of a forensic examiner, but it should never be a substitute for data. Many treating professionals lack a critical mindset; in contrast, some forensic examiners become over-zealous. People with real psychiatric conditions may exaggerate. And the presence of feigning or poor effort cannot automatically be assumed to be malingering, since motivations other than material gain may be present. For example, in factitious disorders, persons adopt a sick role not for monetary benefits, but because it results in care-taking and nurturance from medical personnel or family members: they like the way people treat them when people think they are sick.

Four sources of information are important in determining the possibility of feigning: 1) Semi-structured interviews that cover various aspects of the evaluatee's life; 2) the examiner's observations of manner, both during the interview and outside of it, 3) specialized psychological testing, and 4) collateral information from family members, treatment providers, and jail personnel, and, possibly, 911 tapes or reports from arresting officers and others surrounding the relevant offense giving rise to the evaluation.

Interviews and Observations
Interviews are important to identify collateral sources and to obtain a gauge of the person's credibility. Not uncommonly, a defendant may tell the forensic examiner he left school in the sixth grade after telling jail medical or administrative staff that he completed high school. Credibility issues may also arise in the forms of inconsistencies or the manner in which information is relayed. Psychologists are increasingly identifying behavioral indications of deceit, and it appears possible to train people to be reasonably good lie detectors. ⁷ Some of the best evidence of malingering occurs when the examinee's demeanor changes as he enters or leaves the examination area. At the Harris County Jail, defendants can often be seen talking in an animated manner with four peers in their jail tank before being called for the assessment. Yet, when they meet the examiner at the door, they may cross their eyes and walk past the examiner as if blind.

As in the previous example, many malingerers badly overplay the part, and this may give them away quickly to a seasoned evaluator. Often, malingerers are unfriendly or subtly uncooperative. They may give very brief, unelaborated answers and be reluctant to volunteer information. Sometimes they refuse to sign information release forms so medical records can be obtained, or say they don't know phone numbers of family members.

Before standardized tests were widely available, practitioners improvised a number of approaches to help detect malingerers. For example, they might ask very easy questions that virtually anyone, whether mentally ill, brain damaged, or mentally retarded, should know (e.g., What color is grass?). Psychiatrist Phillip Resnick has written extensively on how persons fake psychosis, detailing the ways in which reports of malingered hallucinations are likely to differ from legitimate ones.⁸ Often, malingerers will report that they have experienced hallucinations "all my life" and experience them "all the time." This is highly atypical for actual psychiatric patients. There are many ways, some subtle, in which a dishonest defendant may become apparent while discussing psychiatric problems if the examiner is attuned to the possibility of faking.

Psychological Testing
Formal, specialized testing is increasingly recognized as important in cases involving claimed psychiatric or cognitive impairment for two reasons: effectiveness and accountability. As indicated earlier, there are many reasons to doubt the ability of most clinicians to separate false from legitimate claims. Further, it is unclear if clinical judgment and unstandardized tests are sufficiently reliable or accepted under FRE and Daubert/Kuhmo. Almost by definition, neither a clinician's judgment nor unstandarized test results will be subject to peer review and rates of error will be unknown.

Feigning of Psychiatric Illness
Most of the traditional psychological tests, such as IQ tests and the Rorschach, are of little use in assessing malingering.⁹ However, the Minnesota Multiphasic Personality Inventory (MMPI; both original and MMPI-II) is useful. Both versions have numerous scales and indices to detect inconsistent responding, defensiveness, or exaggeration/feigning. In fact, scales from the MMPI have shown the greatest resolution of any test or index in separating simulators from actual patients. However, there is a major drawback: the cut scores that are most effective at separating feigners from honest responders have varied greatly from study to study. In other words, although the indices clearly have good validity, the criterion for deciding whether someone is feigning or not has been unreliable.¹⁰

Two new generation tests compete with the venerable MMPI, whose roots go back over sixty years. The Personality Assessment Inventory is a sophisticated, new generation psychopathology inventory that also has indices to detect feigning. Unlike the MMPI, its cut scores have been cross-validated (both with suspected malingerers and those asked to simulate malingering) and have remained fairly stable. However, it is not very sensitive to feigning of nonpsychotic disorders, such as anxiety and depression. This is typically not a problem in criminal justice settings, as most defendants, if they fake, feign psychosis.

The Millon Clinical Multiaxal Inventories (I-III) have validity scales, but these appear less effective and are not as well validated as those for the MMPI and PAI and should not be relied upon. ¹¹

The best-regarded assessment instrument for feigned psychopathology is the Structured Inventory of Reported Symptoms (SIRS). It is administered as a structured interview, with the examiner scoring each response after the question is read to the examinee. There are eight primary scales reflecting different strategies that may be employed to create the impression of severe psychopathology. The judgment of feigning may be based on an extreme elevation on a single score, the combination of three moderately high scores, or on the total number of times endorsed. The decision rules are stringent, as false positive errors (labeling a legitimate psychiatric patient as a malingerer) are viewed as very undesirable. A positive finding indicates a very high probability of exaggeration or feigning (98 percent or better depending on the index). However, sensitivity suffers under these rules, and the SIRS produces a "Definite Malingering" interpretation in only about 50 percent of those eventually identified as such. The SIRS has been cross-validated several times and been shown capable of detecting nonpsychotic conditions (e.g., feigned depression) and to be resistant to coaching.¹² The SIRS takes about 45 minutes to an hour to administer and another 10-15 minutes to score. Since clerical errors are possible it should be scored twice.

Malingering of Mental Retardation or Brain Damage
In the early days of neuropsychological testing, it was believed that it would be impossible to produce a credible profile on the numerous tests that make up a typical assessment battery. However, this belief was shattered in 1978, when a classic study found neuropsychologists unable to identify malingerers at a level that much exceeded chance.¹³ A subsequent study found not one in 42 neuropsychologists identified profiles of children instructed to malinger, and that the clinicians expressed high confidence in their erroneous findings.¹⁴ Since these embarrassments, neuropsychology has developed many procedures to detect cognitive malingering. Nonetheless, clinicians may not use them regularly, particularly if they lack training in testing.

The impact of feigning or poor effort can be dramatic: A recent study found that level of effort (as gauged by a test designed for this purpose) accounted for fully half of the variance in neuropsychological test scores, eclipsing the effect of injury severity:¹⁵ Persons who put forth good effort scored better than those who didn't regardless of the seriousness of their injury. Another study found effort level accounts for four and half times as much variance as actual injury.¹⁶

Measures of cognitive malingering have been around for years. Many rely on the fact that feigners produce implausibly low scores on easy tasks. The Rey 15 Item Test consists of a 3 X 5 array of numbers and letters in sequence. Recent studies have shown that this test lacks both sensitivity (ability to detect malingerers) and specificity (passes legitimate patients), and its continued popularity and acceptance ¹⁷ seem unwarranted. It may be most useful when the examinee produces highly unusual responses that are not found in legitimate neuropsychiatric patients.

Another procedure is the Dot Counting Test, which examines the examinee's performance in counting both organized and disorganized groups of dots. Optimal scoring rules can produce about 80 percent correct classification.

Two more recent tests represent the state of the art. The Test of Memory Malingering (TOMM) is widely respected¹⁸ and appears to be unaffected by depression or other psychiatric conditions. However, two limitations have become apparent. The usual cut score is too high for persons with dementia, and some will be falsely identified as feigning. Secondly, the TOMM may be substantially less sensitive than its rival, the Word Memory Test (WMT).

A recent study found the WMT identified 37 percent of a population involved in litigation as feigning, while the TOMM identified only 12 percent.¹⁹ The WMT authors have shown the WMT to be very easy and to produce very minimal false positives among a diverse range of groups, including the brain injured, depressed, and children.²⁰ Ironically, samples of normals actually produce higher rates of false positives- apparently because some don't put forth good effort. The WMT is explicitly billed as a measure of suboptimal effort, which is broader than the construct of malingering, which connotes intentional failure. Lastly, the WMT appears resistant to detection as a malingering measure, and those instructed to try to beat it have been unable to do so. ²¹

Malingering Specific Deficits
Criminal defendants may limit their malingering to very specific symptoms, such as amnesia, or to a particular legal issue, such as sanity at the time of the offense. This can make assessment more difficult, as most traditional assessment devices focus on feigning of psychosis at the time of testing. However, psychologists have developed some techniques that can target malingering in specific areas. Feigned amnesia can be detected by creating a forced choice memory test. In the simplest version, the examiner assembles perhaps twenty facts that the perpetrator should know and remember from the crime. Two equally plausible answers are generated for each question. Someone with no knowledge of the crime should score 50 percent on such a test, as blind guessing would lead one to be right half the time. A malingerer may overplay the part and score below chance to a statistically significant level, thus implying concealed knowledge of the facts. A recent, more complicated version of this technique resulted in identification of 59 percent of experimental subjects recruited to commit a mock crime and lie to investigators.²² Other defendants may simply drop the act when faced with such a task, as one did when tested by me. Lastly, depending on the legal issue and state law, amnesia may be irrelevant. In Mendenhall v. State of Texas (77 S.W.3d 815, 816 [Court of Criminal Appeals, 2002]), the court held that amnesia might negate mens rea or actus rea. However, they further wrote:

… [I]s the insanity defense available to a defendant who was unconscious or semi-conscious at the time of the alleged offense, so that it might be said of him that he did not know his conduct was wrong only because he did not consciously know of his conduct at all? We conclude the answer to that question is "no."

We have carefully reviewed the legislative history of § 8.01(a), and nothing in it suggests that any legislators intended for the insanity defense to apply to persons who were unconscious or semi-conscious at the time of the alleged offense. ²³

A very recent development is a scale developed to gauge malingering of incompetence to stand trial. The Atypical Presentation Scale (APS) is a companion measure to the Evaluation of Competency to Stand Trial-Revised,²⁴ a standardized competency interview. The APS asks the defendant questions about his alleged symptoms and how they interfere with his perceptions and ability to function in court. Originally presented merely as a screening measure for feigned incompetence, a recent study found the APS highly effective at separating suspected malingers from defendants with legitimate psychiatric disorders. ²⁵

Collateral Sources
Collateral sources are crucial in assessing both malingering and the actual level of functioning or impairment. Traditionally, clinicians have relied primarily on the patient's report, as objective truth was of less concern than the patient's view of the situation. However, such sources are essential for forensic psychological assessment.²⁶ These may include medical, school, financial and employment records, family members, supervisors, previous friends, employers or co-workers, and observations outside of the examination room. Information from law enforcement surrounding the details of the offense giving rise to the evaluation is also essential, since it will contain reports from third-party eyewitnesses about the defendant's conduct.²⁷ The more impairment displayed, the greater the burden to keep up the act over time and different settings. There will be strong temptation to drop the act when the person believes he is not being observed, or when an attractive staff member of the opposite sex wants to talk.

Although generally more reliable than the defendants themselves, collateral sources cannot be taken at face value. Family members may lie to assist the defendant's malingering. Treatment providers may be manipulated to provide credible, sincere but false testimony about the extent of the defendant's mental impairment. Jail personnel may be invested in seeing defendants punished. Records can be incomplete and even wrong. In forensic assessment, virtually no source of information should be assumed reliable.

Conclusion
Detecting malingering is not difficult if an examiner is willing to look for it and has appropriate tools. It is no longer acceptable practice for those doing forensic assessment to accept a defendant's presentation at face value, and evaluations that do not assess the possibility of malingering (when significant impairment is presented) are virtually worthless. Such examiners are opening themselves to being manipulated by criminal defendants. At the same time, the presence of exaggeration or even faking cannot definitively rule out the presence of a mental disorder or the possibility the person might be incompetent (or legally insane at the time of the offense). Information about level of functioning in other settings can provide indirect data on these issues.

¹ Atkins v. Virginia,(00-8452) 536 U.S. 304 (2002), 260 Va. 375, 534 S. E. 2d 312, reversed and remanded. http://supct.law.cornell.edu/supct/html/00-8452.ZS.html

²Richard Rogers, Clinical Assessment of Malingering and Deception (2nd Ed.) 4 (1997).

³H. A. Miller, "The Miller-Forensic Assessment of Symptoms Test (M-FAST): Test Generalizability and Utility Across Race, Literacy, and Clinical Opinion," Journal of Criminal Justice and Behavior (in press).

⁴ American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 467 (1994).

⁵S.A. Greenberg & D. W. Shuman, "Irreconcilable Conflict Between Therapeutic and Forensic Roles," 28 Professional Psychology: Research and Practice, 50 (1997).

⁶Adapted from S.A. Greenberg & D. W. Shuman (1997).

⁷S. Porter, M. Woodworth, & A. R. Birt, "Truth, Lies, and Videotape: An Investigation of the Ability of Federal Parole Officers to Detect Deception," 24 Law and Human Behavior 643 (2000).

⁸P. Resnick, "Malingered Psychosis," In R. Rogers (Ed.), Clinical Assessment of Malingering and Deception (2nd Ed.) 47 (1997).

⁹D. J. Shretlen, "Dissimulation on the Rorschach and Other Projective Measures," In R. Rogers (Ed.), Clinical Assessment of Malingering and Deception (2nd Ed.) 208 (1997).

¹⁰R. Rogers, K.W. Sewell, & R. Salekin, A Meta-Analysis of Malingering on the MMPI-2, 1 Assessment, 227, 234 (1994).

¹¹R. Rogers & S. D. Bender, "Evaluation of Malingering and Deception." In A. M. Goldstein and I. B. Weiner (Eds.), Handbook of Psychology, VOL. 11, Forensic Psychology, 109 (2003).

¹²R. Rogers, J. R. Gillis, & M. R. Bagby, "Detection of Malingering on the Structured Interview of Reported Symptoms (SIRS): A Study of Coached and Uncoached Simulators," 3 Psychological Assessment, 673 (1991).

¹³R. K. Heaton, H. H. Smith, R. A. Lehman, & A. T. Vogt, "Prospects for Faking Believable Deficits on Neuropsychological Testing," 46 Journal of Consulting and Clinical Psychology 892 (1978).

¹⁴D. Faust, K. Hart, T. J. Guilmette, & H. R. Arkes, "Neuropsychologists' Capacity to Detect Adolescent Malingerers," 19 Professional Psychology: Research and Practice, 508 (1988).

¹⁵P. Green, P. R. Lees-Haley, L. M. Allen, III, "The Word Memory Test and the Validity of Neuropsychological Test Scores," 18 Archives of Clinical Neuropsychology, 1 (2003).

¹⁶P. Green, M. L. Rohling, P.R. Lees-Haley, & L. M. Allen, III, "Effort Has a Greater Effect on Test Scores than Severe Brain Injury in Compensation Claimants," 15 Brain Injury 1045 (2001).

¹⁷S. J. Lally, "What Tests are Acceptable for Use in Forensic Evaluations? A Survey of Experts," 34 Professional Psychology: Research & Practice, 491 (2003).

¹⁸S. J. Lally, Ibid.

¹⁹R., Gervais, M. Rohling, P. Green, & W. Ford, "A Comparison of WMT, CARB, and TOMM Failure Rates," 19 Archives of Clinical Neuropsychology 475 (2004).

²⁰P. Green, G. L. Iverson, L. Grant L & L. Allen, "Detecting Malingering in Head injury Litigation with the Word Memory Test," Brain Injury. 13(10), Oct 1999, 813-819.; P. Green, & L. Flaro, "Word Memory Test Performance in Children," 9 Child Neuropsychology

189 (2003). ²¹T. M. Dunn, P. K, Shear, S. Howe, & M. D. Ris, "Detecting Neuropsychological Malingering: Effects of Coaching and Information." 18 Archives of Clinical Neuropsychology 121 (2003). ²²M. Jelicic, H. Merchkelbach, & S. van Bergen, "Symptom Validity Testing of Feigned Amnesia for a Mock Crime," 19 Archives of Clinical Neuropsychology, 525 (2004).

²³There is a certain amount of confusion when it comes to the "I Blacked Out" defense. There is a difference between (1) committing an act and then forgetting the act later, because of voluntary intoxication or some real or imagined mental disease or defect, and (2) committing an act while unconscious. If you force the malingerer to articulate which it is, the issue of malingering is easier to discern. A defendant, for example, who leaves a bar, goes into a parking lot, correctly identifies his own car, finds his keys, unlocks the door, gets in the correct seat in the car, backs out of the parking space, drives to the highway, where he commits vehicular homicide, or continues homeward where he bludgeons his wife to death-is any fact-finder really going to believe that this defendant was unconscious during the offense? A defendant who is forced to articulate that, during the whole evening, his memory and functioning were intact, if not impaired by alcohol or drugs, except for the very few minutes or fraction of an hour in which he bludgeoned his wife to death, faces more legitimate, and effective, skepticism than a defendant who muddies up the issue of alcoholic blackout and unconsciousness during the offense. The fact that the defendant had an alcoholic blackout after the vehicular homicide, or domestic assault, actually makes the prosecution's case stronger in the case of the vehicular homicide: the defendant was really drunk-not just mildly intoxicated. In the case of the bludgeoning, forcing the defendant to articulate the extent of his blackout makes him more susceptible to inquiry, since blackouts are rarely that discrete, or brief. An expert would need to explain to the judge or jury that alcoholic blackouts have to do with an inability to form memory after the fact, not an inability to form criminal intent during the crime. But ascertaining the precise scope of inquiry, and forcing the defendant to specify the nature and extent of his defect, is of great benefit to the truth-finding process.

²⁴R. Rogers, C. E. Tillbrook, & K. W. Sewell, Evaluation of Competency to Stand Trial-Revised Professional Manual. Psychological Assessment Resources, Inc. (2004).

²⁵R. Rogers, R. L. Jackson, K.W. Sewell, & K.S. Harrison, "An Examination of the ECST-R as a Screen for Feigned Incompetency to Stand Trial," 16 Psychological Assessment, 139 (2004).

²⁶Gary B. Melton, John. Petrilla, Norman .G. Poythress, & L.A. Slobogin, Psychological Evaluations For The Courts: A Handbook For Mental Health Professionals and Lawyers (2nd Ed.) 43 (1997).

²⁷Frequently forensic evaluators who are screening defendants in anticipation of competency to stand trial, sanity at the time of offense, or issues of mitigation of punishment, seek information from prosecutors with regard to the details of the offense, in order to gain relevant information about the defendant's behavior. Of course, information that is shared with the evaluator may be shared with the defendant or his attorney, so prosecutors must weigh sharing information with the defense that is not even discoverable against withholding information from an honest evaluator. If the sense is that the evaluator is a "hired gun," providing such information will not change the evaluator's conclusions and may enable him to prepare for your cross-examination more thoroughly. On the other hand, a truly neutral evaluator may take information gleaned from the prosecutor-say, the 911 tape where the defendant is heard calmly and accurately reciting a lengthy address and giving one false explanation for the victim's injuries - and compare it to the defendant's story to him of a frenzied attack that took place without premeditation or intent, and come to the conclusion that the defendant was sane at the time of the offense and currently malingering and lying to avoid responsibility.