Malingering Probability Scale

The Malingering Probability Scale (MPS) was developed in the mid-1990s as a self-report instrument to estimate the probability of malingering, taking account of base rates in underlying populations. The instrument was based on three premises: (1) that psychopathology expresses itself in clearly defined syndromes for which certain symptoms, though perhaps plausibly related, have very low likelihoods of occurrence; (2) that patients will vary in the type of symptoms they feign depending on the context; and (3) that the identification of probabilities of any given individual of malingering should be adjusted according to the base rate of malingering in the population from which he or she was drawn. The MPS was developed by Leigh Silverton, who also designed and conducted many studies; analyzed much of the data; and, in collaboration with Chris Gruber, wrote the MPS manual.

At the time the MPS was developed, there were no other instruments of malingering that provided sensitivity or specificity studies or gave probabilities of malingering predicated on estimates of base rates in the population from which the patient was drawn. The F scale of the Minnesota Multiphasic Personality Inventory (MMPI), which was most widely used, did not distinguish a true-focused response set or random responding. In constructing the MPS, Silverton also tried to address a different type of feigned psychopathology than had been traditionally covered. Past instruments in wide usage had focused on bizarre and psychotic symptoms. Psychotic symptoms tend to be feigned in criminal contexts in which punishment for mentally competent persons judged responsible for their crimes is more aversive than incarceration in a mental hospital. The focus of the MPS is broader, comprising both psychotic symptoms and nonpsychotic symptoms of the type that might be feigned in civil contexts.

Civil litigants should be more apt to feign trauma-related symptoms such as those associated with post-traumatic stress disorder, depression, and dissociation. In civil cases, where money damages are the remedy for a psychological injury, an experienced attorney understands that certain disorders are most likely to stem from trauma and thus yield the highest rewards. Whether through honest questioning by a personal injury attorney attempting to explore damages, through outright coaching, or by self-study of diagnostic material, a litigant may obtain an impression, if not a textbook definition, of the trauma-related syndrome that he or she should emulate to maximize rewards. In such a case, a patient may avoid endorsing bizarre delusional or hallucinatory symptoms represented by the F scale of MMPI-2 or the M test but may endorse posttraumatic stress-like symptoms.

The pseudoclinical items of the MPS cover symptoms related to trauma as well as those related to psychotic phenomena as might be feigned in a criminal context. The symptoms might appear, to the sophisticated faker, to reflect depression, dissociation, post-traumatic disorder, and schizophrenia. Silverton wrote items that seemed, based on the literature and her clinical experience, to reflect genuine psychopathology and those that would appear to reflect genuine psychopathology but did not. One such pair of items for the depression scale is as follows: “I am rarely awakened by sad dreams” F (pseudoitem) and “I sleep well” F (actual item). Depressed people tend to have trouble sleeping and may have sad dreams but are rarely awakened by them. Items such as these were derived rationally and then validated to arrive at a 139-item instrument to detect malingering.

A standardization sample of 1,016 adults, aged 17 and above, was selected from four regions of the United States and tested between 1995 and 1996. The sample was large, contemporary, and nationally representative. Although the sample matched the 1994 U.S. census data adequately in ethnic distribution, the actual number of ethnic minority participants was small, a fact that suggests caution in interpreting results from minorities. Reading level was measured at third to fourth grade, which should be adequate for most populations.

There are reasonably good levels of reliability, using measures of internal consistency, test-retest reliability, and temporal stability and using samples of individuals from the general population in which the malingering base rate is relatively high (prison and civil forensically referred samples) and individuals referred for clinical but nonforensic evaluations. The reliability compares favorably with the MMPI clinical scales, which were in the .70 to .90 range.

Validation studies were performed in a number of contexts. Paradigms were employed to detect dissimulation (for college students and prison inmates) and guided dissimulation in which the subjects were given the Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association criteria for the disorders. Scores were also obtained from samples of inpatients and referred forensic and clinical outpatients. These procedures yielded Malingering (MAL), the scale used to detect malingering.

Test interpretation starts with the Inconsistency (INC) scale. A high INC score indicates that the subject was so inconsistent that the test should not be interpreted. This could be due to a poor reading level or conscious resistance to test taking. If the INC elevation is less than a T-score of 70, it is possible to interpret the MAL scale. In the discussion of MAL or the malinger-ing scale of the MPS in the manual, the authors tried to convey the notion that the conclusion that a person is malingering is a probability statement. The probability relates directly to the base rate of the underlying construct and would be expected to vary from sample to sample. For instance, if a person obtains a MAL T-score of 73 where the base rate of malingering is assumed to be 50%, 20%, or 10%, the concluded probability of malingering is 72%, 38%, or 22%, respectively. An assessor interpreting the test report should, therefore, be able to explain his or her assumptions about base rates in the population he or she tests.

One great advantage of the instrument is also one of its greatest disadvantages: The scoring of the individual items has been protected. The item scoring has not been printed in any publication and has been further protected by providing the test in computer-scored form only. (The Western Psychological Services does, however, make available the individual items and their scoring to qualified persons who apply.) An advantage of this feature is that it is impossible for a prospective subject to study for and, therefore, foil the exam. A disadvantage of this feature is that the test cannot be easily and quickly scored by clinicians and researchers. This may discourage research on this instrument, a particularly important problem for validating the experimental clinical scales.

One advance of this instrument is that it broadened the realm of feigned psychopathology covered in malingering instruments, particularly to include those symptoms related to trauma to capture the more sophisticated feigner. But perhaps the best advance in the construction and presentation of this instrument is the conceptualization of malingering as a probability statement, which depends on the assumptions the diagnostician makes about the base rates of malingering in his or her clinical sample. These are assumptions that are all too often ignored by the forensic clinician, are rarely questioned by even the skilled cross-examiner, and yet are critical to the trier of fact where mental state is at issue.


  1. Fulero, S. (2002). Review of the malingering probability scale. In B. S. Plake & J. C. Impara (Eds.), The fourteenth mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements.
  2. Silverton, L., & Gruber, C. (1998). Malingering Probability Scale (MPS) manual. Los Angeles: Western Psychological Services.

Return to Criminal Responsibility assessment overview.