The Relationship Between Psychiatric and Cognitive Symptom Feigning Among Forensic Inpatients Adjudicated Incompetent to Stand Trial

Abstract

The accurate assessment of feigning is an important component of forensic assessment. Two potential strategies of feigning include the fabrication/exaggeration of psychiatric impairments and the fabrication/exaggeration of cognitive deficits. The current study examined the relationship between psychiatric and cognitive feigning strategies using the Structured Interview of Reported Symptoms and Test of Memory Malingering among 150 forensic psychiatric inpatients adjudicated incompetent to stand trial. A greater number of participants scored within the feigning range on the Structured Interview of Reported Symptoms than on the Test of Memory Malingering. Relative risk ratios indicated that individuals shown to be feigning cognitive deficits were 1.68 times more likely to feign psychiatric symptoms than those not shown to be feigning cognitive deficits. Likewise, individuals shown to be feigning psychiatric deficits were 1.86 times more likely to feign cognitive deficits than those not shown to be feigning psychiatric symptoms. Overall, findings suggest that psychiatric feigning and cognitive feigning are related, but can be employed separately as feigning strategies. Therefore, clinicians should consider evaluating for both feigning strategies in forensic assessments where cognitive and psychiatric symptoms are being assessed.

Keywords

psychiatric feigning feigning cognitive impairment malingering SIRS TOMM

Malingering during a forensic evaluation frequently takes the form of “overstated pathology,” which could include the exaggeration or feigning of symptoms of a mental disorder (Rogers, 2008a, p. 5). The American Psychiatric Association (2013) notes that this symptom exaggeration is purposeful and motivated by external incentives. Examples of external incentives for individuals undergoing forensic evaluations could include a possible reduced prison sentence, acquittal as not guilty by reason of insanity, or being adjudicated incompetent to proceed to trial. The base rate of malingering for defendants undergoing criminal forensic psychological evaluations has been shown to be between 20% and 30% (Frederick, 2000; Mittenberg, Patton, Canyock, & Condit, 2002; Rogers, 2008b), though one study (Ardolf, Denney, & Houston, 2007) reported the base rate to be as high as 54.3% in criminal defendants referred for neuropsychological assessment following a claimed neurocognitive deficit. Accurate assessment and identification of malingering is important due to the cost malingering can have on society and the risk of inappropriate legal outcomes for those who malinger. Specifically, malingering can result in overutilization of valuable mental health services (Kulbarsh, 2009; LoPiccolo, Goodkin, & Baldewicz, 1999; Singh, Avasthi, & Grover, 2007) and inappropriate legal outcomes such as unnecessary hospitalization and/or delay of adjudication. Several feigning strategies may be employed by individuals who engage in malingering (Heinze & Purisch, 2001; Rogers, 2008b; Rogers, Salekin, Sewell, Goldstein, & Leonard, 1998), including exaggeration or fabrication of psychiatric conditions, such as psychosis or trauma responses (Resnick & Knoll, 2008; Resnick, West, & Payne, 2008; Rogers et al., 1998); exaggeration or fabrication of cognitive deficits, such as traumatic brain injuries, memory problems, or intellectual deficits (Bender, 2008; Rogers et al., 1998); exaggeration or fabrication of medical and physical symptoms (Granacher & Berry, 2008; Rogers et al., 1998); and feigned factual knowledge deficits (Colwell, Colwell, Perry, Wasieleski, & Billings, 2008; Otto, Musick, & Sherrod, 2010, 2011; Rubenzer, 2011). Because the definition of malingering requires knowledge of the motivation behind the exaggeration or fabrication of symptoms, the term feigning is used in this article to refer specifically to the behaviors and strategies used by examinees who are attempting to exaggerate or fabricate symptoms without any assumptions about their goals or motivations (see, e.g., Rogers & Bender, 2003).

Two potential means of assessing feigning are through the use of symptom validity tests (SVTs) and performance validity tests (PVTs). As the name suggests, SVTs examine the validity of reported symptoms, whereas PVTs are measures developed to assess for invalid or suboptimal effort performance during cognitive testing (Greiffenstein, Gola, & Baker, 1995; Nelson, Sweet, Berry, Bryant, & Granacher, 2007). Two exploratory factor analyses comparing scores on psychological SVTs and cognitive test performances have revealed that cognitive and psychological response validity tests assess distinct constructs (Greiffenstein et al., 1995; Nelson et al., 2007). Because examinees in forensic contexts may employ more than one feigning strategy when completing psychological assessments, it is important to understand the relationship between common strategies that might affect the reliability and validity of test results. Assessment for different feigning strategies is particularly important, as examinees who feign one type of deficit might not feign other types of deficits that are relevant to the forensic criteria being addressed in a specific evaluation. The majority of previous research to date has focused on individual feigning strategies that an examinee might undertake during psychological assessment and few studies have examined the relationship between different feigning strategies within one sample.

Heinze and Purisch (2001) studied 57 pretrial male defendants who were suspected of feigning and were receiving competency restoration treatment at a large forensic hospital. All participants were administered several tests designed to detect different types of feigning, including the Minnesota Multiphasic Personality Inventory–2 (Butcher, Dahlstrom, Graham, Tellegen, & Keammer, 1989), Structured Interview of Reported Symptoms (SIRS; Rogers, Bagby, & Dickens, 1992a), M Test (Beaber, Marston, Michelli, & Mills, 1985), Structured Inventory of Malingered Symptomology (Smith & Burger, 1997), Rey 15-Item Test (Rey, 1964), Test of Memory Malingering (TOMM; Tombaugh, 1996), and the Atypical Presentation Scale of the Georgia Court Com-petency Test (Gothard, Rogers, & Sewell, 1995). The results of this study indicated that only 12% of the sample scored within the feigning range on all tests, with the exception of the 15-Item Test, leading the authors to conclude that although all of the tests employed in this study were sensitive to multiple types of feigning, the use of only one test in an assessment of feigning might lead to underidentification of the broad spectrum of feigning strategies that might be utilized by examinees.

A large body of research indicates that a common strategy utilized by examinees who are feigning is to put forth invalid or suboptimal effort during assessments of cognitive functioning (Vallabhajosula & van Gorp, 2001; Weinborn, Orr, Woods, Conover, & Feix, 2003). The TOMM (Tombaugh, 1996) is a PVT designed to detect invalid or suboptimal effort during cognitive assessment and is presented to the examinee as a test of memory abilities. Therefore, the TOMM provides useful information to clinicians when assessing for the possibility of feigned memory deficits. The TOMM has been reported to have high specificity across a number of different clinical groups, including samples of psychiatric patients and forensic examinees (e.g., Gierok, Dickson, & Cole, 2005; Vallabhajosula & van Gorp, 2001; Weinborn et al., 2003). For example, Duncan (2005) found that a sample of patients with schizophrenia spectrum disorders obtained average scores above empirically derived cut scores for feigning on the TOMM. Additionally, although patients with documented concentration difficulties scored significantly lower on the TOMM than those without concentration problems, they also obtained mean scores around 45 on both TOMM Trial 2 and the Retention Trial (Duncan, 2005). These findings indicate that the TOMM has low false positive rates among individuals with schizophrenia spectrum disorders. Consequently, low scores on the TOMM among individuals with such disorders raise concern about possible feigning.

Another well-studied feigning strategy during forensic psychological assessments is feigning of psychiatric symptoms such as psychosis. One of the most widely studied SVTs for the detection of psychiatric symptom feigning is the SIRS (Rogers, Bagby, et al., 1992a). In a meta-analysis of 26 studies on the SIRS, Green and Rosenfeld (2011) reported large composite effect sizes (d = 1.53 for the SIRS Total Score and d = 1.53 for the averaged SIRS Primary Scale scores) in the identification of feigned psychiatric symptoms. Additionally, Green and Rosenfeld (2011) found a weighted mean sensitivity of .74 and a weighted mean specificity of .89 among all reviewed studies. The SIRS has also been shown to maintain its discriminability between coached and uncoached samples, between individuals simulating psychiatric symptoms and control groups, and in correctional samples using simulators and controls (Rogers, Gillis, & Bagby, 1990; Rogers, Gillis, Bagby, & Monteiro, 1991; Rogers, Gillis, Dickens, & Bagby, 1991). Additionally, 13 scales within the SIRS have demonstrated the ability to discriminate individuals who were suspected of feigning from nonfeigning psychiatric inpatients (Rogers, Gillis, Dickens, et al., 1991), whereas in a correctional sample comparing simulators with controls, 6 of the 13 scales discriminated the two groups (Rogers et al., 1990). With regard to feigning symptoms of specific disorders, the SIRS has been shown to discriminate a sample of correctional subjects instructed to feign symptoms of schizophrenia, mood, or trauma symptoms from genuine psychiatric patients (Rogers, Kropp, Bagby, & Dickens, 1992b).

Despite the large body of literature examining psychiatric and cognitive feigning instruments individually, limited information exists about the relationship between cognitive and psychiatric feigning strategies within the context of a single assessment in a forensic evaluation setting. Spe-cifically, little is known about the likelihood that forensic samples will engage in these strategies individually or in combination in a single assessment. It is unknown if examinees who feign cognitive symptoms are more likely to feign psychiatric symptoms in comparison with examinees who do not feign cognitive symptoms. Likewise, it is unknown if examinees who feign psychiatric symptoms are more likely to feign cognitive symptoms in comparison with examinees who do not feign psychiatric symptoms.

The current study evaluated the relationship between feigned cognitive symptoms (identified with the TOMM) and feigned psychiatric symptoms (identified with the SIRS) in a pretrial sample of forensic psychiatric patients who were judicially committed as incompetent to stand trial (IST), thereby providing possible incentive to feign one or more types of symptoms. The primary aim of this study was to examine the relationship between these two commonly utilized feigning strategies. Specifically, the current study compared the TOMM performances of examinees who had been identified as feigning on the SIRS with those of examinees who scored in the nonfeigning range on the SIRS. Conversely, we also compared the SIRS performances of examinees who had been identified as feigning on the TOMM with those of examinees who had been identified as giving valid effort on the TOMM. These different analyses allowed us to compare the constructs not only at the scale level but also at the level of the decision rules that are commonly used with both tests when assessing for feigning in forensic psychological evaluations.

Method

Participants

The participants for this study were drawn from an archival database from a larger data collection project studying feigning in a sample of forensic psychiatric hospital inpatients that had been adjudicated as IST. No data from the larger project have been previously published. The hospital is a maximum-security forensic state-funded facility that treats individuals who have been adjudicated IST, Not Guilty by Reason of Insanity, or committed under other state-specific legal statutes. Various treatments are provided to IST patients at the facility including competency restoration groups and psychiatric medication management services. Some patients are court-ordered to take involuntary psychiatric medications under Sell v. United States (2003).

In the jurisdiction of the current study, individuals found IST are initially evaluated in a jail setting by independent court-appointed evaluators. When the courts arrive at a determination of incompetence based on these evaluations, the individuals are transferred to forensic inpatient hospital facilities and efforts are made to restore their competency by means of competency restoration groups and psychiatric medications. The large majority of IST patients in this jurisdiction are not evaluated with SVTs or PVTs in the jail setting prior to commitment to the hospital. Therefore, early in treatment at the hospital, many IST patients are evaluated with SVTs and/or PVTs to assess the validity of their self-reported symptoms. As the current study was specifically focused on examining the relationship between psychiatric and cognitive feigning, only individuals who had completed both the TOMM and the SIRS were included in the final sample.

All assessment protocols completed during a 13-year period were reviewed for the presence of both SIRS and TOMM data during the same assessment. A total of 150 protocols during the relevant time period included both the SIRS and TOMM. All examinees were committed as IST for inpatient competency restoration treatment. Information regarding current criminal charges was coded using the offense categories from the “Criminal Versatility” item of the Psychopathy Checklist–Revised Second edition (Hare, 2003; see Table 1). Because many examinees had multiple charges, all current charges were coded for every examinee. The coding system used to categorize alleged offenses can be found in the manual for the Psychopathy Checklist–Revised Second edition (Hare, 2003). The referral questions were reviewed for every protocol and were coded for the presence/absence of feigning, trial competency, and/or general clinical/diagnostic issues as the focus of the assessment request/referral. It was possible for a single protocol’s referral question to include any combination of these three issues. Diagnoses were obtained from the treatment record for the date of the assessment. Therefore, the diagnoses were not influenced by the results of the SIRS and/or TOMM. Diagnoses were rendered by an interdisciplinary treatment team consisting of a psychologist, psychiatrist, social worker, rehabilitation therapist, and nursing staff. When Diagnostic and Statistical Manual of Mental Disorders–Fourth edition codes for malingering were listed on the diagnosis, they were likely based on a number of factors including inconsistencies in observed versus reported symptoms and inconsistencies between symptoms presented during formal treatment conferences versus symptoms observed on the unit. The demographic characteristics of the sample are provided in Table 1. The first column of Table 1 lists the primary diagnoses and the second column lists all diagnoses regardless of whether they were the primary focus of treatment. The examinees had an average of 1.97 diagnoses (SD = 1.00, range = 0-8; mode = 2.00, median = 2.00) on Axis I and an average of 0.37 diagnoses on Axis II (SD = 0.56, range = 0-3; mode = 0.00, median = 0.00). Nearly 85% of the sample was specifically referred for an assessment of feigning as a major focus of the evaluation request. Therefore, this sample was considered to be at a higher risk for feigning than the typical IST patient in the hospital. Additionally, feigning was likely suspected in the entire sample as the evaluators chose to administer at least two different measures assessing feigning for every protocol included in this study.

Table 1.

Demographic Data.

	N	Range	M (SD)
Age (years)	149	18-65	36.42 (9.91)
Education (years)	100	2-15	9.88 (2.37)
	N	%
Race/ethnicity
African American	61	40.9
Asian	4	2.7
Caucasian	41	27.5
Hispanic/Latino	37	24.8
Other	6	4.0
Gender
Female	25	16.7
Male	125	83.3
	% Yes	% No
Referral question
Feigning	84.8	15.0
Competency	20.0	80.0
General clinical/diagnostic	31.7	68.3
	Primary	All diagnoses (most had >1)
Diagnoses %
Psychotic disorder	78.5	78.1
Mood disorder	8.9	11.6
Substance use disorder	4.8	73.8
Malingering	7.5	13.7
Rule-out malingering	0.07	2.7
Paraphilias	0.07	2.1
PTSD/anxiety	0.07	5.5
Cognitive disorder	—	4.1
BIF/intellectual disability	—	4.8
	%
Charges (most had >1)
Assault/verbal threats	44.4
Theft	33.3
Murder/attempted	18.9
Obstruction of justice	18.9
Robbery	16.7
Drug offenses	11.2
Sexual offenses	5.6
Kidnapping	4.4
Other/miscellaneous	3.3
Criminal negligence	2.2
Arson	1.1

Note. n = 150. BIF = borderline intellectual functioning; PTSD = posttraumatic stress disorder.

Measures

Descriptive statistics for the measures are reported in Table 2.

Table 2.

Sample Means on Relevant Clinical Variables.

	Range	M (SD)
SIRS Primary Scales
Number of Scales in the Definite Range	0-7	1.12 (1.69)
Number of Scales in the Probable Range	0-7	2.18 (1.88)
Rare Symptoms	0-16	5.31 (3.74)
Symptom Combinations	0-20	5.17 (4.42)
Improbable and Absurd Symptoms	0-14	2.92 (3.11)
Blatant Symptoms	0-30	13.33 (8.18)
Subtle Symptoms	0-34	15.44 (9.29)
Selectivity of Symptoms	0-32	16.62 (8.08)
Severity of Symptoms	0-31	11.68 (8.58)
Reported versus Observed Symptoms	0-18	4.28 (3.28)
SIRS Total Score	2-221	100.52 (42.87)
	Range	M (SD)	Score frequency: N (%)
TOMM
Trial 1	5-50	35.23 (11.50)	50-45: 43 (28.7)
			44-39: 26 (17.3)
			38-33: 22 (14.7)
			32-27: 22 (14.7)
			26-20: 22 (14.7)
			19-11: 10 (6.7)
			≤10: 5 (3.3)
Trial 2	5-50	38.69 (12.22)	50-45: 67 (46.5)
			44-39: 16 (11.1)
			38-33: 13 (9.0)
			32-27: 23 (16.0)
			26-20: 12 (8.3)
			19-11: 9 (6.3)
			≤10: 4 (2.8)
Retention Trial	4-50	36.53 (13.13)	50-45: 30 (39.5)
			44-39: 9 (11.8)
			38-33: 6 (7.9)
			32-27: 15 (19.7)
			26-20: 7 (9.2)
			19-11: 6 (7.9)
			≤10: 3 (3.9)

Note. n = 150. SIRS = Structured Interview of Reported Symptoms; TOMM = Test of Memory Malingering.

The SIRS (Rogers et al., 1992a) is a SVT that uses a structured interview for the detection of feigned psychiatric symptoms for use in adult populations. The SIRS contains eight primary scales: Rare Symptoms, Symptom Com-binations, Improbable and Absurd Symptoms, Blatant Symptoms, Subtle Symptoms, Selectivity of Symptoms (SEL), Severity of Symptoms (SEV), and Reported versus Observed Symptoms (RO). For each Primary Scale, four classifications can be made: honest, indeterminate, probable feigning, and definite feigning. A meta-analysis of the SIRS revealed average specificity rates of 73.9% and average sensitivity rates of 89% (Green & Rosenfeld, 2011). Alternatively, a SIRS Total Score can be calculated by summing the Detailed and General Inquires item scores and a score of >76 is suggestive of feigning (Rogers et al., 1992a). The meta-analysis revealed that using the SIRS Total Score produced an average specificity rate of 72.7% and sensitivity rate of 70.3% (Green & Rosenfeld, 2011). The SIRS scales have demonstrated internal consistencies ranging from .66 to .92 (Rogers, Gillis, Dickens, et al., 1991). In the current sample, the Primary Scales of the SIRS demonstrated internal consistencies ranging from α = .68 (RS scale) to α = .80 (RO scale).

The TOMM (Tombaugh, 1996) is a PVT used to detect invalid effort during cognitive assessment. Because it is presented to the examinee as a test of memory, the TOMM provides information regarding the presence or absence of feigned cognitive deficits. The TOMM has two learning trials (Trial 1 and Trial 2) and an optional Retention Trial. Examinees are shown 50 line drawings and then asked to recall each of the previously shown drawings by choosing between two options on each item (the target drawing and a nontarget drawing that had not previously been shown to the examinee). Feedback is provided immediately for each of the 50 items in each trial. Low scores on Trial 2 or the Retention Trial are suggestive of feigning cognitive problems. Individuals who give invalid performance can be identified by using a cut-score of <45 on Trial 2 or the Retention Trial or by identifying those who score statistically below chance (≤19) on any trial. Using the recommended cut-score of <45 on Trial 2, the TOMM has been shown to have high specificity (100%) and sensitivity (80%) in the original validation study with simulated feigners (Tombaugh, 1996). A study with five experiments published after the TOMM was available for purchase reveals similar high rates of specificity ranging from 95% to 100% and sensitivity ranging from 84% to 100% (Rees, Tombaugh, Gansler, & Moczynski, 1998; Tombaugh, 1997).

Procedure

Protocols were classified as “psychiatric feigning” if they obtained three or more SIRS Primary Scales in the Probable range or one or more Primary Scales in the Definite range according to SIRS classification criteria. This cutoff was selected due to the psychometric properties of this decision rule reported in the SIRS manual (Rogers et al., 1992a) and due to its use in other studies (e.g., Sellbom, Toomey, Wygant, Kucharski, & Duncan, 2010). Protocols were classified as “cognitive feigning” if they obtained a score <45 on Trial 2 or the Retention Trial on the TOMM or obtained a score of ≤19 (i.e., statistically below the chance range) on any of the trials (Tombaugh, 1996). Using the SIRS and TOMM decision rules, each protocol could be classified into one of four groups: (a) Both Psychiatric and Cognitive Feigning, (b) Psychiatric Feigning Only, (c) Cognitive Feigning Only, or (d) Neither Psychiatric nor Cognitive Feigning.

Data Analyses

The base rates of both psychiatric and cognitive feigning were calculated using the decision rules described above. Then, three sets of analyses were used to evaluate different levels of data. First, chi-square analyses were performed to determine the association between being classified as psychiatric feigning and cognitive feigning based on these decision rules. Relative risk ratios (RRRs) were also calculated to identify the increased risk associated with being identified as either psychiatric feigning or cognitive feigning when a participant was feigning the other strategy. These analyses treated both feigning strategies as dichotomous constructs. We used RRRs rather than odds ratios because they are generally easier to interpret and because RRRs can provide an estimate of the differential probability of an outcome (e.g., psychiatric feigning) given the presence or absence of a risk factor for that outcome (e.g., cognitive feigning). RRRs can be used when the population at risk (in this case, the population at risk for feigning among examinees who were administered both the SIRS and TOMM) can be estimated. Odds ratios are more difficult to interpret, particularly with high base rate conditions (Sedgwick, 2012), such as those seen in this study. RRRs have been increasingly used in a similar manner in published assessment research (e.g., Gottfried, Bodell, Carbonell, & Joiner, 2014; Tarescavage, Corey, & Ben-Porath, 2015; Tarescavage, Fischler, et al., 2015).

Second, bivariate correlations were performed to examine the associations between SIRS scales and TOMM trial total scores. These analyses treated both feigning strategies as dimensional constructs. Finally, independent samples t tests were performed to compare the TOMM scores of protocols classified as “psychiatric feigning” with those that were classified as not “psychiatric feigning.” Similarly, independent samples t tests were performed to compare the SIRS scale scores of protocols classified as “cognitive feigning” with those classified as not “cognitive feigning.” These analyses treated each dependent variable as dimensional and assumed that the other feigning strategy was either present or absent. Although all three analyses addressed the same basic question (i.e., what is the association between the two constructs?), they addressed the question using different levels of data that might be interpreted in clinical assessment contexts.

Results

Base Rate of Feigning

Based on the cut-scores reported in the professional manuals and used in other empirical studies, 51.3% of the protocols were classified as “cognitive feigning” and 57.3% of the protocols were classified as “psychiatric feigning.” Taken together, 28.0% of the sample answered within the nonfeigning range on both the TOMM and the SIRS, suggesting that 72% of the sample was identified as feigning on at least one of the measures. Specifically, 36.7% obtained scores within the feigning range on both the TOMM and the SIRS, 20.7% elevated the SIRS into the feigning range but responded in the honest range on the TOMM, and 14.7% scored within the feigning range on the TOMM but in the honest range on the SIRS. Results revealed that 71.4% of the individuals who scored within the feigning range on the TOMM also scored within the feigning range on the SIRS and 64.0% of those who scored within the feigning range on the SIRS also scored within the feigning range on the TOMM (χ²[1, N = 150] = 12.85, p < .01, phi-coefficient = 0.29).

RRRs describe the risk or prevalence of a certain event (e.g., psychiatric feigning) happening in one group (e.g., those who feigned cognitive symptoms) compared with the risk of the same event happening in another group (e.g., those who did not feign cognitive symptoms). To assist the reader with interpretation, we provide a description of how RRR was calculated for psychiatric feigning when comparing those who feigned cognitive symptoms with those who did not feign cognitive symptoms. The percentage of examinees in the “cognitive feigning” group who feigned psychiatric impairment was 71.4%. In comparison, 42.5% of the examinees in the “noncognitive feigning” group feigned psychiatric impairment. Dividing the risk if feigning cognitive impairment (71.4%) by the risk if not feigning cognitive impairment (42.5%) yields an RRR of 1.68. This indicates that individuals feigning cognitive impairment were 1.68 times more likely to feign psychiatric impairment than those who did not feign cognitive impairment. If the 95% confidence interval (CI) overlapped with 1.0, the results would indicate an equal risk of psychiatric feigning across the two groups. Because the 95% CI [1.24, 2.28] does not overlap with the value of 1.0, the RRR is statistically significant indicating that the risk of psychiatric feigning is not equal across groups. An RRR was also computed to determine the relative risk for feigning cognitive symptoms for the “psychiatric feigning” group in comparison to the “nonpsychiatric feigning” group. This analysis indicated that examinees shown to be feigning psychiatric symptoms were 1.86 times, CI [1.28, 2.70], more likely to feign cognitive deficits in comparison with those not shown to be feigning psychiatric symptoms. Again, because the 95% CI did not overlap with 1.0, this result was statistically significant.

Associations Between the SIRS Primary Scales and the TOMM Scores

The bivariate correlations between the SIRS Primary Scales and the three trials of the TOMM are presented in Table 3. Bivariate correlations that are .10 are considered small, .30 are medium, and .50 are large (Cohen, 1988). The first two trials of the TOMM were significantly associated with all of the primary scales of the SIRS and the Retention Trial of the TOMM was significantly associated with each primary scale of the SIRS, with the exception of the SIRS RO scale. As expected, the TOMM was negatively correlated with the SIRS scales, as higher scores on the SIRS are more indicative of feigning psychiatric symptoms, whereas lower scores on the TOMM are more indicative of feigning cognitive symptoms.

Table 3.

Bivariate Correlations Between the Primary SIRS Scales and TOMM Trials.

SIRS scores	TOMM Trial 1	TOMM Trial 2	TOMM Retention
Rare Symptoms	−.35	−.31	−.46
Symptom Combinations	−.38	−.34	−.50
Improbable or Absurd Symptoms	−.36	−.29	−.28
Blatant Symptoms	−.40	−.38	−.34
Subtle Symptoms	−.42	−.36	−.36
Selectivity of Symptoms	−.38	−.34	−.29
Severity of Symptoms	−.45	−.40	−.39
Reported versus Observed Symptoms	−.32	−.28	−.14

Note. TOMM Trial 1, n = 150; TOMM Trial 2, n = 144; TOMM Retention Trial, n = 76. SIRS = Structured Interview of Reported Symptoms; TOMM = Test of Memory Malingering.

Comparison of Psychiatric Feigners and Psychiatric Nonfeigners on the TOMM

For these analyses, protocols were first classified as Psychiatric Feigners or Psychiatric Nonfeigners based on the SIRS decision rules. Independent samples t tests were then conducted to determine whether these two groups performed differently on the TOMM trials. Individuals who were feigning psychiatric symptoms scored significantly lower (lower scores are indicative of feigning) on Trial 2 and the Retention Trial of the TOMM than individuals who were not feigning psychiatric symptoms. Table 4 displays these analyses. For the assessment of feigning, Rogers (2008b) has recommended the following standards for interpreting effect sizes (Cohen’s d): moderate ≥ 0.75, large ≥ 1.25, and very large ≥ 1.50.

Table 4.

TOMM Performances Among Psychiatric Feigners and Psychiatric Nonfeigners.

TOMM Trial	Psychiatric feigning		Not psychiatric feigning		t	df	d
TOMM Trial	M	SD (% failed)	M	SD (% failed)	t	df	d
Trial 2	35.01	12.69 (74.1)	43.43	9.81 (34.9)	4.35***	142	0.73
Retention	33.00	14.03 (69.6)	41.93	9.50 (50.0)	3.06**	74	0.72

Note. TOMM Trial 1, n = 150; TOMM Trial 2, n = 144; TOMM Retention Trial, n = 76. TOMM = Test of Memory Malingering; df = degrees of freedom; % failed = percent who scored <45 on TOMM Trial.

p < .05. **p < .01. ***p < .001.

Comparison of Cognitive Feigners and Cognitive Nonfeigners on the SIRS

For these analyses, protocols were first classified as Cognitive Feigners or Cognitive Nonfeigners based on the TOMM decision rules. Independent samples t tests were then conducted to determine whether these two groups performed differently on the SIRS scales. Independent samples t tests indicated that individuals who were feigning cognitive symptoms obtained significantly higher scores on all of the Primary Scales and Total Score of the SIRS than those who were not feigning cognitive symptoms. Table 5 displays these analyses.

Table 5.

SIRS Performances Among Cognitive Feigners and Cognitive Nonfeigners.

SIRS scale	Cognitive feigning		Not cognitive feigning		t	df	d
SIRS scale	M	SD	M	SD	t	df	d
Rare Symptoms	6.05	3.85	4.51	3.47	−2.55**	147	0.42
Symptom Combinations	6.29	4.80	3.97	3.64	−3.30***	147	0.54
Improbable or Absurd Symptoms	3.55	3.33	2.25	2.71	−2.59**	147	0.42
Blatant Symptoms	15.55	8.04	10.90	7.68	−3.57***	145	0.59
Subtle Symptoms	17.92	9.38	12.71	8.43	−3.53***	145	0.58
Selectivity of Symptoms	18.60	7.81	14.44	7.87	−3.21**	145	0.53
Severity of Symptoms	14.04	8.27	9.09	8.20	−3.64***	145	0.60
Reported versus Observed Symptoms	4.90	3.50	3.60	2.90	−2.43*	145	0.40
Total	113.67	44.07	87.36	37.57	−3.64***	126	0.64

Note. n = 150. SIRS = Structured Interview of Reported Symptoms; df = degrees of freedom; Total = SIRS Total Score.

p < .05. **p < .01. ***p < .001.

Discussion

The current study aimed to examine the relationship between psychiatric and cognitive feigning in a pretrial forensic psychiatric sample with significant incentive to feign symptoms. The base rate of feigning in the current sample was higher (51.3% scored in the feigning range on the TOMM and 57.3% scored in the feigning range on the SIRS; 72% were feigning on at least one measure) than what has been cited in the literature (20% to 54.3% across various studies; Ardolf et al., 2007; Frederick, 2000; Mittenberg et al., 2002; Rogers, 2008b). This high base rate is likely attributable to the fact that 84.8% of the protocols in this study noted that the examinee was specifically referred for an assessment of feigning as a major focus of the evaluation request. Although feigning was not specifically listed as the focus of the evaluation request in 15.2% of the sample, it is likely that feigning was suspected in the entire sample because the evaluator chose to administer at least two different measures assessing feigning for every protocol included in this study. Therefore, it would be expected that our sample of forensic inpatients who had been adjudicated IST and administered more than one feigning measure would have a higher base rate of feigning than individuals in other samples. The results also suggested that scoring within the feigning range on one of these measures significantly increased the risk of scoring within the feigning range on the other measure. Therefore, examinees who utilized one feigning strategy were significantly more likely to have utilized the other feigning strategy as well.

Expectedly, an association between the TOMM and the SIRS was observed in the current study. Trials 1 and 2 of the TOMM were significantly correlated with all of the SIRS Primary Scales, with the strongest relationship being observed between the SIRS SEL scale and both Trials of the TOMM. The SEL scale measures symptom endorsement by combining the scores of the Blatant Symptoms and Subtle Symptoms SIRS scales and high scores indicate that the test taker endorsed psychiatric symptoms in an indiscriminate manner (Rogers et al., 1992a). Because individuals who scored high on the SIRS SEL scale are utilizing an indiscriminate manner of feigning, it is not surprising that this score was strongly related to feigning memory impairments on the first two trials of the TOMM. The TOMM Retention Trial was significantly correlated with each SIRS Primary Scale with the exception of the RO scale. The RO scale includes behavioral items in which the evaluator rates whether the examinee reports readily observable symptoms that are not observed by the examiner and/or actually begins displaying symptomatic behavior after being asked about a symptom consistent with that behavior. The TOMM Retention Trial was most strongly correlated with the SIRS Symptom Combinations scale. The SC scale contains item pairs that are uncommonly endorsed by genuine psychiatric patients.

Although a greater number of participants elevated the SIRS than the TOMM, 64% of the participants who were feigning psychiatric symptoms also were shown to produce scores within the feigning range on the TOMM. Participants who were feigning psychiatric symptoms had significantly lower (lower scores suggest feigning) mean scores on the TOMM Trial 2 and the Retention Trial than those who were not feigning psychiatric symptoms. Despite this significant finding, it is notable that participants who were not feigning psychiatric symptoms also obtained mean scores below the recommended cut-score of 45 on the TOMM. Similarly, 71.4% of participants who were feigning cognitive impairments also scored in the feigning range on the SIRS. Those feigning cognitive impairments also had significantly higher (higher scores suggest feigning) mean scores on each of the SIRS scales. The largest effect was demonstrated on the SIRS Total Score followed by the SEV scale. The SEV scale assesses the number of symptoms that are endorsed as being unbearable in severity.

Limitations of the current study include that no feigning measure is perfect, so there were likely some false positive and false negative findings in this sample for both psychiatric feigning and cognitive feigning. Specifically, some examinees identified as feigning on the TOMM or the SIRS may not have been feigning. Conversely, some examinees may have been attempting to feign cognitive or psychiatric symptoms, but were not identified by the SIRS and/or TOMM. Because previous research indicates that both of these instruments generally have acceptable sensitivity and specificity among individuals with psychiatric diagnoses, it is likely that the majority of protocols were classified accurately by the tests. Additionally, as mental illness and feigning are not mutually exclusive, many, if not all, of our participants had a genuine mental illness and this could have affected the results. Future research employing a sample of pretrial defendants who have not had their competency status adjudicated may produce different findings. Another limitation of this study is that the findings might not generalize to a noncriminal forensic setting or assessment context (e.g., disability evaluations). The examinees in our sample were relatively impaired, lower functioning, and were assessed in a pretrial forensic inpatient setting. Higher functioning examinees in other contexts might present with more sophisticated and specific response styles. Finally, this study is limited to assessing the relationship between feigned memory impairment and feigned psychosis. Future research examining the relationship between other types of feigning, such as feigned factual knowledge deficits, would be particularly useful in a sample of individuals undergoing competency assessments.

Although some studies have indicated very low false positive rates on the TOMM among individuals with documented intellectual disabilities (Love, Glassmire, Zanolini, & Wolf, 2014; Simon, 2007), other studies have shown that individuals with intellectual disabilities who are not suspected of feigning frequently score within the feigning range on the SIRS and the TOMM (Graue et al., 2007; Hurley & Deal, 2006). As 4.8% of the sample had borderline intellectual functioning or a mild intellectual disability listed in their diagnosis, this may have created some false positives. It would be beneficial to obtain information regarding intellectual functioning in future similar studies.

Findings of the current study suggest that the strategies of psychiatric feigning and cognitive deficit feigning are related, but can be employed separately between and within forensic examinees. Participants in this study were more likely to feign psychiatric symptoms than cognitive impairments and participants frequently utilized both potential feigning strategies within the same assessment. Additionally, the presence of one feigning strategy was significantly associated with the presence of the other feigning strategy. Therefore, these findings suggest that clinicians should assess for distinct types of feigning in areas relevant to forensic criteria during forensic evaluations.

One major implication of our findings is that although psychiatric feigning and cognitive feigning often co-occur, the relationship between these feigning strategies is not perfect. This point is particularly important, as it is likely that clinicians sometimes erroneously assume that the presence of feigning in one domain indicates that an examinee is feigning in all domains. In a forensic assessment context (e.g., evaluations for trial competency), specific functional deficits that are observed from a clinical standpoint can have strong implications for the assessment of specific forensic criteria. For example, an examinee might be determined to be IST based on delusional beliefs about the legal system that are thought to render the examinee unable to assist counsel in the presentation of a defense. If this same examinee were to fail a cognitive PVT, it would be erroneous to assume that the delusional beliefs affecting trial competency were also feigned based on failure of the PVT. Results from this study indicate that although failure of a cognitive PVT increases the likelihood that the individual might also be feigning psychiatric symptoms, the clinician should not assume that psychiatric symptoms were also feigned without additional assessment data specifically addressing psychiatric feigning. Therefore, it is important to screen for specific types of feigning in domains relevant to forensic criteria in each forensic assessment.

Footnotes

Acknowledgements

The authors wish to thank Anthony Tarescavage for his assistance with an earlier version of this article.

Authors’ Note

The statements and opinions in this article are those of the authors and do not constitute the official views or the official policy of DSH-Patton, the California Department of State Hospitals, or the State of California. Preliminary results of this study were presented at the 2015 American Psychology-Law Society Conference.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

Ardolf

B. R.

Denney

R. L.

Houston

C. M.

(2007). Base rates of negative response bias and malingered neurocognitive dysfunction among criminal defendants referred for neuropsychological evaluation. The Clinical Neuropsychologist, 21, 899-916. doi:10.1080/13825580600966391

Beaber

R. J.

Marston

Michelli

Mills

M. J.

(1985). A brief test for measuring malingering in schizophrenic individuals. American Journal of Psychiatry, 142, 1478-1481. doi:10.1176/ajp.142.12.1478

Bender

S. D.

(2008). Malingered traumatic brain injury. In Rogers

(Ed.), Clinical assessment of malingering and deception (3rd ed., pp. 69-86). New York, NY: Guilford Press.

Butcher

J. N.

Dahlstrom

W. G.

Graham

J. R.

Tellegen

Keammer

(1989). Minnesota Multiphasic Personality Inventory–2: Manual for administration and scoring. Minneapolis: University of Minnesota Press.

Cohen

(1988). Statistical power analyses for the behavioral science (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Colwell

Perry

Wasieleski

Billings

(2008). The test of malingered incompetence (TOMI): A forced-choice instrument for assessing cognitive malingering in competence to stand trial evaluations. American Journal of Forensic Psychology, 26, 17-42.

Duncan

(2005). The impact of cognitive and psychiatric impairment of psychotic disorders on the Test of Memory Malingering (TOMM). Assessment, 12, 123-129. doi:10.1177/1073191105275512

Frederick

(2000). Mixed group validation: A method to address the limitations of criterion group validation in research on malingering detection. Behavioral Sciences & the Law, 18, 693-718. doi:10.1002/bsl.432

10.

Gierok

S. D.

Dickson

A. L.

Cole

J. A.

(2005). Performance of forensic and non-forensic adult psychiatric inpatients on the Test of Memory Malingering. Archives of Clinical Neuropsychology, 20, 755-760. doi:10.1016/j.acn.2005.04.008

11.

Gothard

Rogers

Sewell

(1995). Feigning incompetency to stand trial: An investigation of the Georgia Court Competency Test. Law and Human Behavior, 19, 363-373. doi is:10.1007/bf01499137

12.

Gottfried

Bodell

Carbonell

Joiner

(2014). The clinical utility of the MMPI–2–RF Suicidal/Death Ideation scale. Psychological Assessment, 26, 1205-1211. doi:10.1037/pas0000017

13.

Granacher

R. P.

Berry

D. T. R.

(2008). Feigned medical presentations. In Rogers

(Ed.), Clinical assessment of malingering and deception (pp. 145-156). New York, NY: Guilford Press.

14.

Graue

L. O.

Berry

D. T. R.

Clark

J. A.

Sollman

M. J.

Cardi

Hopkins

Werline

(2007). Identification of feigned mental retardation using the new generation of malingering detection instruments: Preliminary findings. The Clinical Neuropsychologist, 21, 929-942. doi:10.1080/13854040600932137

15.

Green

Rosenfeld

(2011). Evaluating the gold standard: A review and meta-analysis of the structured interview of reported symptoms. Psychological Assessment, 23, 95-107. doi:10.1037/a0021149

16.

Greiffenstein

M. F.

Gola

Baker

W. J.

(1995). MMPI-2 validity scales versus domain specific measures in detection of factitious traumatic brain injury. The Clinical Neuropsychologist, 9, 230-240. doi:10.1080/13854049508400485

17.

Hare

R. D.

(2003). Hare Psychopathy Checklist–Revised (PCL-R): 2nd Edition technical manual. North Tonawanda, NY: Multi-Health Systems.

18.

Heinze

M. C.

Purisch

A. D.

(2001). Beneath the mask. Journal of Forensic Psychology Practice, 1(4), 23-52. doi:10.1300/J158v01n04_02

19.

Hurley

Deal

(2006). Assessment instruments measuring malingering used with individuals who have mental retardation: Potential problems and issues. Mental Retardation, 44, 112-119. doi: 10.1352/0047-6765(2006)44[112:aimmuw]2.0.co;2

20.

Kulbarsh

(2009). Why would anyone want to fake mental illness? Law Officer. Retrieved from http://www.lawofficer.com/article/needs-tags-columns/why-would-anyone-want-fake-men

21.

LoPiccolo

Goodkin

Baldewicz

(1999). Current issues in the diagnosis and management of malingering. Annuals of Medicine, 31, 166-174. doi:10.3109/07853899909115975

22.

Love

C. M.

Glassmire

D. M.

Zanolini

S. J.

Wolf

(2014). Specificity and false positive rates of the Test of Memory Malingering, Rey 15-Item Test, and Rey Word Recognition Test among forensic inpatients with intellectual disabilities. Assessment, 21, 618-627. doi:10.1177/1073191114528028

23.

Mittenberg

Patton

Canyock

Condit

(2002). Base rates of malingering and symptom exaggeration. Journal of Clinical and Experimental Neuropsychology, 24, 1094-1102. doi:10.1076/jcen.24.8.1094.8379

24.

Nelson

N. W.

Sweet

J. J.

Berry

D. T. R.

Bryant

F. B.

Granacher

R. P.

(2007). Response validity in forensic neuropsychology: Exploratory factor analytic evidence of distinct cognitive and psychological constructs. Journal of the International Neuropsychological Society, 13, 440-449. doi:10.1017/S1355617707070373

25.

Otto

Musick

Sherrod

(2010). Inventory of Legal Knowledge™ (Professional manual). Lutz, FL: PAR.

26.

Otto

Musick

Sherrod

(2011). Convergent validity of a screening measure designed to identify defendants feigning knowledge deficits related to competence to stand trial. Assessment, 18, 60-62. doi:10.1177/1073191110377162

27.

Rees

L. M.

Tombaugh

Gansler

D. A.

Moczynski

N. P.

(1998). Five validation experiments of the Test of Memory Malingering (TOMM). Psychological Assessment, 10, 10-20. doi:10.1037/1040-3590.10.1.10

28.

Resnick

P. J.

Knoll

J. L.

(2008). Malingered psychosis. In Rogers

(Ed.), Clinical assessment of malingering and deception (3rd ed., pp. 51-68). New York, NY: Guilford Press.

29.

Resnick

P. J.

West

Payne

J. W.

(2008). Malingering of posttraumatic disorders. In Rogers

(Ed.), Clinical assessment of malingering and deception (3rd ed., pp. 109-127). New York, NY: Guilford Press.

30.

Rey

(1964). L’examen clinicque en psychologie [The clinical examination in psychology]. Paris, France: Presses Universitaires de France.

31.

Rogers

(2008a). An introduction to response styles. In Rogers

(Ed.), Clinical assessment of malingering and deception (3rd ed., pp. 3-13). New York, NY: Guilford Press.

32.

Rogers

(Ed.). (2008b). Clinical assessment of malingering and deception (3rd ed.). New York, NY: Guilford Press.

33.

Rogers

Bagby

Dickens

(1992a). Structured interview of reported symptoms: Professional manual. Odessa, FL: Psychological Assessment Resources.

34.

Rogers

Bender

S. D.

(2003). Evaluation of malingering and deception. In Goldstein

A. M.

Weiner

I. B.

(Eds.), Handbook of psychology: Forensic psychology (Vol. 11, pp. 109-129). New York, NY: Wiley.

35.

Rogers

Gillis

J. R.

Bagby

R. M.

(1990). The SIRS as a measure of malingering: A validation study with a correctional sample. Behavioral Sciences & the Law, 8, 85-92. doi:10.1002/bsl.2370080110

36.

Rogers

Gillis

J. R.

Bagby

R. M.

Monteiro

(1991). Detection of malingering on the Structured Interview of Reported Symptoms (SIRS): A study of coached and uncoached simulators. Psychological Assessment, 3, 673-677. doi:10.1037/1040-3590.3.4.673

37.

Rogers

Gillis

J. R.

Dickens

S. E.

Bagby

R. M.

(1991). Standardized assessment of malingering: Validation of the Structured Interview of Reported Symptoms. Journal of Consulting and Clinical Psychology, 3, 89-96. doi:10.1037/1040-3590.3.1.89

38.

Rogers

Kropp

P. R.

Bagby

R. M.

Dickens

(1992b). Faking specific disorders: A study of the Structured Interview of Reported Symptoms (SIRS). Journal of Clinical Psychology, 48, 643-648. doi: 10.1002/1097-4679(199209)48:5<643::aid-jclp2270480511>3.0.co;2-2

39.

Rogers

Salekin

R. T.

Sewell

K. W.

Goldstein

Leonard

(1998). A comparison of forensic and nonforensic malingerers: A prototypical analysis of explanatory models. Law and Human Behavior, 22, 353-367. doi:10.1023/A:1025714808591

40.

Rubenzer

(2011). Review of the Inventory of Legal Knowledge. Open Access Journal of Forensic Psychology, 3, 70-81.

41.

Sedgwick

(2012). Estimating the population at risk. British Medical Journal, 345, e6818. doi:10.1136/bmj.e6859

42.

Sell v. United States, 539 U.S. 166 (2003).

43.

Sellbom

Toomey

Wygant

Kucharski

Duncan

(2010). Utility of the MMPI–2–RF (Restructured Form) validity scales in detecting malingering in a criminal forensic setting: A known-groups design. Psychological Assessment, 22, 22-31. doi:10.1037/a0018222

44.

Simon

M. J.

(2007). Performance of mentally retarded forensic patients on the Test of Memory Malingering. Journal of Clinical Psychology, 63, 339-344. doi:10.1002/jclp.20351

45.

Singh

Avasthi

Grover

(2007). Malingering of psychiatric disorders: A review. German Journal of Psychiatry, 10, 126-132.

46.

Smith

G. P.

Burger

G. K.

(1997). Detection of malingering: Validation of the Structured Inventory of Malingered Symptomatology (SIMS). Journal of the American Academy of Psychiatry and the Law, 25, 183-189.

47.

Tarescavage

A. M.

Corey

D. M.

Ben-Porath

Y. S.

(2015). Minnesota Multiphasic Personality Inventory–2-Restructured Form (MMPI-2-RF) predictors of police officer problem behavior. Assessment, 22, 116-132. doi:10.1177/1073191114534885

48.

Tarescavage

A. M.

Fischler

G. L.

Cappo

B. M.

Hill

D. O.

Corey

D. M.

Ben-Porath

Y. S.

(2015). Minnesota Multiphasic Personality Inventory–2-Restructured Form (MMPI-2-RF) predictors of police officer problem behavior and collateral self-report test scores. Psychological Assessment, 27, 125-137. doi:10.1037/pas0000041

49.

Tombaugh

(1996). Test of Memory Malingering (TOMM). North Tonawanda, NY: Multi-Health Systems.

50.

Tombaugh

(1997). The Test of Memory Malingering (TOMM): Normative data from cognitively intact and cognitively impaired individuals. Psychological Assessment, 9, 260-268. doi:10.1037/1040-3590.9.3.260

51.

Vallabhajosula

van Gorp

W. G.

(2001). Post-Daubert admissibility of scientific evidence on malingering of cognitive deficits. Journal of the American Academy of Psychiatry and the Law, 29, 207-215. doi: 10.1093/arclin/15.8.847

52.

Weinborn

Orr

Woods

S. P.

Conover

Feix

(2003). A validation of the Test of Memory Malingering in a forensic psychiatric setting. Journal of Clinical and Experimental Neuropsychology, 25, 979-990. doi:10.1076/jcen.25.7.979.16481