Which of the following is an advantage of the personality assessment Inventory PAI

Assessing Behavior and Personality Disturbances in Patients With Cerebral Dysfunction

There is evidence that appropriate treatment of behavior and personality disturbances in patients with acquired brain dysfunction can reduce required care levels and prevent hospitalizations (Chang and Troyer, 2011; Davydow et al., 2013, 2014). Clinical assessment and research of behavior and personality change in individuals with neurological disease and injury are laden with challenges and complexity. Some limitations of the available research are as follows:

Treatment of other symptoms (such as a movement disorder) may mask psychiatric and behavioral symptoms.

Most available neuropsychiatric assessment tools use conventional psychiatric terminology based on idiopathic psychiatric illness, which sometimes fails to distinctly reflect the symptoms associated with acquired disease and/or trauma.

There is overlap between symptoms of cerebral dysfunction and symptoms of behavior and personality disturbances; for example, psychomotor retardation or reduced energy, libido, or appetite might reflect an underlying syndrome (Parkinson disease [PD]), an acquired injury (e.g., TBI), or a major depressive episode.

Cognitive impairments may confound the detection of behavioral changes. For example, language and memory deficits occurring in individuals with cerebral dysfunction can limit self-reports and can restrict the ability to assess changes in mood or insight.

The validity of the behavioral dysfunction assessed can vary depending upon the source. Ample research shows that clinical ratings acquired from the patient, a collateral or spouse, and a healthcare worker can vary widely (see, e.g.,Hoth et al., 2007). Patients with cerebral dysfunction may have impaired insight; thus they may underreport behavioral difficulties. Similarly, caregivers may also provide biased information because their current mood or degree of caregiver burden may influence their reporting of behavioral symptoms.

Nevertheless, clinically meaningful and objective measures of behavioral symptoms are very important. In the clinic, an unstructured but targeted interview with the patient and the caregivers separately can be useful. Inventories and scales based on semistructured interviews give valuable insight when used with appropriate training.

Personality Assessment Inventory (Morey, 1991)

The Personality Assessment Inventory (PAI; Morey, 1991) is a self-administered, multiscale personality inventory providing psychological assessment information about psychopathology, personality, and psychosocial environment, measuring constructs relevant to clinical diagnosis and decision-making, suitable for use with individuals aged 18  +. It applies to diverse types of settings, with surveys of practicing clinicians indicating the PAI as one of the most popular, widely used, clinical measures in mental health, forensic/correctional, screening, and training contexts (e.g., Archer, Buffington-Vollum, Stredny, & Handel, 2006).

The PAI is organized into four sets of scales—four validity scales, 11 clinical scales, five treatment scales, and two interpersonal scales, with many of the full scales also including subscales. The questionnaire includes 344 items, and completion typically takes 45–60 min using either paper and pencil or computer; it requires a fourth-grade reading level to complete. Response options are on a four-point scale, ranging from “totally false, not at all true” to “very true.” Items are summed into raw scores which are then transformed to T-scores (mean of 50, standard deviation of 10), and interpretation is provided relative to a standardization sample of 1000 community-dwelling adults from the United States, selected to match US census characteristics based on gender, race, and age. Internal consistency alphas for the full scales are generally found to be in the .80 in a variety of different settings, and test-retest reliability in nonclinical samples yields similar estimates. Such reliability values provide an estimate of the standard error of measurement of roughly three to four T-score points, with 95% confidence intervals of ± six to eight T-score points. For a comprehensive presentation of available validity evidence, see the PAI manual (Morey, 2007).

The four validity scales of the PAI—Negative Impression Management (NIM), Positive Impression Management (PIM), Inconsistency (ICN), and Infrequency (INF)—were developed to assess profile distortion, which can result from either random (assessed with ICN and INF) or systematic (assessed with NIM and PIM) sources of response distortion. Systematic profile distortion can be further broken down into effortful (i.e., intentional) or noneffortful forms. Effortful distortion occurs when respondents intentionally present themselves in a manner that is at odds with their experience or historical fact (i.e., they might malinger, or fake good). In noneffortful distortion, respondents may present themselves in a manner which is consistent with their subjective experience, but which an experienced clinician might see as lacking insight (most commonly, when overly positive), or as an exaggeration (if overly negative) of symptoms. The PAI provides supplemental indices that assist in distinguishing between these different forms of systematic distortion. Because of the potential reporting biases that might be present in the assessment of psychopathy, the availability of validated response validity scales on the PAI represents an important advantage for test interpretation.

While several of the substantive scales from the PAI are relevant to the assessment of psychopathy, the most commonly researched scale for this application is the Antisocial Features (ANT) scale. This scale includes three conceptually distinct subscales intended to measure facets of antisociality: ANT-A (Antisocial Behaviors), which assesses a history of conduct problems and criminality; ANT-E (Egocentricity), which measures a self-centered, callous, and remorseless interpersonal style; and ANT-S (Stimulus Seeking), which reflects a tendency to take risks and a strong desire for novelty. The ANT scale has shown to be a robust predictor of theoretically relevant forms of social deviance, such as institutional misconduct and criminal recidivism (e.g., Reidy, Sorensen, & Davidson, 2016). A recent meta-analytic review (Gardner, Boccaccini, Bitting, & Edens, 2015) indicated that ANT scores consistently emerged as a small to moderate predictor of institutional misconduct in both correctional and treatment settings.

Research has generally shown that ANT scores are moderate to strongly associated with both self-report and interview-based assessment of both an antisocial personality disorder and psychopathy (Douglas, Guy, Edens, Boer, & Hamilton, 2007; Edens, Poythress, & Watkins, 2001; Guy, Poythress, Douglas, Skeem, & Edens, 2008). However, there is some evidence to suggest that the ANT scale may be more strongly related to behavioral deviance than to interpersonal or affective features of psychopathy (Douglas et al., 2007; Lilienfeld, Fowler, & Patrick, 2006). This is supported by findings of higher associations with Factor 2 (the social deviance factor) of the Psychopathy Checklist-Revised (PCL-R; Hare, 2003), with smaller associations between the ANT Scale and affective and interpersonal deficits as assessed by the PCL-R (Douglas et al., 2007; Edens, Hart, Johnson, Johnson, & Olver, 2000; Kucharski, Petitt, Toomey, & Duncan, 2008; Walters, Duncan, & Geyer, 2003). While research suggests the pragmatic utility of the PAI in criminal justice settings, with predictive validity, estimates often comparable to those provided by the PCL-R (Walters & Duncan, 2005), the two instruments should be viewed as complementary rather than as a proxy for one another, as both seem to contribute somewhat independently to prediction of important outcomes (e.g., Buffington-Vollum, Edens, Johnson, & Johnson, 2002).

Another PAI scale that is theoretically relevant to the assessment of psychopathy is Aggression (AGG). The AGG scale assesses anger, hostility, and aggression and is comprised of three subscales: Aggressive Attitude (AGG-A), Verbal Aggression (AGG-V), and Physical Aggression (AGG-P). Similar to the ANT scale, the AGG scale has proven to be a robust predictor of institutional misconduct in both correctional and treatment settings, as well as of postrelease recidivism (Gardner et al., 2015; Reidy et al., 2016). Additionally, the AGG scale has demonstrated incremental utility above PCL-R Factor 2 in the prediction of institutional misconduct (Walters et al., 2003). Given that psychopathic individuals tend to be described as prone to violence and aggression, research has reported the AGG scale to demonstrate moderate correlations with self-report and interview-based indices of psychopathy (Edens, Poythress, & Watkins, 2001; Walters, Duncan, & Geyer, 2003). For example, Edens, Poythress, and Watkins (2001) reported AGG, in addition to the ANT scale, as being a significant behavioral correlate of the Psychopathic Personality Inventory (PPI; Lilienfeld & Andrews, 1996).

In addition to the basic scales of the PAI, supplemental indicators for the PAI have been developed that may provide useful information regarding the assessment of psychopathy and violence risk. The Violence Potential Index (VPI; Morey, 1996) incorporates 20 different features of the PAI profile that serve as putative risk factors for violent behavior. Such features include anger, hostile control in relationships, sensation seeking, impulsivity, agitation, antisocial behavior, grandiosity, and alcohol and drug abuse, among others. Research has shown moderate-sized effects for VPI in predicting both violent and nonviolent recidivism (Boccaccini, Murrie, Hawes, Simpler, & Johnson, 2010; Reidy et al., 2016). The VPI has been shown to correlate with a self-report measure of psychopathic features such as Hare’s 1985 self-report measure of psychopathy (SRP-R) and a diagnostic interview for ASPD (Edens et al., 2000; Morey, 1996).

Aside from the scales most commonly used to assess psychopathy with the PAI (i.e., ANT and AGG), there are a variety of other PAI scales that are theoretically related to the construct. Relevant PAI scales such as Alcohol Problems (ALC) and Drug Problems (DRG), Borderline Features (BOR), Mania (MAN), Paranoia (PAR), Stress (STR), and Treatment Rejection (RXR) have been investigated as correlates of PCL-R defined psychopathy (e.g., Douglas et al., 2007) and of the Psychopathic Personality Inventory (e.g., Patrick, Edens, Poythress, Lilienfeld, & Benning, 2006). Moreover, as interpersonal style theoretically overlaps with core personality traits of psychopathy, several researchers have examined the relationship between the interpersonal scales of the PAI (e.g., Dominance and Warmth) and psychopathic personality. The Dominance (DOM) scale assesses the extent to which one is controlling and independent in personal relationships, while the Warmth (WRM) scale assesses the extent to which a person is interested in supportive and empathic personal relationships. Previous research has shown particular support for the use of the DOM scale as a relevant indicator of interpersonal features of psychopathy (Douglas et al., 2007; Edens, 2009; Patrick et al., 2006). For example, in a sample of 281 Canadian federal offenders, Douglas et al. (2007) found that only the DOM scale predicted Factor 1 (the interpersonal and affective deficit factor) of the PCL-R.

In tandem with the use of specific scales of the PAI to assess psychopathy, recent research has investigated the ability of higher-order factors of the PAI to capture the construct. Such higher-order factors include the broad psychopathological domains of internalizing (INT) and externalizing (EXT) features. In a large correctional sample, Blonigen et al. (2010) explored convergent and discriminant relations between factors of psychopathy measures and latent criteria of INT and EXT derived from the PAI (Blonigen et al., 2010). Using a multimethod approach of both self-report and interview/file review, results indicated scores reflective of affective and interpersonal traits of psychopathy were negatively associated with INT, whereas scores reflective of social deviance exhibited positive associations with both INT and EXT of the PAI.

Personality assessment

Do not rely too heavily on a patient's self-assessment of personality, although it is important to hear how he views himself. Depressed patients may paint a picture of themselves as incompetent, blameworthy and lacking in confidence, which may be far from their premorbid personality. To avoid such misrepresentations, try to rely on objective information in the history, and where possible talk to an informant who knows the patient well and what he is usually like. The important areas to cover in assessing personality are shown inBox 8.11. In this area of assessment a collateral history is mandatory.

Summary

1.

The PAI is not a substitute for the PCL-R. Like other self-report measures, it should never be used for clinically diagnosing psychopathy, or forming “psychopathic groups” in research studies.

The PAI can be reliably used with forensic populations including female offenders to measure antisocial attitudes, potential violence, paranoia, borderline features, and substance abuse. It cannot be used to categorize someone as psychopathic.

The PCL-R total score/Factor 2 and PAI scales of ANT, DRG, and BOR have been correlated, suggesting that the PAI may be capturing the behavioral component of psychopathy and ASPD.

The PAI correlates mostly with PCL-R Factor 2 and not Factor 1. Like the Rorschach, it adds to understanding individual differences among those scored with the PCL-R.

Elevated scores on the ANT, AGG, DRG, ALC, PAR, and DOM scales with forensic samples have been found (Edens et al., 2002; Morey & Quigley, 2002; Reidy et al., 2016).

We have found that on the PAI, psychopathic women display higher levels of borderline features, paranoia, mania, aggression, antisocial behaviors, dominance, and violence potential than non-psychopathic women (Smith et al., 2020b). We have also found similar levels of paranoia, borderline features, aggression, antisocial behaviors, dominance, and violence potential in a sample of female sex offenders (Smith, Gacono, Kivisto et al., 2019).

The Personality Assessment Inventory

The Personality Assessment Inventory (PAI) is one of the newest objective psychological tests available.17 The PAI was developed using a construct validation framework with equal emphasis placed on theory-guided item selection and the empirical function of the scales. The PAI employs 344 items and a four-point response format (False, Slightly True, Mainly True, and Very True), which generates 22 scales with non-overlapping items. The PAI has some psychometric advantages over other self-report instruments. Unlike the MMPI-2 and MCMI, one scale elevation on the PAI will not result in a second, by proxy, scale elevation simply because those scales share items. This characteristic allows for more direct interpretation of each scale. The 22 scales of the PAI consist of four validity scales (Inconsistency [INC], Infrequency [INF], Negative Impression Management [NIM], Positive Impression Management [PIM]); 11 clinical scales (Somatic Complaints [SOM], Anxiety [ANX], Anxiety-Related Disorders [ARD], Depression [DEP], Mania [MAN], Paranoia [PAR], Schizophrenia [SCZ], Borderline [BOR], Antisocial [ANT], Alcohol [ALC], Drug [DRG]); five treatment scales (Aggression [AGG], Suicide [SUI], Stress [STR], Non-support [NON], Treatment Rejection [RXR]); and two interpersonal scales (Dominance [DOM], Warmth [WRM]). The PAI possesses outstanding psychometric features and is an excellent test for broadly assessing multiple domains of relevant psychological function.18–20 The validity and clinical utility of the PAI is also well established.21,22

As was discussed with both the MMPI-2 and the MCMI, the PAI also has validity scales that were created to help detect deviant response styles. The first validity scale, INC, is a collection of item pairs that are expected to be answered in the same direction (similar to the VRIN from the MMPI-2). Typically, elevations on this scale are an indication of confusion, reading problems, or even cognitive impairment.23 INF consists of items that are expected to be answered in a certain direction. Half of these items should be true (e.g., “Most people prefer to be happy”) and half should be false (e.g., “I really enjoy paying taxes”). Elevation of the INF suggests that the respondent was overinterpreting items, was careless, or has reading difficulties. If either INC or INF is above its respective cutoff, it invalidates the test results and further interpretation of the clinical scales is not recommended. The NIM scale is designed to identify respondents who are attempting to present themselves in an overly negative light. Similar to the F scale of the MMPI-2, NIM items are rarely endorsed by most people (e.g., “I have not had a single day of happiness”). Another use for this scale, however, is to identify patients who may want their treaters to know how much psychological distress they are in (e.g., a cry for help). It is not uncommon for NIM to be elevated in clinical samples. When evaluating psychiatric inpatients, NIM scores can reach well into the 80s, which is 3 SDs above community norms. The PIM attempts to identify respondents who are trying to present themselves in an overly positive light. Research has shown that the PIM scale is highly sensitive to efforts to present oneself in an overly positive manner (“I never feel bad”); even modest elevations (T-scores ≥ 57) can raise questions about profile accuracy.

Abstract

The development of personality assessment emerged in the 19th and early 20th centuries. Self-report inventories evolved from behavioral observations and informant rating scales. Various scale construction methods were used to develop self-report inventories to measure normal personality, specific clinical problems like anxiety or depression, or a broader range of mental health problems. This chapter covers four multidimensional measures of psychopathology: the Minnesota Multiphasic Personality Inventory (MMPI) and its revision, the MMPI-2; the Personality Assessment Inventory (PAI); the Millon Clinical Multiaxial Inventories (MCMI); and the recent MMPI-2-Restructured Form (MMPI-2-RF). The MMPI-2, given its wider use, is highlighted. A PsycINFO search from 2003 to 2016 reveals that the MMPI-2 remains the most widely researched inventory (1646 publications). The PAI is the second most researched inventory (880 publications), followed by the MCMI (411 publications), and the MMPI-2-RF (322 publications). A trend of declining research on the MMPI-2 is considered in light of its publisher’s current marketing, research, and development practices.

Extracting situational “items” and systematizing measurement: Challenges

Today, smartphone and emerging sensor technologies afford innovative measurement of personality (e.g., Harari et al., 2016, 2019; Harari, Gosling, Wang, & Campbell, 2015). These methods, however, are constrained by ethical and privacy issues (e.g., recording conversations) as well as consistency issues (e.g., noisy sensor data) across participants. Nonetheless, there are strong alternatives to systematically assess within-person measurement in response to each situation, such as measuring the time it takes for individuals to choose what to do (given known multiple options associated with varying goals). For example, we may measure how long it takes for one to initiate an interaction with another or respond to another’s communication.

Conclusion

The goal of this chapter has been to discuss personality assessment in daily life as a complement to traditional assessment methods in the field of personality development. Assessing personality under real-life and real-time conditions would provide a better understanding about the ways in which personality processes are assembled and unfold over time. The use of ambulatory assessment to capture personality change processes in real-life contexts would offer interesting novel assessment perspectives for the field of personality development. Emerging developments in sensor-enabled mobile technologies to assess daily contexts and individual experiences, perceptions, and behaviors using auditory, visual, olfactory, and smartphone sensing data, will create new opportunities for researchers to study personality development and dynamics in daily life.

Example 3

Sinclair et al. (2013) recently developed a new index for the Personality Assessment Inventory (PAI) as a means of better predicting risk factors for increasing level of care (ILOC). The PAI (Morey, 1991) is a broadband measure of psychological functioning and interpersonal style that was developed using the construct validation approach. It consists of 344 items that fall under 4 validity scales, 11 clinical scales, and 2 interpersonal scales. Sinclair et al. (2015) sought to extend the study of Sinclair et al. (2013) by examining the validity of the level of care index (LOCI) in two independent psychiatric samples. In Study I, differences in LOCI scores were compared across levels of care in a mixed sample of psychiatric inpatients and outpatients to establish construct validity and were also evaluated in terms of their associations with known risk factors for increased level of care (e.g., suicide risk). Likewise, the incremental validity of the LOCI was also evaluated to see whether the index had predictive value above and beyond other PAI indexes that could also provide information about level of care. In Study II the construct validity of the LOCI was further evaluated in a separate inpatient sample by examining associations with several criteria variables, such as previous admissions to the hospital for the prior 6 months and whether someone’s current admission was related to suicide risk.

Results revealed that, in addition to differentiating inpatients from outpatients, the LOCI was also found to be meaningfully associated with a number of risk factors for increased level of care, such as suicide risks and self-harming behaviors, with effect sizes in the moderate to high range. Furthermore, results generally supported the incremental validity of the LOCI, which would further indicate that it contains unique indicators of the potential need for increased level of care that extend beyond suicide and violence risk.

Incorporating personality in clinical professional practice

Finally, De Fruyt and Van Leeuwen (2014) recently argued to better integrate personality assessment in clinical professional practice. A description of a patient’s standing on the traits of a general personality descriptive model such as the FFM is not a standard part of clinical assessment. This is rather remarkable given the complex interplay between personality traits and psychopathology and the importance of personality to understand an individual’s daily functioning. Moreover, personality self- and observer descriptions (partner or parent) can be easily obtained and efficiently scored and normed electronically, without taking psychologist’s time. The present review has tried to propose an integrative framework, relying on trait-activation theory, with a focus on explaining the personality–psychopathology relationship. It was further highlighted where the position of psychological assessment and decision making is in this process, and how elements of this framework may form target areas of different types and schools of psychological interventions. We hope that this chapter may contribute to achieve a better integration of personality assessment in the diagnostic process.