Assessment

By WHITBOURNE, S.K., Halgin, R.P

Edited by Paul Ducham

CHARACTERISTICS OF PSYCHOLOGICAL ASSESSMENTS

A psychological assessment is a procedure in which a clinician provides a formal evaluation of an individual’s cognitive, personality, and psychosocial functioning. In Ben’s case, a comprehensive assessment proved valuable in helping to understand the nature of his symptoms and potential directions for treatment.
    Clinicians conduct assessments under a variety of conditions. In many cases, clinicians use the assessment process to provide a diagnosis, or at least a tentative diagnosis, of an individual’s psychological disorder. However, clinicians may also use assessments for a variety of other reasons. For example, in forensic assessments, clinicians determine whether a suspect meets the criteria of being competent to stand trial. Clinicians might also provide information that employers can use to evaluate an individual’s appropriateness for a particular job. Assessments are also useful when clinicians consult about an individual’s level of functioning in a specific area. An older woman experiencing memory problems may seek neuropsychological assessment to determine whether she has a cognitive impairment that will require further intervention.
    For Ben, the assessment process is critical to understanding the nature of his current symptoms. The clinician must evaluate the potential roles of both brain injury and what may have been the appearance of symptoms unrelated to the train incident. His immediate treatment plan and his long-term psychological development will depend on the outcome of the evaluation. Dr. Tobin provided an initial evaluation, and as a result of this assessment, she decided to refer Ben to a neuropsychologist.
          To be useful, clinicians must hold assessments to standards that ensure that they provide the most reproducible and accurate results. The reliability of a test indicates the consistency of the scores it produces. In other words, it should produce the same results regardless of when it is given, and the individual should answer test items in similar fashion. The test’s validity reflects the extent to which a test measures what it is designed to measure. An intelligence test should measure intelligence, not personality. Before using a given test, clinicians should be aware of its reliability and validity, information that is readily available in the published literature about the instrument.
      The profession strives to design tests so that the results they produce don’t vary from clinician to clinician. The criterion of standardization clearly specifies a test’s instructions for administration and scoring. Each individual receiving the test should have the same amount of time, and each person scoring the test should do so in the same manner according to the same, predefined criteria. Furthermore, a given score on the test that one person obtains should have a clear meaning. Ideally, the test’s designers have a substantial enough database against which to compare each testtaker’s scores.

        In addition to determining a test’s reliability and validity, it is important to take into account its applicability to test-takers from a diversity of backgrounds. Increasingly, test publishers are designing their measures for usage with a variety of individuals in terms of ability level, first language, cultural background, and age. For example, clinicians may need to adapt assessment instruments for use with older adults, who may require larger print, slower timing, or special writing instruments for use with those who have arthritis (Edelstein, Martin, & McKee, 2000). Even so, clinicians need to ensure that they are using the most appropriate instrument for a given client. When interpreting test results, clinicians need to ensure that they don’t fall into the trap of the so-called “Barnum Effect.” Named after legendary circus owner P. T. Barnum, this is the tendency for clinicians unintentionally to make generic and vague statements about their clients that do not specifically characterize the client. Here’s an example of a Barnum Effect statement: “Julia is often shy around other people, but at times she can be very outgoing. When presented with a challenge, she can often perform very well, but she occasionally becomes nervous and intimidated.” These two sentences could apply to Julia, but they could also apply to most other people. Therefore, they don’t say anything special about Julia. Furthermore, most people would find it difficult to disagree with this feedback. You are most likely to encounter the Barnum Effect in situations such as reading your horoscope or a fortune cookie, which are written so generally that they could apply to anyone. These are relatively harmless situations, unless you decide to invest a great deal of money on the basis of such an unreliable prognosticator. In a clinical situation, the problem is that such statements are not particularly insightful or revealing and do not help inform the assessment process.
        Clinicians should keep up with the literature to ensure that they are using the best assessment methods possible. Evidence-based assessment includes (1) relying on research findings and scientifically viable theories; (2) using psychometrically strong measures; and (3) empirically evaluating the assessment process (Hunsley & Mash, 2007). By following these guidelines, clinicians ensure that they will evaluate their clients using the most current and appropriate materials available. For example, a seasoned clinician may have a preference for using the assessment methods she learned about in graduate school, but she should be constantly alert for newer procedures that rely on newer technology or research. According to criterion (3), she should also develop evaluation methods to assess whether her assessments are providing useful information about her clients.
     For example, consider the case of an assessment suggesting that a client is experiencing significant depressive symptoms even though she seeks help for what she describes as attacks of anxiety. Following the criteria for evidence-based assessment, the clinician would determine whether the tool or tools she used to make the diagnosis provided an accurate characterization of the woman’s symptoms as they evolved over the course of treatment. Similarly, in Ben’s case, the clinician must validate the findings from neuropsychological assessment carefully by obtaining multiple measures to assess possible brain injury.

CLINICAL INTERVIEW

Clinicians typically begin their assessment with the clinical interview, a series of questions that they administer in face-to-face interaction with the client. The answers the client gives to these questions provide important background information on clients, allow them to describe their symptoms, and enable clinicians to make observations of their clients that can guide decisions about the next steps, which may include further testing.
    The least formal version of the clinical interview is the unstructured interview, which consists of a series of open-ended questions regarding the client’s symptoms, health status, family background, life history, and reasons for seeking help. In addition to noting the answers to these questions, the clinician also observes the client’s body language. By noting the client’s nonverbal cues, the clinician can gain an understanding of whether the client is experiencing, for example, anxiety, attentional difficulties, unwillingness to cooperate, or unusual concern about testing. The clinician may also use cues from the client’s appearance that give further indication of the client’s symptoms, emotional state, or interpersonal difficulties. The typical clinical interview covers the areas outlined in Table 3.2. The clinician can vary the order of questions and the exact wording he or she used to obtain this information.
            In Ben’s case, the clinical interview provided Dr. Tobin with key information about his history including not only his educational and vocational background, but also his relationship history. She determined that, prior to the incident, he enjoyed engaging with others, so his current isolation is a change from his previous pattern of social functioning.
       Dr. Washington obtained more detailed information from his clinical interview with Ben. Upon further questioning about his symptoms, Ben stated that he has difficulty concentrating, but that his main symptom is the occurrence of “very strange thoughts” that have been quite troubling to him. Specifically, he feels too afraid to leave his apartment because he believes that the police will apprehend and arrest him as punishment for “what [he] did.” He worries that others blame him for killing people in the incident and that if he returned to work the passengers would turn on him, thus resulting in his apprehension. He states that he spends several hours a day worrying about the consequences of the incident and sometimes hears accusatory voices blaming him for hurting people and telling him that he is a “monster.” He reported that he has only heard these voices a few times in the past four weeks. As Dr. Tobin noted, though, no one was injured in the incident.
        Though Ben reports that he feels distressed about his recent psychological problems, he stated that he had no thoughts of hurting or killing himself. Ben also reported that he has been unable to get a full night’s sleep since the incident. At times he is unable to fall asleep, and when he does, frequent nightmares awaken him about the incident, and he feels that those he believes he killed in the incident are “haunting” him.
        Ben stated that although he was worried about what he had been experiencing recently, he had been too embarrassed to tell anyone, worrying that he was “going crazy.” As he had not spent time with friends or family, and had not been to work, the people in his life had been unaware of the extent of his psychological problems following the incident. Ben took a leave of absence for the first two weeks following the incident and has since been calling in sick daily. When the suggestion of psychological testing came up, Ben reports that he was hoping that it might help reveal the nature of his troubling symptoms.
         As you can see, the clinical interview is a key step in the diagnostic process because of the information it provides regarding the client’s current symptoms, history, and availability of social support. In addition, Dr. Washington used the interview as an opportunity to establish rapport with Ben. Over the course of a 30- to 45-minute interview, the clinician must help the client feel as relaxed as possible. Because the client is providing highly personal information, the clinician attempts to draw the client out with questioning that is respectful, but also matter-of-fact. The clinical interview is not like an ordinary conversation in that respect.
             Unlike the clinical interview, the structured interview provides standardized questions that are worded the same way for all clients. A structured interview can either provide a diagnosis on which to further base treatment or classify the client’s symptoms into a DSM disorder.
         One of the most widely used clinical interviews is the Structured Clinical Interview for DSM-IV Disorders (SCID). Though the title uses the word “Structured,” clinicians who administer the SCID modify the wording and order of questions to accommodate the particular individual whom they are examining. Clinicians use the SCID-I to make Axis I diagnoses and the SCID-II to make Axis II diagnoses. Both SCIDs are designed so that the clinician can adapt to the interviewee’s particular answers. The questions are worded in standard form, but the interviewer chooses which questions to ask based on the client’s answers to previous questions. For example, if a client states that she experiences symptoms of anxiety, the interviewer would follow up with specific questions about these symptoms. The interviewer would only ask follow-up questions if the client stated that she was experiencing anxiety symptoms. If she stated that she had different symptoms, such as depressed mood, then the follow-up questions would inquire further about her mood (First & Gibbon, 2004). The SCID-I takes 45 to 90 minutes to administer, depending on the complexity of the client’s symptoms.
        An advantage of a structured interview is that it is a systematic approach that is less subject to variations from clinician to clinician than an unstructured interview. Furthermore, anyone with the proper training can administer the SCID, not necessarily just licensed mental health professionals. This has practical value in that clients can receive initial screening prior to their beginning a course of therapy. Furthermore, there is a research version of the SCID that professionals can use to provide systematic diagnostic information across different investigations. Researchers can feel confident that an SCID-based diagnosis of a mood disorder means the same thing regardless of who conducted the study.

Table 3.2

MENTAL STATUS EXAMINATION

A clinician uses a mental status examination to assess a client’s current state of mind. In conducting a mental status examination, the clinician assesses a number of features of the client including appearance, attitudes, behavior, mood and affect, speech, thought processes, content of thought, perception, cognition, insight, and judgment. The outcome of the mental status examination is a comprehensive description of how the client looks, thinks, feels, and behaves.
        The Mini-Mental State Examination (MMSE) is a structured tool that clinicians use as a brief screening device to assess dementia. The clinician administers a set of short memory tasks and compares the client’s scores to established norms. If the client scores below a certain cutoff, the clinician then can (and should) continue to more in-depth testing of potential cognitive impairments.

      In Ben’s case, Dr. Washington noted that Ben was not experiencing an altered mental state at the time of the interview. He arrived on time and was alert and fully oriented. His conversational speech was normal in tone, rhythm, volume, rate, and prosody. The clinician noted that his receptive language appeared intact, and he was able to understand novel test instructions. However, his general appearance was slightly disheveled (i.e., his clothes were wrinkled and he was unshaven). His affect was appropriate, meaning that his apparent emotional state matched that expected in the situation, though he made a few jokes when he became frustrated during the administration of a test he found difficult. In general, though, throughout testing he cooperated with the examiner and appeared motivated and interested in the tests themselves. Dr. Washington decided not to administer a formal mental status examination. Instead he proceeded directly to further neuropsychological and personality testing.

INTELLIGENCE TESTING

Many professions (i.e., mental health, academia, business, government agencies) use intelligence (IQ) tests for a variety of assessments, including overall cognitive evaluation, diagnosis of learning disabilities, determination of giftedness or mental retardation, and prediction of future academic achievement. Clinicians also sometimes use IQ tests in the diagnosis of neurological and psychiatric disorders, in which cases they are a component of a more comprehensive assessment procedure. Finally, human resource departments often use IQ tests in personnel selection to evaluate the potential for employees to perform in specific conditions.
      Through intelligence testing, clinicians can obtain standardized scores that allow them to evaluate the cognitive strengths and weaknesses of their clients. The most commonly used intelligence tests in clinical settings are on a one-to-one basis, providing a comprehensive view of the client’s abilities to perform a range of perceptual, memory, reasoning, and speeded tasks.

Stanford-Binet Intelligence Test

First developed in the early 1900s by Alfred Binet, the Stanford-Binet is now in its fifth edition, known as the Stanford-Binet 5 (SB5). Children taking this test receive a deviation intelligence (IQ) score, calculated by converting their raw scores to standardized scores that reflect where a child stands in relation to others of similar age and gender. The average deviation IQ score is set at 100 with a standard deviation of 15. If a child receives an SB5 IQ score of 115, this means that the child stands at above the IQ of 68 percent of the population.
    In addition to yielding an overall IQ score, the SB5 yields scores on measures of scales labeled Fluid Reasoning, Knowledge, Quantitative Reasoning, Visual-Spatial Reasoning, and Working Memory (Table 3.4). These scales are intended to provide greater understanding of the child’s cognitive strengths and weaknesses not necessarily conveyed in an overall IQ score.

Wechsler Intelligence Scales

The first comprehensive individual test that researchers specifically designed to measure adult intelligence was the Wechsler Adult Intelligence Scale (WAIS). Originally developed in 1939 by David Wechsler as the Wechsler-Bellevue test, the WAIS, first published in 1955, is now in its fourth edition (WAIS-IV) (D. Wechsler, 2008). Researchers subsequently developed parallel tests for children based on the same format as the adult scales. Those currently in use are the Wechsler Intelligence Scale for Children–Fourth Edition (WISC-IV) (D. Wechsler, 2003) and the Wechsler Preschool and Primary Scale of Intelligence–Third Edition (WPPSI-III) (D. Wechsler, 2002).
        Wechsler originally sought to develop a tool for use in clinical settings, not just schools. He also believed that it was important to include both verbal and nonverbal tests. Originally he labeled these two categories “Verbal” and “Performance.” For many years, clinicians reported the WAIS scores in terms of these two categories of subtests. However, over time it became increasingly evident that these two categorical scores didn’t adequately capture the full complexity of intellectual functioning. Thus, the WAIS-III was substantially revised in 2008 to become the WAIS-IV, which now includes new tests and a different scoring system.
    The WAIS-IV, like its predecessors and the SB5, produces an overall IQ score based on an age-normed mean of 100 and standard deviation of 15. However, the full scale IQ is not as useful for clinical purposes as are scores on Verbal Comprehension, Perceptual Organization, Working Memory, and Processing Speed (Table 3.5). The intent of the WAIS-IV is to allow clinicians to examine in more depth the client’s cognitive functioning along these key dimensions.
        You can think of scores from the WAIS-IV as forming a triangle (see Figure 3.1). At the top is the Full Scale IQ (FSIQ), which reflects general cognitive functioning and is the best single predictor of school achievement on the WAIS-IV. Beneath the FSIQ score at the top of the pyramid are four index scores. Verbal Comprehension Index (VCI) assesses acquired knowledge and verbal reasoning skills. Perceptual Reasoning Index (PRI) measures visualspatial and fluid reasoning. Working Memory Index (WMI) measures the capacity to hold and process information in memory. Processing Speed Index (PSI) measures the ability to process nonverbal information quickly. Beyond interpreting the index scores, clinicians propose hypotheses about individuals’ performance based on an interpretation of clinical clusters, which are comprised of various combinations of individual subscale scores.
         Because the WAIS-IV is given on an individual basis, clinicians have ample opportunities to observe the test-taker’s behavior during the test, possibly gaining valuable diagnostic information to complement the test scores. In fact, the instructions for the WAIS-IV scoring include suggestions for the examiner to include behavioral observations such as the individual’s fluency in English; physical appearance; problems with vision, hearing, or motor behavior; difficulties with attention and concentration; motivation for testing; and any unusual behaviors that the test-taker shows.
         Table 3.6 shows Ben’s performance on the WAIS-IV. Ben’s FSIQ was 115, indicating that he has an above average level of performance (higher than 68 percent of the population). However, if you look across the entire pattern of his scores, you’ll notice that Ben demonstrated high variability among the index scores that make up the FSIQ. This type of variability suggests that there is a wide range to Ben’s cognitive abilities. The clinician found it noteworthy that Ben’s Processing Speed cluster score was low (higher than only 40 percent of the population). This would suggest that Ben struggles with the perception of visual patterns and stimuli, particularly when speed is a factor. The appearance of this low test score where he should have performed well given his job suggests that there may be injury to the areas of his brain involved in processing of spatial information.
            While administering the WAIS-IV, Dr. Washington carefully recorded Ben’s behaviors, further fleshing out the picture provided by the test scores themselves. Ben stated several times throughout the testing session that “they give you fake confidence early on,” referring to his frustration as the test became more difficult. Despite his frustration, Ben remained determined to complete the test. For example, he took nearly six minutes to complete the final Block Design item, and finally stated, “It doesn’t make sense—there aren’t enough blocks.” On the Matrix Reasoning subtest, Ben took nearly one minute for each response toward the end of the task. While completing the Figure Weights subtest, Ben commented on what the shapes in the stimulus book looked like and made several jokes throughout the subtest administration. On the Letter-Number Sequencing task, before giving his response to each item, Ben described how each correlated with the name of a different type of army ship or plane. As the tests became more difficult toward the end of the testing session, Ben appeared visibly restless and began to tap his fingers and tap on his legs. On tasks requiring verbal responses, Ben provided long elaborations, and when the test required a short answer, he would sometimes respond in a sing-song voice.FIGURE 3.1

Table  3.4

Table 3.5

Table 3.6

SELF-REPORT TESTS

A self-report clinical inventory contains standardized questions with fixed response categories that the test-taker completes independently either on paper or on the computer. Test-takers rate the appropriateness of the item to themselves on a fixed scale. These tests are objective in the sense that the scoring does not involve any form of subjective judgment on the part of the examiner. In fact, computers can both score these tests and produce brief explanatory reports. However, clinicians need to balance the advantages of their objectivity against the possibility that these reports are subject to the Barnum Effect. Because computer programs rely on a set of algorithms to produce their reports, they run the risk of being overly generic and not tapered to the particular test-taker’s idiosyncrasies.
         Nevertheless, a major advantage of self-report inventories is that they are relatively easy to administer and score. Consequently, large numbers of people can take these in an efficient manner. Clinicians can take advantage of the wealth of information on the validity and reliability of the better-known self-report inventories when interpreting the scores of their own clients.
       There are literally hundreds of self-report clinical inventories, many of which researchers developed specifically to investigate particular clinical problems or research areas. These inventories can range from a few dozen to several hundred items. As with tests such as the MMPI, clinicians must take care to ensure that the measures meet standards of reliability and validity before interpreting their scores.
        The most popular self-report inventory by far is the Minnesota Multiphasic Personality Inventory (MMPI), originally published in 1943. The current version of the test is the 1989 revision known as the MMPI-2 (Table 3.7). There are 567 true-false items on the MMPI-2, which are all in the form of statements that describe the individual’s thoughts, behaviors, feelings, and attitudes. The original intent of the MMPI was to provide scores on 10 so-called “clinical scales” corresponding to major diagnostic categories such as schizophrenia, depression, and anxiety. The text developers built an additional 3 “validity” scales into the test in order to guard against people trying to feign either exceptional psychological health or illness.
        In the decades after its publication, researchers and clinicians became aware of limitations in MMPI-2 clinical scale scores. They did not correspond to the original clinical categories, meaning that they could not interpret them as the scale’s developers originally planned (i.e., a high “Schizophrenia” scale score did not imply that the individual had a diagnosis of schizophrenia). Consequently, MMPI-2 users are incorporating the MMPI’s newer, restructured clinical scales (RC’s). In fact, the newest version of the MMPI is the MMPI-2-RF, published in 2008 (Table 3.8). The MMPI-RF is based entirely on the restructured scales. Containing only 338 items, this latest version of the MMPI-2 also provides scores for so-called “higher order” factors that indicate a client’s overall emotional, cognitive, and behavioral functioning.
           The Personality Assessment Inventory (PAI) (Morey, 1992) consists of 344 items organized into 11 clinical scales, 5 treatment scales, 2 interpersonal scales, and 4 validity scales. One advantage of the PAI is that clinicians can use it with clients who may not have the language or reading skills to complete the MMPI-2. A second advantage is that, unlike the MMPI, one calculates the validity scale independently of any of the content scales.
                The SCL-90-R (Derogatis, 1994) measures the test-taker’s current experiencing of 90 physical and psychological symptoms. One advantage of the SCL-90-R is that it focuses on the client’s current status rather than asking about symptoms over a previous period of time. Consequently, clinicians can track the progress of their clients over multiple occasions.
          Less oriented toward clinical use is the NEO Personality Inventory (Revised) (NEOPI-R)(Costa & McCrae, 1992), a 240-item questionnaire that measures five personality dimensions, or sets of traits. The scales are designed so the test-taker can complete them as well as individuals who know the test-taker, such as spouses, partners, or relatives (Form R). People use the NEO-PI-R less in clinical settings than in personality research or in personnel selection, although it can be of value in describing a client’s “personality” as distinct from the client’s symptoms.
        Clinicians and researchers may also use specific self-report inventories designed to investigate specific disorders or research questions for which a general test may not be as relevant. There are literally hundreds of these developed for such specific purposes. These inventories may also supplement more general assessment methods.
           You can see Ben’s MMPI-2 scores in Figure 3.2. His scores on the Paranoia scale were slightly elevated and he endorsed several critical items relating directly to psychosis such as “I have no enemies who really wish to harm me” (False), “I have strange and peculiar thoughts” (True), and “At times my thoughts have raced ahead faster than I could speak them” (True). According to his responses, he may have unusual thought content and may often feel suspicious that others are saying bad things about him. As a result, he may feel disconnected from reality. He may believe that his feelings and thoughts are controlled by others. His abnormal thought content was evidenced at times throughout the WAIS-IV administration in his unusual reactions to certain test items. His scores on the MMPI-2, however, indicate that he does not tend to be impulsive or take physical risks and generally follows rules and laws. These may be protective factors for Ben in that he may be able to maintain some control of his abnormal thoughts, which may differentiate him from those with diagnosable psychotic disorders.
        From looking at Ben’s scores, Dr. Washington concluded that his limited coping resources may be a more situational than long-standing problem. Further, his clinical profile suggests that he may be excessively sensitive and overly responsive to the opinions of others. He may overemphasize rationality and be moralistic and rigid in his attitudes and opinions. As a result, he may be argumentative and have a tendency to blame others and act suspicious, hostile, and guarded in relationships. This may account for his report of having few close friends at school and his preference to be alone in his dorm room.
          Based on his scores on the MMPI-2, it appears that Ben has a traditional sense of masculinity and may have stereotypically masculine preferences in work, hobbies, and other activities. In the clinical interview, he reported having had no previous significant romantic relationships with women, which may be a result of his tendency to be guarded and hostile in his relationships with others. Ben’s scores on the MMPI-2-R suggest that he does endorse aberrant experiences, but his score on persecution is within the normative range. He also received scores above the norm on somatic complaints.
        Clinicians typically interpret the MMPI-2 scores in the context of other test scores. They also may use the content scales to flesh out the profiles provided by the basic 10 clinical scales of the MMPI-2. The Restructured MMPI-2 also provides a different perspective on a client’s current psychological state, because the content scales provide a more descriptive summary of the client’s symptoms. In Ben’s case, Dr. Washington noted that his score was high on the Demoralization scale, suggesting that Ben felt discouraged and hopeless about his current life situation.

Figure 3.7

table 3.8

figure 3.2

PROJECTIVE TESTING

A projective test is a technique in which the examiner asks the test-taker questions about an ambiguous item. The underlying idea behind projective tests is that people cannot or will not provide accurate statements on self-report inventories. For example, clients may not wish to say that they are experiencing unusual symptoms or have qualities that they deem negative. On projective tests, clients may be less guarded about their responses because they don’t know how the assessor will interpret their answers. Projective tests are most useful when combined with self-report inventories rather than used as the sole basis for diagnosing or evaluating a client.
        The most famous projective technique is the Rorschach Inkblot Test, named after Swiss psychiatrist Hermann Rorschach, who developed the method in 1911. To administer the test, the examiner shows the test-taker a set of 10 cards (5 black and white, 5 with color), one by one. The test-taker’s job is to describe what the inkblot looks like. Although the method sounds simple enough, over the last century researchers and clinicians continue to refine the scoring methods.
         The Thematic Apperception Test (TAT) presents test-takers with a very different task than does the Rorschach. Test-takers look at black-and-white drawings that portray people in a variety of ambiguous situations. Their task is to tell a story about what is happening in each scene, focusing on such details as what the characters in the picture are thinking and feeling. The TAT’s original purpose was to evaluate motivation such as the need for achievement or need for power. Like the Rorschach, its use has evolved over time and clinicians can administer it as part of a larger test battery.
     In Ben’s case, Dr. Washington decided not to conduct projective testing until he completed the neuropsychological assessment. Clinicians typically do not administer these instruments as part of a standard battery, particularly if there is the potential that the client’s symptoms are related to a trauma or injury.

BEHAVIORAL ASSESSMENT

Unlike psychological tests, behavioral assessments record actions rather than responses to rating scales or questions. The target behavior is what the client and clinician wish to change. Behavioral assessments include descriptions of the events that precede or follow the behaviors. We call events that precede the behavior antecedents and events following the behavior consequences.

         For example, a child in a classroom may be unusually disruptive immediately following recess, but not immediately following lunch. When clinicians record behavior in its natural context, such as the classroom or the home, this is called in vivo observation. However, it’s not always possible or practical to conduct an in vivo observation. The teacher or a teacher’s aide is most likely too busy to record the behavior of one child, and having a clinician in the room would create a distraction or influence the behavior he or she is observing.
        Analog observations take place in a setting or context such as a clinician’s office or a laboratory specifically designed for observing the target behavior. A clinician assessing the disruptive child would need to arrange a situation as comparable as possible to the natural setting of the classroom for the analog observation to be useful.
           Clients may also report on their own behavior rather than having someone observe them. In a behavioral self-report the client records the target behavior, including the antecedents and consequences of the behavior. Self-monitoring is a form of behavioral self-report in which the client keeps a record of the frequency of specified behaviors, such as the number of cigarettes he or she smoked or calories he or she consumed, or the number of times in a day that a particular unwanted thought comes to the client’s mind. Clinicians may also obtain information from their clients using behavioral interviewing in which they ask questions about the target behavior’s frequency, antecedents, and consequences.

MULTICULTURAL ASSESSMENT

When psychologists conduct an assessment, they must take into account the person’s cultural, ethnic, and racial background, performing a multicultural assessment. Clinicians evaluating clients who speak English as a second language, or do not speak English at all, must ask a number of questions: Does the client understand the assessment process sufficiently to provide informed consent? Does the client understand the instructions for the instrument? Are there normative data for the client’s ethnic group? Even if clients appear as fairly fluent, they may not understand idiomatic phrases for which there are multiple meanings (Weiner & Greene, 2008).
         Publishers of psychological tests are continually re-evaluating their instruments to ensure that a range of clients can understand the items. At the same time, graduate trainees in clinical programs are trained to understand the cultural backgrounds of the clients who they assess. They are also learning to evaluate assessment instruments critically and to recognize when they need further consultation (Dana, 2002).

NEUROPSYCHOLOGICAL ASSESSMENT

Neuropsychological assessment is the process of gathering information about a client’s brain functioning on the basis of performance on psychological tests. Clinicians use neuropsychological assessment measures to attempt to determine the functional correlates of brain damage by comparing a client’s performance on a particular test with normative data from individuals who are known to have certain types of injuries or disorders. There is no one set procedure for conducting a neuropsychological assessment. Particular clinicians may have preferences for certain tests, but these preferences are not set in stone. Moreover, neuropsychologists typically choose tests that will help them understand the client’s presenting symptoms and possible diagnoses. The client’s age is another factor that the clinician takes into account. Tests appropriate for older adults are not necessarily either appropriate or useful for diagnosing a child or adolescent.
      Certain neuropsychological tests are derived from or the same as tests on the WAIS-IV, such as Digit Span (used to assess verbal recall and auditory attention) and Similarities (used to assess verbal abstraction abilities). Developers think that each of these tests is related to brain damage in particular areas. Developers have also created other tests, such as the Trail Making Tests, also called “Trails.” Figure 3.3 shows a sample item from the Trail Making Test A. This test evaluates frontal lobe functioning, and focuses on attention, scanning of visual stimuli, and number sequencing.
        In a neuropsychological assessment, the clinician can choose from tests that measure attention and working (short-term) memory, processing speed, verbal reasoning and comprehension, visual reasoning, verbal memory, and visual memory. A number of tests evaluate what clinicians call “executive function,” the ability to formulate goals, make plans, carry out those plans, and then complete the plans in an effective way. There are a variety of available tests within each category. If a clinician wishes to investigate one area in depth for a particular client, then he or she will administer more tests from that category.
       There are a large number of tests that measure visuospatial ability. Many neuropsychologists rely on the Clock Drawing Test (Sunderland et al., 1989) a simple procedure that involves giving the client a sheet of paper with a large predrawn circle on it. Then the examiner asks the client to draw the numbers around the circle to look like the face of an analog clock. Finally, the examiner asks the client to draw the hands of the clock to read “10 after 11.” The clinician then rates the client’s drawing according to number of errors. The most impaired clients are unable to reproduce a clock face at all, or make mistakes in writing the numbers or placing them around the clock.
         The Wisconsin Card Sorting Test (WCST) (see Figure 3.3) requires that the client match a card to one of a set of cards that share various features. Originally developed using the physical cards, the clinician typically administers the test in its computerized format. The test requires that the client shift mental set because the basis for a correct match shift s from trial to trial. In the example we’ve shown in Figure 3.3, the client could match the card on the basis of color, number of items, or shape. The profession regards the WCST as a test of executive functioning (Rabin, Barr, & Burton, 2005) that is sensitive to injury of the frontal lobes, but also assesses damage in other cortical areas (Nagahama, Okina, Suzuki, Nabatame, & Matsuda, 2005).
          Neuropsychologists use the Boston Naming Test (BNT) to assess language capacity. Containing 60 line drawings of objects ranging in familiarity, clinicians can use the test to examine children with learning disabilities and adults who suffer from brain injury or dementia. Simple items are those that have high frequency, such as a house. The client must choose from among four choices to identify correctly the object (e.g., chimney, church, school, and house).
        The Paced Auditory Serial Addition Test (PASAT) assesses a client’s auditory information processing speed, flexibility, and calculation ability. The tasks on the PASAT require that the client keep a running total of a series of numbers. For example, the examiner reads the numbers “6” and “8”; the correct total is 14. If the next number were “3,” then the correct total would be 17. The client must respond before hearing the next digit to score the response correctly. In addition to its usage to assess traumatic brain viduals with multiple sclerosis (Tombaugh, Stormer, Rees, Irving, & Francis, 2006).
       Other neuropsychological tests investigate a variety of memory functions, for example, the Wechsler Memory Scale, now in its fourth edition (WMS-IV). The WMS-IV includes tests of working (short-term) and long-term memory for visual and verbal stimuli. Examiners can choose from among the WMS-IV subscales according to which areas they believe are most critical to evaluate in particular clients. For example, when testing an older adult, the examiner may use only the scales assessing Logical Memory (recall of a story), Verbal Paired Associates (remembering the second in pairs of words), and Visual Reproduction (drawing a visual stimulus). However, neuropsychologists are cautious in accepting the newer versions of both the WMS-IV and the WAIS-IV, as they are too new to have accumulated sufficient validational data.
         Increasingly, neuropsychologists are relying on computerized test batteries, which are easier to administer than paper-and-pencil tests. One advantage to computerized testing is that it provides the opportunity for adaptive testing, in which the client’s responses to earlier questions determine the subsequent questions presented to them.
         The Cambridge Neuropsychological Testing Automated Battery (CANTAB) consists of 22 subtests that assess visual memory, working memory, executive function and planning, attention, verbal memory, and decision making and response control. Before deciding whether to move to a computerized test, the clinician must weigh the advantages of ease of administration and scoring against the potential disadvantages that may exist for clients who are disadvantaged in their ability to use computers, such as young children (Luciana, 2003). However, given the relatively rapid growth of this field, more extensive normative data will be available that will allow clinicians to feel more confident about their utility.
           Dr. Washington chose to administer tasks that would be sensitive to the type of injury that Ben might have sustained given his low WAIS-IV Coding score, which suggested that Ben may have suffered brain damage that led to changes in his ability to focus his visual attention and perform quickly on a psychomotor speed task. Unfortunately, because Dr. Washington saw Ben two weeks aft er the possible brain injury, Dr. Washington could not administer the Glascow Coma Scale (GCS), a common test that clinicians use in cases of possible traumatic brain injury. Included in the GCS are ratings, for example, of the individual’s ability to hear and obey commands, open the eyes, and speak coherently.
          Ben’s completion time on Trail Making Test Part A was in the marginally impaired range. On the Clock Drawing Test, Ben received a score of 5 out of a possible 10, erroneously crowding the numbers at one end of the clock. He received a score within the normal range on the PASAT, suggesting that the injury did not affect his auditory attentional functioning. On the WCST, Ben showed evidence of perseverative errors, meaning that he was unable to switch mental set in categorizing the cards according to different criteria. Ben’s performance on the WMS-IV was within normal range, a finding consistent with his relatively high scores on the Verbal scales of the WAIS-IV, indicating that he has not suffered either short- or long-term memory loss.

Toward the DSM-5

Changes in Diagnostic Assessments

The DSM-5 Task Force is recommending a major shift in diagnostic assessments. First, they propose including dimensional assessments along with the existing categorical diagnoses. Such assessments would allow clinicians to employ “measurement-based care,” meaning that they could more readily quantify the impact of treatment. The technology in this new approach would include traditional paper-and-pencil measures, but would also expand to computerized adaptive testing. Dimensional assessment also makes it possible to evaluate a client’s functioning in areas that span several diagnostic categories. For example, clinicians could measure depression, suicidality, and anxiety, symptoms that occur in people who would have a variety of possible diagnoses. By having access to this information, clinicians could establish a baseline of the client’s functioning. As the client continues through treatment, the clinician could then measure whether changes are occurring in these dimensions of functioning.
          The changes that the committee is considering in DSM-5 assessments would mean that clinicians would ask clients to complete a brief “Level 1” screening in which they rate themselves in several key areas, simply described, and on a 0 to 4 or 5 scale. If they answer positively to these screening questions, they would then receive a “Level 2” set of ratings that would more precisely tap into that symptom category. For example, clients who rate themselves as “5” on anxiety, would then move on to questions that ask them to rate themselves on more specific symptoms such as whether or not they experience panic attacks.
      One advantage of the new system is that clients (or informants) can answer these questions on their own without the need for a clinician to administer these questions. This would mean that clinicians could evaluate clients in a wider range of treatment settings. The National Institute of Health is supporting this effort as part of its larger “PROMIS®”initiative, a “Patient-Reported Outcomes Measurement System” (http://www.nihpromis.org). PROMIS will provide an opportunity to improve health care outcomes by “giving decision makers hard data on how health care affects what patients are able to do and how they feel.” Clearly, with the development of more objective indexes of client functioning across a range of physical and psychological domains, clinicians will better be able to evaluate the effectiveness of their approaches both to assessment and treatment.

 

figure 3.3

 

NEUROIMAGING

Neuroimaging provides a picture of the brain’s structures or level of activity and therefore is a useful tool for “looking” at the brain. There are several types of neuroimaging methods that vary in the types of results they provide.
    The electroencephalogram (EEG) measures electrical activity in the brain. EEG activity reflects the extent to which an individual is alert, resting, sleeping, or dreaming. The EEG pattern also shows particular patterns of brain waves when an individual engages in particular mental tasks. Clinicians use EEGs to evaluate clients for conditions such as epilepsy, sleep disorders, and brain tumors.
Computed axial tomography (CAT or CT scan) is an imaging  method that clinicians and researchers use to provide an image of a cross-sectional slice of the brain from any angle or level. CT scans provide an image of the fluid-filled areas of the brain, the ventricles. The method is useful when clinicians are looking for structural damage to the brain. Magnetic resonance imaging (MRI) uses radiowaves rather than X-rays to construct a picture of the living brain based on the water content of various tissues. The person is placed inside a device that contains a powerful electromagnet. This causes the nuclei in hydrogen atoms to transmit electromagnetic energy (hence the term magnetic resonance), and activity from thousands of angles is sent to a computer, which produces a high-resolution picture of the scanned area. The picture from the MRI differentiates areas of white matter (nerve fibers) from gray matter (nerve cells) and is useful for diagnosing diseases that affect the nerve fibers that make up the white matter. However, like the CAT scan, the MRI produces static images so it cannot monitor brain activity. Positron emission tomography (PET) scan, or a variant known as single photon emission computed tomography (SPECT), does provide images of brain activity. Specialists inject radioactively labeled compounds into a person’s veins in very small amounts. The compounds travel through the blood into the brain and emit positively charged electrons called positrons, which they can detect much like X-rays in a CT. The images, which represent the accumulation of the labeled compound, can show blood flow, oxygen or glucose metabolism, and concentrations of brain chemicals. Vibrant colors at the red end of the spectrum represent higher levels of activity, and colors at the blue-green-violet end of the spectrum represent lower levels of brain activity. Proton magnetic resonance spectroscopy (MRS) is another scanning method that measures metabolic activity of neurons, and therefore may indicate areas of brain damage (Govind et al., 2010).
    Functional magnetic resonance imaging (fMRI) provides a picture of how people react to stimuli virtually in real time, making it possible to present stimuli to an individual while the examiner monitors the individual’s response. Researchers are increasingly using fMRIs to understand the brain areas involved in the processing of information. One major advantage of the fMRI is that it does not require injection of radioactive materials, like the PET or SPECT scans. However, because the fMRI uses magnetism to detect brain activity, people with artificial limbs made from metals such as titanium cannot use this testing method.
      Brain scans can produce evidence of specific areas of damage, but they do not necessarily correspond to a specific loss of behavioral functioning (Meyers & Rohling, 2009). Upon evaluating Ben through neuropsychological testing, Dr. Washington decided to administer a CT scan (de Guise et al., 2010), because Ben showed some signs of frontal lobe damage (personality changes, some perseveration on the WCST, and marginal errors on the TMT). However, it was necessary to rule out parietal lobe damage, which can also contribute to this performance pattern as well as to visual attentional deficits. The CT scan revealed that Ben had suffered a traumatic brain injury as a result of the incident, perhaps in the form of a brain hemorrhage.