Analogue behavioral observation is a behavioral assessment method in which a clinician observes a client's behavior in a contrived environment (e.g., a waiting room, play room, clinical setting) to assess variables hypothesized to influence behavior. Although analogue assessment is a direct measure of behavior, the target behavior is observed outside of the individual's natural environment. A special section of the journal Psychological Assessment (Vol. 13, No. 1) is devoted to a discussion of analogue behavior observation.
There are several classes of analogue behavioral assessment. Each class includes many different instruments for measuring behavior. A role-play is an analogue behavioral assessment class in which a client performs one or more behaviors in a contrived social situation. An experimental functional analysis is a structured observation session in which clients are observed while variables hypothesized to control or maintain a target behavior are systematically introduced and withdrawn. Family and marital interaction tasks is a class of analogue behavioral assessment in which members of a familial unit interact with one another on a task specified by the assessor that is relevant to the family or couple. Behavioral avoidance tests are often used to assess a person's response to a feared stimulus by measuring proximity to the stimulus (e.g., how close, in feet, a person with a snake phobia can come to a snake in a glass cage) or other approach behavior.
Other analogue behavioral observation techniques include enactment analogues, in which the individual is observed performing a newly acquired skill; contrived situation tests, in which a novel situation is presented to an individual to determine whether the individual can apply newly learned skills; think-aloud procedures, in which a person reports his or her thoughts during performance of a behavior; and response generation tasks, in which an individual engages in, or generates, one or more response options to a stimulus event. Variants of analogue behavioral observation have been used to assess a wide array of behavioral problems and phenomena. Hundreds of studies have employed the technique to assess marital interactions, child behavior problems, adult social functioning, and countless other behavioral disorders.
The utility of the analogue observation in helping the clinician generate hypotheses about functional relations depends on the degree to which the analogue context includes the elements that affect the behavior problem in the natural environment. For example, McGlynn and Rose in 1998 observed that anxious patients usually fear stimulus classes, rather than a single stimulus, and that one analogue session would be unlikely to include the myriad of feared stimuli present in the client's natural environment. Analogue behavioral observation is most likely to be a cost-effective alternative to naturalistic behavioral observation when the targeted behavior(s) occurs with high frequency in the analogue situation and is not reliably or accurately measured using other less costly assessment methods (e.g., questionnaires, rating scales).
Analogue observation sessions are more likely to provide important information regarding functional relations than behavioral rates. For example, couples may exhibit a higher frequency of negative comments toward one another during an initial assessment interview than they would at home where they could more easily avoid their spouse or partner. Alternatively, behaviors that occur only in private contexts may not occur at all in an analogue situation (e.g., battering, verbal threats). Researchers have long noted that partial-interval recording in analogue settings underestimates high rate responding, does not produce valid estimates of behaviors of short durations, and can misrepresent temporal relationships between behaviors and events.
The novelty of the assessment environment may make it more likely that irrelevant behaviors, rather than the target behavior, are observed. For example, a child who is defiant toward his parents may be stimulated by unfamiliar objects and toys in an observation room and may not interact with a parent as a result. Nonetheless, role-play activities may still permit observation of related variables and their dimensions (e.g., tone of voice, eye contact, frequency of reflective statements made by each partner).
All forms of analogue behavioral assessment require a coding or rating system in which the assessor quantifies a dimension of behavior. For example, Heyman and Vivian in 1993 developed the Rapid Marital Interaction Coding System (RMICS) to facilitate analogue observation of marital communication styles. In 2001 Heyman and colleagues found that observation periods as brief as 15 minutes were sufficient to obtain stable estimates of most RMICS codes in maritally distressed couples. However, in happily married couples, some variables and behaviors (e.g., dysphoric affect, withdrawal) were observed too infrequently to be reliably coded in brief laboratory interactions.
Analogue behavioral assessment methods and their respective coding systems have generally not been subjected to the type of psychometric rigor common for other psychometric instruments. For example, reports of analogue assessment methods often do not include information about (1) the goals of the analogue assessment, (2) the specific behaviors, functional relations, constructs, and facets to be measured, (3) the response modes and dimensions to be measured, (4) the methods of data collection, (5) how the specific scenarios, situations, and instructions might affect client behavior, or (6) a discussion about how dimensions of individual differences (e.g., sex, religion, age, ethnicity, sexual orientation) might influence responses. One of the major difficulties in evaluating the usefulness of analogue assessment, especially in the assessment of child behavior problems, is the lack of standardization demonstrated by most available measures.
Closely related to standardization is the issue of reliability, or consistency of measurement. The reliability of analogue behavioral observation coding systems is generally not well studied. For example, one researcher concluded that only 20% of published marital communication studies included reliability information for the constructs that were studied. Another researcher concluded that no test-retest reliability data are available for parent-directed-play coding systems or free-play behavior coding systems.
Additionally, the external validity of most analogue assessment measures has not been well investigated. External validity, in the analogue context, is the degree to which behavior observed in the analogue setting is representative of the client's behavior in his or her natural environment. Norton and Hope in 2001 concluded that the evidence concerning the external validity of role-play methods is "equivocal" and data on the external validity of other analogue assessment classes are either insufficient or absent.
B. Behavioral Rating Scales and Behavioral Checklists
A behavioral rating scale is an assessment instrument completed by a clinician or a third party (e.g., significant other, teacher, parent, peer) that includes items that assess one or more targeted client behaviors. A behavioral checklist is similar to a behavioral rating scale but often includes fewer items and may include di-chotomously scored response options. Many behavior rating scales and behavioral checklists have been standardized using a normative sample of individuals and aggregate raw data into standardized scale scores or global scores.
Behavioral rating scales are frequently divided into two classifications: narrow band behavior rating scales and broad band behavior rating scales. Narrow band behavior rating scales include items that sample from a small number of domains and are not intended to be global measures of an individual's behavior. Broad band behavior rating scales usually include more items, sample from a wider spectrum of behaviors, and are often used to screen for more than one disorder or behavioral syndrome.
For example, behavioral checklists and behavioral rating scales are the most popular methods of gathering information in assessing ADHD. Narrow band measures include the 55-item Social Skills Rating System, which divides item content into three narrow domains: problem behaviors, social skills, and academic competence. Another narrow band instrument is the Disruptive Behavior Rating Scale (DRS). The DRS includes item content covering oppositional defiant disorder, ADHD, and conduct disorder. Broad band behavioral rating scales include the Child Behavior Checklist (CBCL) and the Conners Parent and Teacher Rating Scales. Both the CBCL and the Conners Scales provide several scale scores and include versions for parents, teachers, and youths to complete. The popularity of these behavioral assessment methods can be attributed to their cost-efficiency, ability to quantify the opinions of important persons in a client's life, and their ease of administration. In addition, the most widely used instruments (e.g., the CBCL) rest on an extensive foundation of empirical literature that testifies to their reliability and validity.
Although behavioral rating scales and behavioral questionnaires are popular, it should be emphasized that they are indirect measures of behavior. As indirect measures, data collected using behavioral rating scales and behavioral checklists reflect a rater's retrospective impression of a client's behavior rather than an objective recording of the rate at which behavior occurs, as with naturalistic behavioral observation methods. Consequently, all behavioral rating scales and behavioral checklists are subject to rater bias regardless of the rigor with which the instrument is designed. Although indirect observation of behavior can be useful in behavioral assessment, its limitations need to be understood by the behavioral assessor.
In addition to being indirect measures of behavior, behavior rating scales and behavioral checklists rarely provide information pertaining to the functional relations of variables. Most behavior rating scales and behavioral checklists include items that measure topographical behavioral dimensions rather than functional relations. To some degree, the contextual variability of behavior can be addressed by having multiple informants complete the instrument provided each informant observes the client in different contexts (e.g., having a parent and a teacher complete the same rating scale). A thorough functional assessment, however, requires greater attention be paid to other variables that may be maintaining the behavior (e.g., the type of reinforcement received for an oppositional behavior; whether the problem behavior results in avoidance of an aversive event or situation).
Psychophysiological assessment involves recording and quantifying various physiological responses in controlled conditions using electromechanical equipment (e.g., electromyography, electroencephalography, electrodermal activity, respiratory activity, electrocar-diography). Which response or response system is measured depends on the purpose of the assessment. Psychophysiological measurement has been used to assess autonomic balance (e.g., heart rate, diastolic blood pressure, salivation), habituation to environmental stimuli, reactivity to traumatic imagery, orientation response, and other physiological systems.
Frequently, the behavioral assessor is not so much interested in the behavior measured by the equipment as what may be inferred from the behavior. For example, a large literature exists with regard to the psychophysio-logical measurement of responses to anxiety-eliciting stimuli. Keane and co-workers in 1998 showed that male military veterans with posttraumatic stress disorder (PTSD) exhibited greater changes in psychophysiological responding (i.e., increased heart rate, skin conductance, systolic and diastolic blood pressure) when presented a series of trauma-related cues than did veterans without PTSD. Other studies have found increased physiological responsivity in females with PTSD and increased heart rate responses to startling tones in individuals with PTSD.
Selection of the eliciting stimulus, or stimuli, and the response modes to monitor during a psychophysiologi-cal assessment are important considerations, especially when investigating responses to trauma-related cues.
For example, research has consistently shown that individuals are more physiologically reactive to scripts detailing their own personal experiences than to standardized scripts detailing either neutral scenes or traumatic situations. Synchronous responding to stimuli across physiological modes has not, however, been generally observed. For example, Blanchard, Hickling, Taylor, Loos, and Gerardi in 1994 found that heart rate and electrodermal activity, but not systolic or diastolic blood pressure, were responsive to audiotaped scripts describing a motor vehicle accident the participant survived.
All of these studies demonstrate how psychophysio-logical assessment can be used to identify behavioral differences in individuals, provide criterion-related validity for psychiatric diagnoses, and can be used as a clinical marker of client change since clinical improvement has been associated with changes in physiological indices. However, psychophysiological assessment is often cumbersome, expensive, and, depending on the client and his or her problems, may not provide information that sheds light on the functional relations of variables operating in a client's life. In addition, psy-chophysiological information does not inherently possess greater validity or is more objective than other behavioral assessment methods. Data from a psy-chophysiological assessment require interpretation in the context of convergent evidence from other assessment methods (e.g., a behavioral interview, analogue behavioral observation) in order for the information to be clinically meaningful.
Was this article helpful?