clinical psych exam two

PSY2301

goals of assessment and prediction

screening, diagnosis/case formulation, prognosis/prediction, treatment monitoring and evaluation

fundamental properties of psychometrics

standardized, reliable, valid, and useful

AUDIT

best available BRIEF screener for alcohol dependence; interview is better than self report; has all fundamental properties

what the AUDIT covers

consumption, drinking behavior/dependence, & alcohol related problems

cutoff value

score used to differentiate those who may have AUDIT versus those who may not (8 points)

sensitivity

the ability of a tool to correctly identify individuals with a disorder (true positive)

specificity

the ability of a tool to correctly identify those without a disorder (true negative)

clinical assessment

completely unstructured assessments where the clinician selects and measures risks factors according to experience and judgement

actuarial assessment

completely structured assessments where risk assessment is determined by prediction equation and established cutoff scores (linear regression)

VRAG

moderately predicts incarcerated mens violent recidivism

base rate

proportion of people who meet the criteria; base rates of violent recidivism are typically very low

overstate

we often_________ our ability to predict low frequency events

Clinical vs actuarial approach to A&P in clinical psychology

evidence overwhelmingly supports actuarial approach, but clinical approach overwhelmingly used in practice

“gold standard” result when comparing interview assessment and questionnaire

the decision from the clinical interview about whether the problem is present

completely clinical assessment and prediction of violence risk involves

relying solely on ones clinical judgement

structured interviews

generally provide more reliable information than unstructured interviews

mental status exam

a commonly used system of behavioral observation in psychiatric contexts

self-monitoring

is suggested by research that it is an empirically supported treatment for a number of clinical problems

issues with self-monitoring

inaccuracy, low validity, low reliability, distortion

singular model of intelligence (spearman)

intelligence is singular and reflects one ability; “g” for general (single) intelligence

plural model of intelligence (thurstone)

intelligence is made up of multiple abilities that may not relate to one another

examples of plural intelligence

verbal comprehension, logical reasoning, perceptual speed, numeric facility, word fluency, associative memory, spatial visualization

hierarchal model of intelligence

blends singular and plural models and is defined by a global/overall intelligence factor (“g-factor”) that includes relative ability specific factors

hierarchal

the model most intelligence tests are based on

uses of intelligence tests

academic planning, tracking cognitive changes over time, predict future achievement, influence approach to assessment, influence approach to therapy (vocab & written material)

WAIS-5

assesses cognitive functioning and adult intelligence across five indexes: verbal comprehension index, visual spatial processing index, fluid reasoning index, working memory index, processing speed index

verbal comprehension index (VCI)

verbal concept formation and verbal reasoning; aka one’s ability to extrapolate (why is a raven like a writing desk?)

visual spatial processing index (VSI)

spatial processing/perception and visual-motor integration; block designs and visual puzzles which break down barriers to the test

fluid reasoning index (FRI)

nonverbal abstract problem solving, inductive reasoning and logical thinking; figure weights

working memory index (WMI)

capacity to store, transform, and recall incoming info/data in short term memory; “repeat the last three digits I say”

processing speed index (PSI)

ability to rapidly and accurately process simple info; requires individual to hold & use a pencil to complete coding and symbol search tasks

FSIQ (full scale iq)

the intelligence quotient based on total performance aka the g-factor

general index (GAI)

based ONLY on scales from verbal comprehension, visual spatial processing, and fluid reasoning, which allows for estimate of intelligence without motor skills-related measures

WAIS-5 scores

individuals score is scaled relative to scores of same age peers in normative sample

stratification

what external validity is enhanced by

100

average IQ for age with the average range being between 90 and 109

>130

gifted WAIS-5 score

121-129

superior WAIS-5 score

110-120

high average WAIS-5 score

80-89

low average WAIS-5 score

70-79

borderline impaired WAIS-5 score

<70

impaired WAIS-5 score

WAIS-5 Psychometrics

high standardization (same questions and administrative strategy), high reliability for FSIQ and sub-scales (internal consistency and inter-rater reliability is high), good validity for FSIQ and sub-scales (measures what its supposed to measure), high utility (used in variety of contexts like schools or hospitals)

drawbacks of WAIS-5

time intensive, scores can be influenced by factors unrelated to intelligence (anxiety or testing conditions), questions about construct and content validity

what does the WAIS-5 predict

grades in school, number of school years completed, academic performance through first year of college, landing a job

genetic influences on WAIS-5

heritability estimates are often lower in low-socioeconomic families in the u.s. with intelligence clearly being polygenic

environmental influences on intelligence

lack of micronutrition & low SES decrease IQ scores while breast feeding and educational interventions in low SES children increase IQ scores

projective personality assessment

unstructured and ambiguous stimulii work to project unconscious; not very good psychometrics

objective personality assessment

administration of standard set of questions or statements with fixed set of response; potentially high psychometrics but malingering can happen

MMPI

most psychometrically-sound self report measure of personality and psychopathology; widely used to predict and asses clinical phenomena in a multitude of contexts

mmpi 1

challenges external validity of norms and developed using a purely empirical approach

mmpi 2

eliminated offensive items and revised outdated language from fist edition; developed using a theoretical approach

empirical approach

includes items that discriminate people who are and are not experiencing clinical phenomena of interest aka clinical and non-clinical groups; low content validity

theoretical approach

prioritizes content validity by including items that clearly assess the clinical phenomena of interest

mmpi 3

new items created using a blend of empirical and theoretical with the norms being updated to reflect population of the 21st century

t-scores

standardized score with an average of 50 and standard deviation of 10

>70 or 80

t-scores that are of potential clinical interest; anything higher may indicate an invalid profile

F-scale

infrequency scales aka over-reporting/overselling, possible malingering (i have no friends)

L-scale

lie scale or uncommon and excessive virtues and underreporting (i am the perfect person and nothing is wrong with me!)

K-scale

defensiveness scale and unrealistically positive (i don’t mind criticism at all)

RCd-Demoralization

measure of overall emotional discomfort common across multiple features of psychopathology; good inclusion criteria of distress we see across the DSM

strengths of MMPI-3

high standardization/reliability/concurrent & predictive validity, ability to detect untruthful responses, incredibly useful

drawbacks of MMPI-3

still pretty long, concerns that the validity scales are not as sensitive as MMPI2, reading and attention requirements, emphasis on pathology not strengths, relies on self report

what clinicians should do to create a full picture

pair MMPI with SCID and behavioral observation

sensitivity equation

true positive/(true positive + false negative)

specificity equation

true negative/(true negative + false positive)

>.80

ideally the amount we want specificity and sensitivity to be above but could be changed by calibrating cutoff values

naturalistic behavioral observation

observation within normal situational contexts and under naturally occurring reinforcements and consequences; good external and poor internal; inexpensive but may lose reliability

structured behavioral observation

observation in role-played scenarios in which one person engages in one or more simulated social interactions; poor external good internal; better reliability and more expensive

antecedent

what happens before person partakes in behavior

behavior

what person does/is doing

consequence

what happens after person partakes in behavior