Class Notes (806,895)
Canada (492,510)
Psychology (7,610)
PSYC31H3 (65)

PSYC31 chapter 6.pdf

15 Pages
Unlock Document

University of Toronto Scarborough
Konstantine Zakzanis

PSYC31 Chapter 6 – Neuropsychological Examination: Interpretation NATURE OF NEUROPSYCHOLOGICAL EXAMINATION DATA - Basic data of examinations = behavioural observation - In order to get meaningful data of patient’s behaviour  need to have made or obtained reports of many different kinds of observations including historical and demographic information Different kinds of Examination Data : Background Data - Are essential for providing the context in which current observation can be best understood - Need developmental and medical history, family background, educational and occupation accomplishment (or failures), and patient’s current living situation and level of social functioning - Examiner must taken into account a number of patient variables: sensory and motor status, altertness cycles and fatigability, medication regimen and likelihood of drug and alcohol dependency - Importance of background data when interpreting examination observation  e.g. test score of millwright must be at least average but more likely to achieve high average vs. Executive chief at least high average ability but many would perform at superior level o But keep in mind motivation to reach goal is important too  professor at average ability vs. Shoe clerk who has exceptional ability Different kinds of Examination Data : Behavioural Observation - Naturalistic observation can give useful info about patient’s function outside the formalized, highly structured and possibly intimidating examination setting o Rarely done so  but reports form nursing personnel or family members may help to understand what the examiner should look for - Especially important when justly formal examination finding conclude about person capability more or less than they actually are  this error likely to occur when examiner confounds observed performance w/ ability o Behavioural characteristics that comprise their adequate and sometimes even excellent skills are not elicited in usual examination o E.g. frontal lobe damage – show no cognitive deficits but is apparent to people who live them vs. Others who show deficits but can cope despite it using various strategies - Patient’s conduct in examination is useful too  documented & evaluated as attitudes towards the examination, conversation or silence, the appropriateness of their demeanour and social responses Different kinds of Examination Data : Test Data - Testing differ from other psychological data  it elicits behaviour samples in a standardized, replicable, and more or less artificial and restrictive situation - in the sameness a of test situation for each subject: o strength is ability to compare the samples b/w individual over to time or w/ expected performance level o weakness – is that observations are limited to behaviours prompt by test situation - examiner extrapolates from limited set of observations to the patient’s behaviour in real life situation o extrapolations likely to accurate as the observations on which they are based on are pertinent, precise and comprehensive, as situations are similar and as generalization are apt. - Most case  examiner rely on their common sense judgements and their piratical experience in their making test-based predictions about a patient’s real-life functioning o Studies on predictive validity and ecological validity of tests show many have good predictive relationship w/ variety of disease characteristic Quantitative and Qualitative Data - Observation expressed as either numerically (=quantities data) or descriptively (=qualitative data) - 2 approaches: - 1) Actuarial system exemplifies the quantitative method  relies on scores, derived indices and score relationships for diagnostic predictions - 2) Clinical approach – based on richly described observations w/out objective standardization - Together provide the observational frames of reference and techniques for taking into account, documenting and communicating the complexity, variability and subtleties of patient behaviour - Conditions necessary for actuarial predictions to be more accurate than clinical ones: o There are only a small number of probable outcomes (e.g. left or right cortical lesions, right cortical regions, diffuse damage, no impairment ) o Predictions variable be known (limit info that can be processed by formula to info on which the formula was based on) o Data from which the formula was derived be relevant to questions asked - Actuarial people over look the fact that in this era  most assessment are not taken for diagnostic purposes bur for the patient’s neuropsychological status & even if so  is unique ot individual case that can’t be know from simple formulas - Studies trying to support actuarial judgments  give examiners w/ only score index not patient live index - This debate extend into “fixed” vs. “flexible” approaches  practical and clinical experience support use of flexible selection of tests to address referral questions & problem /issues raised in consultation Quantitative and Qualitative Data : Quantitative data - Scores are summary statements about observed behaviour  obtained for any set of behaviour samples that can be categorized according to some principle o Scorer determined if each behaviour fits into a predetermined category & give into place on numerical scale - Commonly used scale for individual test item  two points: good/pass or poor/fail o 3 point scales - add “fair or barely pass” often used for grading ability test items - Scored tests w/ more than one item produces summary score  usually simple average of all individual items w/ occasionally some incorporating correction for guessing scoring so that final is not just simple summation - Final test score may misrepresent the behaviour under examination under at least 2 counts: o Based on only one narrowly defined aspect of set of behaviour sample o It is 2 or more steps removed from original behaviour - Global or aggregate or full scale by summing averages of test scores are 3-4 steps removed - Summary Index scores based on items scores that have had their normal rage restricted to just 2 points of either pass or fail or w/in normal limits or brain damaged (many steps removed) o if each 2 tests produces a different score pattern or normative distribution or sensitive to particular kind of brain dysfunction  then 2 test treated individually - test scores satisfy the need for objective, replicable data cast in form that permits reliable interpretation and meaningful comparisons. o Standardization allow for comparison of any one test performance score w/ patient’s all other scores or w/ any group or performance criteria - Also different behaviour e.g. writing vs. Visual reaction can be compared on single numerical scale  receive high on writing but low reaction time Quantitative and Qualitative Data: Problems in the evaluation of quantitative data - Interpretation of test scores  keep in mind the artificial and abstract nature of it - Reification of test scores can lead to overlook or discount direct observations o approach that minimize the importance of qualitative data can result in serious distortions in interpretation, conclusions, and recommendations - To be neuropsychological meaningful  scores should represent as few kinds of behaviour or dimensions of cognitive functions as possible - If test scores are over-conclusive e.g. summed scores or averaged test battery scores  becomes virtually impossible to know what behavioural or cognitive characteristic it represents o E.g. Memory quotient by averaging WMS was the same for very different kind of memory disorders o Same principle of multi-determinants holds for single test scores too as similar errors lowering scores in similar ways can occur for different reasons e.g. attention deficits and motor slowing - Range of observations by examiner is restricted by the test  especially in m/c paper pencil test and button/ mechanized activity that limits opportunities for self-expression o For most automated tests  how patients solves the problem is unknown or is relied on insubstantial info as heaviness or neatness of pencil marks - Fine-grained scaling by sophisticated instruments are not suited for many of behavioural symptoms of cerebral neuropathology o E.g. behaviour deficits that can consider species-wide norms (i.e. that occur at a developmentally early stage and performed effectively by all but most severely impaired school-aged children e.g. speech and dressing) are readily apparent o E.g. aphasic patient using a finely scaled vocabulary test is useless - Evaluation of test scores in context of direct observations is essential in neuro assessment  for many test scores alone give relatively little info about patient’s functioning o Important is often how patient solves the problem or approaches a task than scores  many was to fail and more than one way to pass - Different individuals may obtain the same test score on a particular test for very different reasons e.g. WIS-A oral format makes it test of attention and STM for most adults not just what is thought to measure (i.e. arithmetic)  called pitfall of face validity - Famous 2 studies that illustrated how psychologist drew erroneous conclusions from test scores faked by 3 pre-adolescent and 3 adolescent  2 interpretations of studies: o 1) valid interpretations of neuropsychological status cannot be accomplished by reliance on scores alone o 2) training, experience and knowledge are prerequisites for neuropsychological competence Quantitative and Qualitative Data : Qualitative Data - They are direct observations  about test-taking behaviours, appearance, verbalization, gestures, tone of voice, mood and affect, personal concerns, habits, and idiosyncrasies - Specific to test situations  observations of patient’s reactions to examination, approach to different kinds of test problems and their expressions of feelings and opinions about how they are performing o Wording of responses, nature & consistency of errors and success , fluctuations in attention and perseverance, emotional moment as they interact w/ examiner and w/ different kinds of test material are all qualitative data Quantitative and Qualitative Data: Limitations of qualitative data - There are different kinds of methodological and examination problems w/ this  all of standardizations, reliability and validity problems inherent in the collection and evaluation of data by single observer are threats to objectivity o Vagaries of neurological impairment also compound these problems – e.g. communication disability - When neuropsychological insult does not produce specific defects but rather reduces efficiency in the performance of behaviours that tend to be normally distributed e.g. response rate, or recall  standardized tests are better o E.g. many of early behavioural deficits of deterioration disease & much of expression of TBI or little strokes can occur as quantifiable rather than qualitative distortion - Development of internalized norms based on clinical experience build over theyears Quantitative and Qualitative Data: Blurring the line b/w quantitative and qualitative data - Hybrid technique - Quantification of the qualitative aspects of test responses  thus, adaptable for retest comparison and research - Quantified qualitative errors provide info about lateralized deficits that summary scores alone cannot give o E.g. quantifying broken configuration errors on Block Design discriminated seizure patients of left vs. Right hemisphere since right hemisphere made more errors while raw scores were identical b/w them Quantitative and Qualitative Data: Integrated Data - Integrated use of both data treat these 2 types of data as different parts of the whole data base - Need both  quantitative data w/out qualitative is meaningless in their individual applications but clinical not quantifiable testing lack comparability needed for many diagnostic and planning decisions Common Interpretation Errors - If this, then that: the problem of over generalizing - E.g. Walsh – 2 different diagnostically different groups e.g. right hemi damage and chronic alcoholism generate 1 similar cluster of scores (in the WIS test that is particular sensitive to right hemi damage)  by looking at only that score can lead to wrong conclusion Common Interpretation Errors - Failure to demonstrate a reduced performance: the problem of false negatives - Absence of low scores or other evidence of impaired performance can happen when brain damaged patients have not been given an appropriate examination - E.g. testing situation is controlled w/ minimal external stimuli  does not replicate real world circumstances which is challenging for neurologically impaired individual Common Interpretation Errors - Confirmatory Bias - Seek and value supportive evidence at the expense of contrary evidence when the outcome is presumably known Common Interpretation Errors - Misuse of salient data: over and under interpretation - Single dramatic finding (can be just be normal mistake) given much greater weight than not so interesting history that extends over years (e.g. base rate data) vs. Sometimes cluster of abnormal findings that corresponds w/ patients may provide important evidence of disorder even when most scores reflect intact functioning Common Interpretation Errors - Under-utilization or misutilization of base rates - Base rates important when evaluating diagnostic sings or symptoms  when sing occurs more frequently than conditions it indicates, using it as diagnostic indicator is erroneous o Regard any sing that can occur w/ more than one condition as possibly suggestive but never pathognonomic  e.g. slurred speech is abnormal but it is common among acute alcoholism, righ hemi damage etc., so have to do other test - Proper base rate studies of large scale ,prospective , done independently w/ several types of clinical disorders examined w/in population  these studies are rare but necessary Common Interpretation Errors - Effort effects - American Academy of Clinical Neuropsychology & National Academy of Neuropsycholgy support the use of effort testing in assessment as means to address the validity of an assessment o E.g. under performance due to insufficient effort can result in impair performance when it is not EVALUTAION OF NEUROPSYCHOLOGICAL EXAMINATION DATA Qualitative Aspects of Examination Behaviour - 2 kinds of behaviour are of special interest: - 1) Behaviour that differs from the normal expectations or customary activity for the circumstances o certain abbreviations are associated w/ certain neurological conditions & can alert examiner to look for other evidence of the suspected condition  e.g. setting the block on stimulus card or satisfaction w/ blatantly distorted response o can also tell about how patients think, how they perceive themselves, world & its expectations  e.g. who put the block on card  does not comprehend instructions & also not aware of this failure vs. Patient happy w/ incorrect response – not aware of failure but also show demonstrate self-awareness and some self-expectations this performance satisfied - 2) Gratuitous responses – comments patient make about their test performance, or while they are taking tests, or elaborations beyond the necessary requirements of task o Value of gratuitous responses well recognized in interpretation of projective test similar to neuro assessment  e.g. unnecessary details of picture tell of patient’s involvement w/ details at expense of practical considerations Test Scores - Can be expressed in a variety of forms  rarely use raw score instead report scores as values of a scale based on the raw scores made by standardization population (=group of people tested for obtaining normative data) o So each score becomes statement of its value relative to all other scores on that scale - Brooks et al. – 4 themes underlying the interpretation & reporting of test scores and neuropsychological findings: o 1) adequacy of the normative data for the test administered o 2) inherent measurement error of nay test instrument ( ceiling and floor effects) o 3) what represents normal variability o 4) what represents a significant changes over time w/ sequential testing Test Scores – Standard scores - Test scores can come from many different sources =complex - Techniques employed in assessment of diff aspects of cognitive functioning been developed at diff times, in diff places, on diff populations, for diff ability & maturity levels , w/ diff scoring and classification , and for diff purposes  these scores are NOT directly comparable w/ one another - So to make comparison , many disparate test scores must be convertible into one scale w/ identical units  this scale can be “lingua fanca” permitting direct comparison b/w many different kinds of measurements o Best scale in one derived from normal probability curve & based on SD  so most widely used scales is based standard score - SD is the measure of spread or dispersion of a set of scores around their mean  position of test score on SD unit scale defines the proportion of people taking test who will obtain scores above or below given score - Score based on SD (=standard score) can be estimated from a percentile (=most commonly used nonstandard score in adult testing) - Evaluation of test score depends on significance of their distance from one another or from the comparison standard  in general – differences of 2 SD or more may be significant vs. Diff of 1 - 2SD suggest a trend - Standard scores come in diff forms but are all translated on the same scale based on mean & SD - Z-score – basic unelaborated standard score from which all others can be derived  it represent in SD units the amount a score deviates from the mean of population (M) from which it is drawn o Mean of normal curve set as 0 & SD unit as 1  scores above M = positive value vs below M =negative value - Elaboration of z-score =derived scores  is same info as z but is expressed in scale units that are more familiar to most users than z scores - When standardization populations are similar, all of the of diff kinds of standard scores are directly comparable w/ one another (w/ SD & it relationship to normal curve as key) - Few test makers still report standardization data in percentile or IQ score equivalents  standard score approximation can be estimated o If standardization population is normally distributed- standard score equivalent for a percentile score can be estimated from table of normal curve functions Test Scores – Exceptions to the use of standard scores - Only compare scores from diff tests when standardization populations of eac
More Less

Related notes for PSYC31H3

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.