Lecture 2: Neuropsychological Assessment

Methods of assessment (research and clinical):

•        As Neuropsychologists, we are interested in sometimes unfortunate, sometimes dramatic and tragic instances when individuals have brain damage.

•        Brain Damage could be caused by (not an exhaustive list):

•        Accidents (e.g., car accidents, war injuries, sports injuries, work injuries)

•        Surgery (to “fix” another problem such as epilepsy)

•        Epilepsy (can lead to damaged tissue, especially in memory areas of the brain)

•        Lack of oxygen (sometimes happens at birth or after drug overdoses, also called anoxia)

•        Dementia and the “progressive” brain disorders

•        Infections (e.g., encephalitis)

dissociation:

Can remember what he did a few seconds ago

Can’t remember childhood events

= a single dissociation

Can remember what he did a few seconds ago

Can’t remember childhood events

Can’t remember what he did seconds ago

Can remember childhood events

= a double dissociation

What is our comparison? What is ‘normal’?

Enter, The Control Group or Normative Population

The brain: CT scan

• Computerised Tomography

• Sometimes referred to as CAT scans

• Essentially an X-ray of the brain

• Provides a rough guide to brain damage (e.g., bone structure and haemorrhages)

• Guides later assessment and therapy

The brain: PET

• Positron Emission Tomography

• A functional imaging technique

• Involves injecting participants with a radio-isotope that has a short half-life (i.e., it disappears after around 2 mins)

• It is able to detect oxygen and glucose uptake and also uptake of specific neurochemicals (e.g., dopamine)

• This method detects indirect activity via concentration of radiotracer in the brain (not neural activity)

The brain: structural MRI:

• High resolution

• Good impression of density of neurons

• Can determine the volume of certain regions (i.e., volumetrics)

• Small “slices” can be taken at different orientations. The one on the right is taken from the horizontal plane

The brain: functional MRI

• Known as fMRI

• Can be used in healthy or brain damaged individuals

• Is complementary to Neuropsychological data with brain injured patients

• E.g., we might see a link between semantic dementia regions and fMRI of regions involved in semantics

• Good spatial resolution (sometimes to the mm) but poor temporal resolution (averaged over around 5 secs)

The brain: electroencephalography

• Known as EEG; when averaged to look at the effect of an ‘event’, this is called ERP (Event-related potential)

• Can be used in healthy individuals or patients (i.e., it has research and clinical functions)

• It represents the electrophysiological activity in the brain of millions of neurons at millisecond accuracy

• It has poor spatial resolution but good temporal (time) resolution

What is the traditional Neuropsychological approach in relation to studying brain damaged individuals?

Typically:

• Compare performance of a brain damaged person with a small group of age and education matched neurologically normal controls

• The control group often has a small N size (i.e., small number of participants)

• Statistical methods exist to take the small N size into account

• A full Neuropsychological profile (of unimpaired and impaired functions) is presented (usually in one Table in the article)

• Results consist of comparing the patient with the small control group

• Comparisons can be made based on standard Neuropsychological tests (e.g., stroop, trail making test) or on specially designed tasks (e.g., mirror drawing)

Why are Neuropsychological tests useful?

• Standardised to a large group. (hint: The normal healthy data is typically called the ‘norms’ of a test)

• Can determine which aspects of cognition are impaired

• Can determine which aspects of cognition are unimpaired

• Can establish the difference between a poor function and a clinical or pathological impairment

• Can approximate premorbid (pre-injury) function

Types of neuropsychological tests:

•        Comprehensive Test Battery

• Includes a variety of different functions assessed within the same test

• Includes subtests of memory, attention, executive function etc.

• Therefore one can compare different functions within one patient, as well as comparing with the Norms.

• The most famous is the Wechsler Adult Intelligence Scale; it has a general IQ but also assesses

•        Individual Tests

• Assesses a specific function (e.g., language)

• Good if the patient does not have the necessary attention to follow a 1-2 hour test battery

• The most famous is the Wisconsin Card Sorting Card, a test of executive function

WAIS: an example

• First WAIS published in 1955 Recent one is version IV (2008). This is in the YSJ test library.

• Version III in 1997, then version IV in 2008.

• Measures general IQ as well as other functions

Why do we need to know about this?

• Raw scores are the actual amounts people get correct on a test (e.g., how many details of a story remembered)

• Raw scores are interesting to assess various test scores in one patient, but we usually need to see if their scores are clinically meaningful.

• In short, do their scores represent a dramatic enough decrease from what is expected from how the ‘normal’ population completes the task

The principles of neuropsychological testing and scoring:

• To work out standardised scores for one patient, we need to know how well people of a similar ability do on that task.

• Therefore, each test is completed by large cohorts of participants of different ages:

• E.g., 5-6 year olds, 55-65 year old, 75-85 year olds.

• When calculating an individual’s standardised score, we need the norms.

• This is because the raw score we would expect from an 8-year old will be different from that of a 30 year old, and an 80 year old.

WAIS: An example and the normal distribution

• Normal distribution is the same we use in statistics (e.g., t-test) – refer back to Statistics Module if needed.

• The mean is 100 for the main IQ score; or 0 for SDs and z-scores.

• We use this to establish how many standard deviations a patient is from the mean. A standard deviation is 15 points on the Wechsler tests. If they are -2 SDs below the mean this is usually taken as a clinically meaningful or clinically significant impairment. Therefore, an IQ of 70 is seen as below average. -3 SDs from the mean is seen as extremely impaired performance and equates to an IQ of 55.

• On individual tests of the WAIS, patients have a raw score which is converted into a standard or scaled score (mean=10, SD=3). Like the SD, z, and IQ, this is standardised.

• We also use percentiles. This tells us the percent of the general population who would score less than you. If 50% this means half the population would score less. If 90%ile. 90% would score less than you (this is superior). If 10%ile, this means 10% would score less than you (this shows an impairment)

• In working out of a patient is ‘impaired’ on a certain test we can compute z-scores, which is like a standard deviation from the small group mean.  Z-scores have a mean of 0 and a standard deviation of 1. They are statistically similar: They appear exactly the same on the Graph above!

Patient TB – standardised scores

•        Standardised scores are typically used in Neuropsychological tests. In these, 10 is usually a z-score of 0.0.

•        A standard score of 4 would represent 2 standard deviations from the mean, or a z-score of -2.0. 4 would indicate an impaired performance, as would any value below that (e.g., 0, 1,,2, or 3).

•        What would a standard score be for a z-score of -1.0?