Flashcards in Rehab and Single Case Design Deck (18):
What are the four main approached to remediation of cogntive impairments?
1. surgical (e.g., stem cells)
2. Pharmacological (e.g., dexamphetamine)
3. Stimulation (TMS/TDCS) (still a long way to go, and in combination with behavioural)
4. Behavioural [we will focus on behavioural]
What key assumption underlines the cognitive neuropsychological APPROACH to rehabilitation?
"Treatments will be maximally effective only when the direction of treatment is determined by precise knowledge of the individual's processing strengths and weaknesses."
Use of models of normal processing as an aid to rehabilitation. Model-based assessment can be used to define what the specific of the treatment should (and should not) be.
Thus, it starts with the discovery of the individual's processing stengths and weaknesses (e.g., damaged 'modules') via a neuropsychological assessment!. NOT just based on surface symptoms, as these can arise in a number of ways. BUT, this does not necessarily tell us WHAT TO DO to treat the impairment (e.g., which dyslexia programme to use).
Taking a cognitive neuropsychological approach doesn't allow you to _______
But it does aid rehabilitation by ______
Doesn't AID rehab:
*knowledge of what is going wrong (via cognitive neuropsychological assessment) does not tell us WHAT TO DO to treat the impairment.
*Analysis won't help strategic decisions regarding the importance of each impairment to the individual (human-factor!)
- able to use detailed theory-drive assessment to detemine scope of impairment and the ways it parralels normal performance
-Knowledge of what is going wrong, limits the theoretically motivated choices for rehabilitation that are left open to the therapist
- avoid treating intact processes (e.g., can reject reading programs that focus on phonics when a child has a reading problem but with good phonic skills!).
- might suggest an order of attack (fix what comes first in the model!)
- able to use data and theory from unimpaired individuals to intepret our treatment results and constrain treatment parameters (e.g., treating word retrieval in aphasia)
- Expand on models of cognition and treatment mechanisms based on treatment outcomes
Why do we need to have a 'Theory of therapy'? (in addition to theories of cognitive processes e.g., language)
[What are the minimal requirements for a 'theory of therapy'?]
While cognitive neuropsychological assessment is a GOOD START in developing an approach to treatment It is NOT ENOUGH. It gives us an idea of how the cognitive model works (e.g., language system) and a hypothesis of how it is broken (and some ideas how we might start to remedy it) - BUT we also need to have an idea of how THERAPY will work
For example, whether therapy results in change pre-to-post ; how that change came about (what is the mechanism, what has the treatment actually done); and what individual factors affected that outcome
[From lecture slides: minimal requirements for a theory of therapy]
1. Analysis of pre and post therapy (damaged) states of the cognitive process in question (E.g., language system)
2. A specification of how the change between pre-and-post occured (E.g., treatment which improves word retrieval - is it because they 're-learned' lost words, or were the words inaccesible and now it is retrieved easily)
3. Determination of which characteristics of the individual and their profile of impairments are relevant to the treatment of outcomes (e.g., what other cognitive abilities need to have for the treatment to work, e.g., intact self-monitoring/memory etc).
We know far less about 2 and 3; relative to 1.
How can cognitive neuropsychology be used to help inform treatment? (2 things)
a) by helping interpret treatment results intensive treatment may not be warranted OR cumulative semantic inhibition (harder to retrieve subsequent words from same semantic catergory) --> treating in semantically related blocks is not beneficial)
What is a 'restorative' approach to rehabilitation?
a) Improving the functioning of defective processes
b) reteaching of missing information, rules or procedures (or regaining retrieval of that information)
e.g., in a child with prosopagnosia who has poor recognition of internal face features [eyes, nose, mouth] - TREATMENT focuses on perception and analysis of facial features for familiar faces.
What is a 'compensation' approach to rehabilitation?
a) Teaching a different way to perform the same function using intact skills within the same cognitive domain. (e.g., in a child with prosopagnosia who has poor recognition of internal face features [eyes, nose, mouth] and good recognition of external face features [hair, face shape] can be taught to compensate using the latter ability.
b) Teaching a way to compensate for the lost function using different skills. (e.g., using a memory aid/pages for memory impairment)
Is compensation or restoration 'better'?
It depends on the situation. While restoration may seem like the ideal, some processes are very difficult to remediate (e.g., memory) and outcomes are often restricted to the items that have been treated (does not generalise # generalisation is more common with strategic approaches).
Thus, compensation can be a very practical and effective mechanism.
The most successfull treatment is one which effects improvement no only for th eitems use din therapy, but also for any other item, in any other context (unfortunately this is a VERY HARD OUTCOME TO ACHIEVE).
What does it mean to say a treatment 'works'?
i) things get better (performance improves/symptoms reduce) compared to before treatment
ii) This change is due to the treatment and not any other factor (e.g., placebo effects, hawthorne/charm effects, practice effects, bias in the experimenter)
What other factors can be responsible for pre-post changes in a treatment program?
Placebo effects - the tendency of any medication or treatment, even an inert or ineffective one, to exhibit results simply because the recipient believes that it will work
Hawthorne/charm effects - The alteration of behaviour by the subjects of a study due to their awarenes sof being observed/positive emotional effect due to the perception of a sympathetic or interested observer
Practice effects - Improved performance from repeated testing of the outcome measure
Bias in the experimenter -"i've worked really hard on this treatment, I really WANT it to work" :P
Why might randomised controlled trials not be the best measure of treatment outcomes?
What are some benefits to RCTs?
- RCTs can only answer the question "does the treatment improve performance ON AVERAGE?", Thus RCTs/group studies mask individual differences in outcome, some people (Due to their individual characteristics/other factors) may benefit, others may not.
- RCT's only control for other causes of improvement (e.g., placebo, hawthorne etc) at the GROUP level, for any one individual it might well just be placebo etc.
*Random allocation - reduces selection bias (but not always when small ns
*Blinding of experimenters, participants, assessors to group - reduces performance bias.
What is the main question when deciding on a
Which particular tretment will be effective for my particular patient?
#There will never be a 'prescription' guide for treatment, we need to evaluate treatments for every individual we see - thus we need to evaluate if it is working or not with a particular individual (i.e., everycase needs to be a single-case design).
What are the benefits to case studies?
Answers the questions "Does the treatment significantly improve performance for this individual" (with there specific characteristics).
Requires REPLICATION of findings.
How might you control for recovery/practice effects in a single case design?
What doesn't this control for?
Test more than once before treatment starts - this allows you to establish a 'rate-of-change' and extent of variability without therapy.
Still possibility of placebo/hawthorne effects DURING the training/intervention phase.
How might you control for the possibility of placebo/hawthorne effects during training/intervention.
Include a set of items that are exposed/probed as often as treated items, but are not treated. (but could still be charm/placebo OR cumulative practice) [e.g., naming task]
Include a set of items [e.g., unseen items on the naming task] that are only included before and after. (can help distinguish effects of exposure from charm/placebo effects).
If there is greater improvement during treated than non-treated phases
If you have a treatment task and a control task (e.g., naming training control that is exposed and probed as often as the treated items) and the individual improves on both the treated task and naming control but NOT on an unseen naming control list (ie.., a list of unseen items only delivered/tested pre-and-post and not -trained!).
Than what are possible explanations for this finding?
*there is a greater improvement during treated than non-treated phases.
* The effect is due to mere EXPOSURE, rather than being due to the treatment (or placebo/spontaneous recovery).
If you have a treatment task and a control task (e.g., naming training control that is exposed and probed as often as the treated items) and the individual improves on both the treated task and naming control AND ALSO ON an unseen naming control list (ie.., a list of unseen items only delivered/tested pre-and-post and not -trained!).
* There is generalisation to other tasks
* we cannot exclude this being due to charm/placebo effects (everything improves and generalises due to charm/placebo!)
YOU NEED A CONTROL TASK THAT IS UNLIKELEY TO CHANGE! (e.g., memory task, when the treament is language based - but should not be at ceiling)