WEEK 3 - research methods Flashcards
What is the Independent variables (IV)
- Variable that is manipulated in a study
- E.g., amount of drug administered/ type of drug administered
What is the Dependent variables
- Variable that is measured in a study (to observe effect of IV)
- E.g., behaviour/mood/cognition/motor performance
What is a between-subject designs (independent groups)
- Experiments are conducted with 2 (or more) different
groups - (IV is operationalised as different groups)
- E.g., Group A given a new drug; Group B given CBT
- Usually interested in the difference between the means of
these groups.
What are the advantages of a between subject design?
- Easier & more time efficient to run
- Allows observation of variables that are not stable (i.e., habituation, practice effects, etc.)
What are the disadvantages of a between subject design?
- Many variables are unable to be controlled (e.g., systematic
differences between the groups)
–> need more participants so these effects average out - Results are presented in terms of group differences; masks changes within the individual
- Method of allocating to groups (randomised controlled trial; consecutive case design?)
What is a Within-subject designs (repeated measures)?
- Same participants involved in every level of the experiment
- IV is operationalised as different levels/testing occasions that all participant receive
- E.g., same participants are tested for attention span before
& after taking a drug - Usually interested in difference in the mean of the 2(+)
testing occasions
What are the advantages of Within-subject design?
- Requires fewer participants
- Each participant acts as their own control
What are the disadvantages of a Within-subject design?
- Not amenable to measurement of unstable variables
- Time & money
- May need alternate forms to assess DV to counteract practice effects
- May need to counteract cross-over effects (Landauer, 1975)
WHat are Control groups/conditions
- Used to ensure the effect observed is due to the variable wemanipulated & not some other variable
- Especially important in between-subject designs
- A control group will be identical to the groups being tested, except for the manipulation
For example:
Group A: has depression & given new antidepressant (exp. group)
Group B: has depression but receives no treatment (control group) - participants with same symptoms/severity, demographics etc. as experimental group
- If no difference is observed between the 2 groups, then it appears the new antidepressant did not work
What are Placebo controls
- sometimes in drug research a placebo condition will be used
- A placebo control condition will be similar to the experimental condition, except rather than receiving the drug (or no drug) they receive a substance containing no active ingredients (a placebo)
E.g., Group A: has depression, receives new antidepressant
Group B: has depression, & receives sugar pill
- Placebo controls are extremely useful for investigating whether any benefits derived from a drug are due to placebo effects
What are Three-groups designs
3 groups may be used
1. Given a new drug
2. Given a proven drug
3. Given a placebo
- Allows comparison between new drug & placebo
- Allows comparison between new & established drugs
- Allows experimenter to see if measures are sensitive enough to detect change (i.e., compare proven drug & placebo)
What are specificity of drug effects:
- e.g., include groups comprising people with different mental illnesses → does the drug work in depression exclusively?
What is longitudinal design
ffects may also be assessed over time (multiple measurements)
What is cross-sectional design
once of design
What is bias
- Bias results in systematic errors in measurement or prediction
- Bias can enter an experiment from many different sources
- Experimenter & participant expectations/bias (double blind
studies)
WHat are selction biases
- Demographic differences (age/gender)
- Cultural differences
- Personality differences
- Education etc. etc. etc
Rate which forms of study has the most control (1) to least control (7)
- experiment
- correlation
- Test
- survey
- case history
- Naturalistic observation
- Introspection
What are Ethical & legal constraints on drug research
In most drug studies there are severe ethical & legal constraints
* E.g., alcohol studies: legal constraints are relatively few (except min. age), but there are still ethical problems (e.g., how do people get home after experiment?)
- administering alcohol (& other legal drugs) to participants to participants can be costly (time & money)
- retrospective consumption questionnaires are often used (or daily consumption questions (diaries))
What are the potential varibles or domains we look for effects in during drug research?
changes to
* Arousal
* Cognition
* Perception
* Motor function
* Mood
* We might also want to measure side effects / biochemical or physiological drug effects
How/why do we measure perfomrence within a drug test
- The more measures we use:
- the more comprehensive our assessment
- the more costly our experiment (financial & time)
- the greater the need to consider experimental controls &
logistics (e.g., order we administer measures). - Assessing change/differences in performance across performance domains (e.g., arousal, cognition, perception, motor function) tells us something about how the drug works, & the effects it produces (primary & side effects)
Measuring arousal
- Arousal level changes naturally throughout the day, but can also be affected by a number of things – including drug use
- Most drugs are thought of as either ‘ stimulants’ or ‘depressants’, when given in large doses (at low doses this categorization does not hold)
- ‘uppers’ & ‘downers’ imply a relationship between arousal & mood
– not necessarily that clear - High arousal does not high activity
How is arousal measured?
- Since various drugs have an effect on arousal how we measure arousal is important
- Electroencephalograph (EEG): a measure of arousal/ detects potential differences between points on the scalp & neutral points on the body
- Other ways to measure arousal:
- Introspection?
- Unstructured Introspection
- Systematic Introspection – e.g., self-report scales (e.g., VAS)
- Ask an observer?
What are the levels of arsoual
Death
Coma
Sleep
Drowsy
Normal
Aroused
Highly excited
Mania
Convulsions
Death
Measuring mood
- Drugs can impact significantly on mood
- Can be studied experimentally (e.g., we can test drug effects by inducing mood states) or using a between-groups design (e.g., depressed vs. non- depressed patients)
- Measurements could be by self-report, doctor’s assessment, informant report, questionnaire, tests of biological markers of depression etc.
- Consider the advantages & disadvantages of these ways of
assessing drug effects on mood
Measuring perception
- Refers to the detection & integration of external stimuli
- Visual & auditory perception are most often studied
- Perception research often investigates the changes in the sensitivity of a person’s perception brought about by changes in the internal or external environment (e.g., drug use)
- These changes in sensitivity = thresholds (two types)
- Absolute Threshold –> Lowest value of stimulus detectable by an organ (e.g., eyes)
- Difference Thresholds –> Organ detects a change in level of stimulation
What are absoloute thresholds?
The lowest value of a stimulus that can be
accurately detected 50% of the time. Example:
- Vision: candle flame on a dark, straight road, at 60km
- Smell: 1 drop of perfume in a 6-room apartment
- Hearing: watch tick 6 m away
- Taste: 1 tsp sugar in 8 L water
- Touch: wing of a fly falling on your cheek from 1cm
What is difference thresholds?
- measure the ability of an individual to detect a
change in a stimulus - E.g., when can you tell the light is brighter than it was before?
- Critical frequency at fusion (CFF) – sensitive to drug effects
Measuring cognitive performance
- Ability to process, store & retrieve information (e.g., attention, memory, learning, etc.); can also include higher-level processes such as planning, set-shifting & response inhibition
- Often manipulated experimentally
–> Effect of drug at different stages of cognitive process can be assessed (e.g., may affect storage of information, or earlier attention processes)
Measuring motor performance
- Drugs can have a significant impact on motor tasks like coordination (e.g., alcohol)
- How might we measure these?
- ? simple or choice reaction time, tapping, pursuit rotor, etc.
Measuring side effects/physiological effects
Some methods of assessing side effects:
* Questionnaire (e.g., yes/no questions; rating scale etc.)
* Biochemical assay
* Doctor’s check-up/ physical examination
* Informant report
* Strengths & weaknesses of these methods
Some methods of assessing physiological effects: biochemical assays, MRI scans, PET scans
How do you identify a good research design?
What is the research question: does the chosen design allow that question to be answered?
- What are the strengths & limitations of the measures used to assess effects?
–> How sensitive/ appropriate are the measures?
–> How precisely can measurements be made? - What experimental controls have been put in place?
- Effect size: how meaningful is the change observed in terms ofeveryday functioning?
What are some special factors to consider with drug studies when trying to identify a good research design?
- Washout effects (esp. in w/g designs)
- Deprivation study designs (withdrawal)
- Level of drug administered
- Dose of drug used (e.g., compliance with instructions)
- Inclusion of biochemical assays
- Controls for routine / habitual or other drug use
- Non-specific treatment effects
- Special group selection factors (mild, moderate, & severe
depression?) - Timeframe for assessing effect
- Reporting of adverse events/non-compliance / attrition
- If in-vitro cellular trials are used: how generalizable are results to intact living systems?
- If animal trials are used: how similar is the species, apparatus, domain of function being tested?
What are the Major considerations in drug development?
- Medical need
- Commercial potential
- Feasibility for mass production
- “Orphan drugs”
What is the FDA approval processes: 4 stages
- Preclinical investigation (basic science; tests on cells &
animals; effects of drugs noted for side effects, toxic effects,
addictions, cancerous tumors, fetal deformities,
pharmacodynamics) - Clinical investigation (human testing, 4 phases, incl. 3 phases of clinical trials)
- Review of NDA (New Drug Application)
- Postmarketing studies
What are some Specific limitations of animal studies,
- Cats: markedly different sleep-wake cycles (e.g., sleep 65-80% of the time)
- Monkeys: develop obvious behavioural abnormalities from
confinement - Rabbits: different enzymes from humans (e.g., can eat some plants that would be poisonous to humans)
Despite limitations, animals studies generally provide a valid
indication of the likely drug effect in humans
In the drug development & animal testing (Stage
1:step 1) what is the first step of the process
Step 1. Determining toxicity
* Acute toxicity (for single doses)
* Subacute toxicity (short term use of drug)
* Chronic toxicity (longer-term drug use)
* Special toxicity (carcinogenic? teratogenic?)
In the drug development & animal testing (Stage
1:step 2) what is the second step of the process
Step 2. Pharmacological studies (direct & indirect)
- Direct:
– to locate direct measures of changes induced by drugs via tools such as: - rating scales, objective measures (e.g., “jiggle-cage”/ “tail
suspension test”), conditioning studies (see next slides), sites- of-action studies - Indirect:
- to identify markers or signs of effects rather than evidence of the effect itself or - to observe effects on behaviours that are induced in animals
What are the general principles of Direct pharmacological studies- conditioning
- Stimulant substances: rapid learning/acquisition of conditioned response, poor discrimination / ready generalisation, slow termination/extinction
- Depressant substance: slow acquisition/learning, poor
generalisation/easy discrimination, rapid extinction
Measuring conditioned behavior in nonhuman (what are the Schedules of Reinforcement)
- Schedules of Reinforcement: pattern that determines when
reinforcements are to be given - Ratio Schedules
- Fixed ratio (FR)
- Variable ratio (VR)
- Interval Schedules
- Fixed Interval (FI)
- Variable Interval (VI)
- Avoidance-Escape Task
- Punishment
- Stimulus Properties of Drugs
- stimulus properties
- ability to act as a discriminative stimulus in a discrimination learning task.
Measuring conditioned behavior in nonhuman (what are the Reinforcing properties of drugs)
- Rate of Responding
- Progressive Ratio
- Breaking point (organism will stop responding)
- Choice
- 2 levers; one has consequences. Exposure to 2 drugs over
separate sessions; then observe animal’s choice of drug A or B. - Conditioned Place Preference
- Animal will spend time in area of reinforcement
FDA approval processes
- Stage 1 Pre-clinical (animal studies to determine suitability as therapeutic agent; measurement of ED50 , LD50 & therapeutic index) – i.e., toxicity & pharmacological studies
- Stage 2 Clinical investigation (human volunteers):
- Phase 1: Pharmacokinetic & Safety Testing - human testing to determine toxicity & side effects in healthy human (paid) volunteers
- Phase 2: Small-scale Effectiveness Testing - human testing in clinical samples (i.e., patient groups) to assess potential therapeutic effects (+adverse)
- Phase 3: Large-scale Effectiveness Testing - expanded clinical trials using basic 3-group design
–> approval or rejection of new drug for licensing & marketing - Phase 4: accumulation of data on drug effects
In the Phases of human testing (clinical trials) what is phase one?
- Phase 1
- 20-100 healthy volunteers
- Informed consent mandatory
- evaluate side effects
- establish final, correct dosage
- pharmacokinetics studied
- generally takes 2 years
In the Phases of human testing (clinical trials) what is phase two?
Phase 2
* 100-300 patients who have disease drug intended to treat
* drug given on experimental basis
* determines therapeutic effect
* usually takes 2 years
In the Phases of human testing (clinical trials) what is phase three?
Phase 3
* 1000-3000 ill patient volunteers
* administered same way that it will be used on the market
* performance vs. other drugs currently on the market (3-group design with placebo; single or double-blind)
* usually lasts 2-4 years
Completion of Phase 3
* drug company submits all documentation to FDA in a New Drug Application (NDA)
* waits for final FDA decision → approval or denial
* only 20% NDAs receive final FDA approval for marketing
In the Phases of human testing (clinical trials) what is phase four? (postmarking study)
- After NDA review completed & approved
- Lasts on average, 15 years
- New drug placed on the market
- Surveillance: check for new harmful effects in larger group of humans
- Drug removed from market if serious problems occur
