Structural Neuroimaging (types)

day5

Computed Tomography (CT)

Magnetic Resonance Imaging (MRI)

Functional Neuroimaging (types)

day5

Positron Emission Tomography (PET)

Single Photon Emission Tomography (SPET/SPECT)

functional Magnetic Resonance Imaging (fMRI)

Magnetic Resonance Spectroscopy (MRS)

Event-Related Potentials (EEG/ERP)

PET Scan

day5

MRI (and fMRI)

day5

-uses magnetic fields and radio waves to first align all your particles together, then lets them go

- they tilt back to where they were at different speeds depending on the tissue they are.

-fMRI measures "blood oxygenation level dependent signal" (BOLD)

-people with free-floating metal (like aneurism clip, not brain screw) cannot have these done.

-those with anxiety or panic disorders have a difficult time getting these done

EEG/ERP

day5

Magnetric Resonance Spectroscropy

day5

Transcranial Magnetic Stimulation

(TMS) day5

Creates a strong magnetic field that can be used to study the brain and/or activate or calm sections of the brain (used to treat depression).

The worst side effect can be a seizure but that's only if it's calibrated incorrectly - usually the worst side effect is a headache.

Some people are combining imaging with this to measure how it changes brain function.

Fun fact: if you hold the device over the motor cortex it will cause the patient's hand to twitch.

the significance of ventricle size

day5

Gray Matter

day5

What have post-mortem studies found out?

day5

Brain function paradigms

day5

Symptom Capture

day5

Symptom Provocation

day5

Cognitive Activation

day5

Issues with NeuroImaging methods

day5

-group matching: need to match patients and controls (brain volume, weight, education, handedness, drugs)

-medication status

-comorbidity

-good experimental and control tasks

-lots of confound with post-mortem studies

NEUROPSYCHOLOGY (what is it?)

day5

Visuospatial memory tasks

day5

-rey-osterreith complex figure task (look at order they do it, what they miss)

-draw a clock

Executive functioning tasks

day5

-frontal lobe tasks involving planning, problem solving, and logic all at once

-EXAMPLE: wisconsin card sort

-EXAMPLE: tower of london (also called hanoi) --> schizophrenics make the same mistake over and over again

Reliability

day2/day4

-stability of a psychiatric diagnostic system

-internal consistency and test-retest and interrater

how to maximize kappa

day2

-use DSM

-use SCID

-base findings off of a joint interview

-use well-trained clinicians

Discriminant Validity

day2

concurrent validity

day2

correlation of a measure with a criterion at the same point in time

(ex: test scores and current school performance)

(part of criterion validity)

convergent validity

day2

when measures of constructs that are expected to correlate actually do correlate

(state anxiety and trait anxiety)

Predictive Validity

day2

Construct Validity

day2

criterion validity

day2

how well the measure measures the construct as compared to other measures established for that construct

-includes concurrent and predictive validity

How to read journal articles

day2

-correct topic?

-read methods very carefully

-how were subjects sampled?

-how were subjects diagnosed?

-what measures were used, and are they reliable and valid?

-how many subjects?

-was there a control group?

-do interpretations match the data without forcing it or going beyond?

-what are the limitations?

-so what?

confound

day2

Psychiatric Epidemiology

day2

Prevalence

day2

lifetime prevalence

day2

Cross Sectional Design (how it's done)

day2

-data collected at one point in time

-geographically or otherwise defined populations

-random sampling, voter registration, census lists

Cross Sectional Design (purposes)

day2

-examine prevalences of disorders, subtypes, and risk

-examine relationships between disorders in a population

-examine relationship between risk factor and disorder

-determine change in incidence of a disorder over time

Cross Sectional Design (limitations)

day2

-subjects only tested at one time

-correlational and not causational

-retrospective reporting is unreliable

-incidence can change

Longitudinal Design

day2

-subjects tested at more than one time

-allows real time to elapse

Longitudinal Designs (limitations)

day2

Case-Control Experimental Design

day2

-individuals selected for having a specific disorder of interest

-not taken from various geographical locations

-usually at inpatient hospitals, clinics, or small communities

-case-by-case matching of experimental and control groups

Limitations of Case-Control Experimental Designs

day2

-imperfect matching btwn groups

-ascertainment is an issue

-relatively small n's

Cronbach's alpha

day2

Types of Research Questions

day1

treatment efficacy

description

diagnostic validation and subtyping

risk factor identification

prevention

etiology

Treatment Efficacy

day1

Descriptive Research

day1

Diagnostic Validation and Subtyping

day1

Risk factor identification

day1

Prevention Research

day1

Etiology

day1

diagnosis

day1

diagnostic sign

day1

diagnostic symptom

day1

Reasons for classifying disorders

day2

description and communication treatment planning

prediction

standardization of research

unintended consequences of diagnosis

day2

central tendency

day2

variability

day3

T-TEST equation

day3

examines mean difference over variability you want a large mean difference and little variability can measure no more than two groups at two times (see further slides for specific t-tests)

df = (n1+ n2) - 2

independent t-test

day3

paired samples t-test

day3

same group, repeated measures

also, if it's a married couple, a pair of twins - something tested together

one-sample t-test

day3

correlations

day3

assessing the strength of the relationship between two continuous variables

EXAMPLE: age and education

EXAMPLE: tests and school performance

EXAMPLE: age and brain volume pearson's correlation coefficient = r (-1 - 1)

when you say correlation you are referring to (r), say relationship otherwise

chi-square

day3

Psychiatric Genetics

day4

family studies

day4

-studying how often a disorder pops up in a family (how prevalent is it in the family?)

-confound: families share environments use of a control family

-proband = disordered individual whose family members are studied

adoption studies

day4

-we look at proband adoptee's biological and adoptive family

-look at rates of disorder in relatives

-these are better than family studies because they share different environments

-even better than family or adoption studies, since they share 100% or 50% of their genes

-if it's completely genetic the concordance between monozygotic twins will be 100%, dizygotic twins will have 50% concordance

proband

day4

concordance rates

day4

linkage studies

day4

-studies if genetic markers linked with genes for a psychiatric disorder

-assumes one major gene, which is bad, since most studies show that an interaction of multiple genes (polygenetic) more likely creates a diathesis

Assessment

day4

-measurement of a psychological concept (construct)

-used as a dependent variable in experiments

construct

day4

measure/instrument/questionnaire

day4

item

day4

internal consistency

day4

-part of RELIABILITY, correlating items within the same questionnaire

-the degree of consistency or homogeneity of items

-measured with cronbach's alpha

-item intercorrelations

-item-total-score correlations (are there any items that don't correlate with the final score?)

-split-half reliability --> does one half of the measure correlate with the other half -factor analysis

test-retest reliability

(day2/day4)

 -extent to which the same person tested twice on the same measure will have similar/same results

-also phrased as stability of scores over time 

interrater reliability

day4

-extent to which two trained independent raters will agree on diagnosis of a patient (this is specific to clinical)

-to maximize this, we use DSM and SCID

-we use percent agreement and kappa to measure amount of agreement. kappa takes into account the likelihood they'd agree on change, whereas % agreement does not.

step 1 of constructing a questionnaire

day4

step 2 of constructing a questionnaire

day4

decide on the purpose of the questionnaire

-response of the treatment?

-letting them into a study?

step 3 of constructing a questionnaire

day4

step 4 of constructing a questionnaire

day4

step 5 of constructing a questionnaire

day4

create item scale

-decide on if you're being overinclusive/including subscales

step 6 of constructing a questionnaire

day4

step 7 of constructing a questionnaire

day4

step 8 of constructing a questionnaire

day4

step 9

day4

step 10 of creating a questionnaire

day4

calculate reliability

-internal consistency, item-total correlations, split-half correlation

-figure out what items you want to be correlated with each other, and don't forget to flip the reversed measures

step 11

day4

step 12

day4

all steps (sans explanations) of making a questionnaire

day4

Determine construct to be measured (dock)

Decide on purpose (dope)

Review the literature (reliably,)

Decide on a format (doofus,)

Create item scale (creates)

Write items clearly (whisker)

Circulate item pool (caps.)

Pilot test (pot)

Remove/revise items (removes)

Calculate reliability (cheap)

Revise and readminister Test Validity (reapers)

When do you use an ANOVA?

day6

F-test

(day6)

the test performed within an ANOVA

puts variability among group means

over variability of scores within groups

large F is good, means smaller p-value

One-Way ANOVAs

day6

Repeated Measures ANOVA

(day 6)

Mixed Model ANOVA

(day6)

-aspects of both one-way and repeated measures

-when you have 2 or more groups over 2 or more measures

-this is used most IRL

Effects of Mixed Model ANOVA

(day6)

1) main effect of group

2) main effect of measure/condition

3) interaction effect of group and measure

Main effect of group

(day 6)

Main effect of condition/measure

(day 6)

Interaction Effect

(day 6)

does the pattern of the means differ between groups and conditions over time?

another way of phrasing:

is the difference between conditions different between groups?

Difference Score

(day 6)

Cognition

(day7)

-attention

-memory

-tests of cognition are helpful in understanding psychopathology

cognitive measures/

measures of cognition

(day 7)

-response times (RTs)

-accuracy or error rates (ER)

-these are objective measures

cognitive defecits (examples)

(day 7)

cognitive biases

(day 7)

Stoop Task

(day 7)

-standard: colors and color names match/don't match

-modified/emotional: colored words are positive, negative, and neutral relating to their condition - they'll take longer on negative stimuli

interference (in cognitive task)

(day 7)

the difference between the RT/ERs of neutral and negative stimuli

dot probe task

(day 7)

1) fixation cross

2) show two pictures in opposite corners, one neutral, one emotional

3) take pictures away

4) show letter/dot/etc where one of the pictures was

--> people will respond slowest when dot is in opposite corner from emotional picture - your attention is captured

-this is an attention task

Dichotic Listening task

(day 7)

1)subjects are played different things in opposite ears, are told to listen for target words (which will be played in either ear)

2) they are more likely to take note of words relating to their condition

-this is an attention task

attention biases

(day 7)

-most often found in people with anxiety disorders or high trait anxiety

-bias is towards disorder-related stimuli

(high trait anxiety = sort of sad people, concerned eaters, not MDD or AN/BN)

Free recall test

(day 7)

-memory task

1) present word lists of mixed up +/-/0 words

2) have them do unrelated buffer task

3) ask them to remember as many words as possible

4) compare the categories of +/-/0, which had the most?

Cued recall

(day 7)

Autobiographical memory task

(day 7)

-give subjects positive, negative, and neutral cues, ask them to provide an anecdote, and to be as specific as possible

-EX: patients with depression have overgeneral memories, which causes them to think their problems are global, internal, and stable

memory biases

(day 7)

-more often found in mood disorders

-biases towards disorder-related stimuli

peripheral psychophysiology

(day 7)

skin conductance

(day 7)

-skin's ability to conduct electricity

-the sweatier you are, the higher conductance

-used to measure arousal in psychophysiological testing

muscle tension

(psychophysiological testing)

(day 7)

eye movements: startle response

(in psychophysiological testing)

(day 7)

-measured in orbicularis oculi

-emotional modulation of startle: pos=least amount of startle; neutral=mild amount; negative=most amount

--> this is diminished in antisocial personality disorder

eye movements: smooth pursuit

(in psychophysiological testing)

(day 7)

watch something that looks like slow tennis

-schizophrenic patients, and their relatives, to a degree, have issues with this (52-85% of schiz)

-lithium can also interfere

Methods considerations and problems of psychophysiological testing

(day 7)

-some ethnicities have lower skin conductance

-group matching

-medication

-comorbidity

questions asked in treatment research

(day 7)

-Does a treatment work?

-How does a treatment work?

-Does a treatment work better in certain groups of people?

-Are the effects of a treatment statistically significant?

-Do the effects of a treatment last?

Inclusion Criteria

(group design)

day 7

Exclusion Criteria

(group design)

day 7

what is the effect of increasing the homogeneity of patient groups in group designs?

(day 7)

what's the deal with random assignment in group designs?

(day7)

manualized treatment

(day 7)

-There are instructions, down to the minute, for what you do with a patient

Advantages:

-standardizes across clinicians,

-more conducive to treatment,

-can disseminate techniques to other clinics,

-serves as manipulation check (if not adhered to, can be a source of error)

what problems does not having a comparison condition in a group design create?

(day 7)

what are the benefits of having a waitlist/no treatment group?

(day 7)

what's the difficulty with having a comparison group in a group design be "treatment as usual"?

(day 7)

-Both treatments can be compared to no treatment

-may be hard to distinguish sigdiff between treatments a and b because they could both work

Dismantled treatment (comparison condition in group design)

(day 7)

-using treatment A with one peice missing

-can show what parts of a treatment are unnecessary/which are necessary

additive treatment (as comparison condition in group design)

(day 7)

parametric variation (as comparison condition in group design)

(day 7)

crossover condition (as comparison condition in group design)

(day 7)

outcome measure: symptom measure

(group design)

(day 7)

continuous: are severity scores different?

categorical: do they still meet criteria for the disorder?

outcome measures: assessment of function

(group design)

(day 7)

when are outcome measures measured?

(group design)

(day 7)

analysis of treatment research (group design)

(day 7)

1) dropouts: either delete their data entirely or carry over data from their last measure until the end

2) average the continuous measures pre and post treatment for each group

3) look for significant interaction effects

4) figure out of non-responders and responders differ on some pre-treatment variable like education, age, sex, symptom severity, or comorbidity --> this can also serve as a prediction method for treatment response

Completer's Analysis

(day 7)

-excludes dropout data

\_-> makes it difficult if there's a lot of dropouts

\_-> skews results (can make treatment look more effective if those who dropped out did so because they weren't experiencing results)

\_-> defeats the purpose of random sampling

"intent to treat" analysis

(day 7)

-dropout data is carried over from their last measure

-this is very conservative, making it harder to find a significant effect (so if you find significance, it's awesome)

Single Case Design

(day 8)

-done when there's a rare disorder or rare presentation of a disorder

-you end up with a unique patient group and treatment, since you don't have the luxury of drawing on a large group for subjects

-it's a within-subjects design

how many observations are in a single case design?

(day 8)

describe what the measures of a single-case design need to be (even before treatment)

(day 8)

-need to be stable dependent variables at the beginning, that way changes from A to B aren't from erratic readings

-ex: skin conductance (PTSD), #purging episodes (BN)

ABAB Design

(single case design)

(day 8)

baseline --> treatment 1 --> baseline --> treatment 1

-reversals can imply causality due to relationship between improvement and treatment (though clinically we don't like reversals)

multiple baseline design

(single case design)

(day 8)

changing criterion design

(single case)

(day 8)

1) set criterion or goal for patient for a period of time

2) once goal is met, reward

3) make goal more stringent

4) gradual changes over time --> shaping behaviors

Analysis of Single-Case Design

(day 8)

-visually based

-change in mean=avg rate of behavior with each phase (ABAB)

-change in level=change from last measurement of a phase to first measurement of next phase

-change in slope=the direction of the slope changes from phase to phase (within phase slope, compare them to each other)

-latency of change = how long does it take for a phase to have effect

Limitations of single-case design

(day 8)

-difficult to generalize to other patients

-n=1, so its difficult to determine treatment moderators

-no p-values or stats = no way of making concrete decisions

-only large effects are obvious

-pattern has to be fairly clear or another round of AB should happen

Ethical Issues with Deception

(day 8)

-does the risk outweigh the benefits?

-it must be vital to the experiment for IRB to approve it

What is a debriefing?

(day 8)

How do we establish confidentiality in an experiment?

(day 8)

-names and other identifiers are removed from the data

-participants' IDs are never released without explicit permission

What is informed consent?

(day 8)

-process that informs the participant about (1)procedures, (2)risks and benefits, (3)confidentiality, (4)their voluntary participation and freedom to leave at any time, (5)gives them an alternative to participation

-they usually sign a consent form, though sometimes consent can be verbal

-very important that no coersion occurs (large sums of money = coersion)

who can give consent?

(day 8)

all those who are not

(1)minors (need parent's consent and their assent), and

(2)those without disorders that incapacitate them from (psychosis/retardation)

Intervention Ethics Issues: all

(day 8)

-all treatments must be described, along with the risks and benefits to being on placebo vs experimental

-explain random assignment to groups

-explain that they can end up experimental or placebo group

-witholding treatment in control group can make them worse

-informed consent can increase attrition (people leaving?)

ethical issue of witholding treatment in control groups

(day 8)

-treatment is delayed

-condition may worsen

-safety monitoring boards deal with patients who worsen, decide if they should be given treatment/if trial should be stopped

IRBs

(day 8)

-institutional review boards

-must submit protocols to IRB (procedures, consent forms, advertisements, questionnaires)

-must have their approval to research

-grants cannot be given without their approval

-their goal is to protect patients

if you were an unethical psychologist/psychiatrist/ph. d, you might indulge in these ethical issues

(day 8)

-fraud (fudging data)

-plagiarism

-not giving authorship where its due

-sharing materials badly

-not keeping confidentiality

-conflicts of interest due to finances

what sort of pressure are professionals under?

(day 8)

PUBLISH OR PERISH

teaching is half the job

PUBLISHING STEPS

(day 8)

1) submit research to journal

2) it gets peer-reviewed

3) peers send critiques

4) get a response (accept, reject, accept with revisions/revise and resubmit)

what makes a journal a higher tier journal?

(day 8)

how critical is it to get published in a good journal?

(day 8)

what future directions are there for psychological research?

(day 8)

-revising DSM

-genetics

-placebo response

-endophenotypes

-neuroimaging and treatment prediction

-psychotropic medication in children

-exposure therapy in virtual reality

future direction of genetics research

(day 8)

group by gene/allele combo, use neuroimaging as measure

ex: s vs l allele in serotonin transporter system - amygdala activation is highest with ss combo

what are the current findings/future research for the biological basis of the placebo response?

(day8)

future directions: endophenotypes

(day8)

-endophenotype measures (imaging abnormalities) come between genetics and behavior

-it's a biological measure

future directions: neuroimaging and predicting treatment response

(day 8)

future directions: psychotropic drugs in children

(day 8)

-FDA put black box warning on SSRIs in kids, thinking they caused them to commit suicide more often as adolescents

-murky as to what it actually does, though there is research pointing either way, but suicide has increased since warning in US and netherlands