Midterm Exam Flashcards
Stanley Milgram Paper (1963)
Obedience & How it relates to the ability to Harm
Author Name: Stanley & Milgram
Background: atrocities during WW2, what motivates humans to harm others, relationship between authority and capacity to harm.
Hypothesis: Most people wouldn’t go to the danger zone, only a few might. We fundamentally underestimate our capacity to harm when following orders.
Main Structure: fake experiment (memory test), 3 people (experimenter, subject/teacher, victim learner)
Methods: subject must administer shocks in memory test (danger done). At 300 volts the learner stops answering and pounds for help, silence is considered wrong, and if the subject hesitates they are given encouragement.
IV: N/A
DV: the highest shock level the teacher would go to without stopping.
Results: We abandon out values to obey authority, and we will try to justify these actions.
Critiques:
- causes emotional distress
- more diversity / larger sample size
-clear instructions and guidelines to minimize confusion and risk
Hard (or Natural Sciences) vs. Social Sciences is there a difference?
Hard science: Physics, chemistry, astronomy, geology
explanation , understanding, and predictability through observation and experimentation
Social sciences: Psychology, sociology, anthropology, economics, and political science
Why do we study psychology? And why do we study research methods in psychology?
- To understand human behavior
- To understand ourselves as individuals
- To understand others
- Interventions: counciling, family therapy, risky behaviors, education (in school), training (on the job), political psychology
- A solid foundation for other psychology courses
-Other psychology courses are about content…
-…but research methods is about process
-We see how sound, scientific research is conducted
-We learn how to do our own research to answer our own questions
-We are better able to read and understand the research of others
-We learn how to be skeptical of other research
-We learn how to adapt and apply existing research findings to our own specific situation - We learn how to communicate about research in speaking and in writing
-Writing APA style
-speaking-Professional Presentations
-Posters-Scientific Presentations
-The Charter of W&M
-we read good science and write good science and discuss good science we will learn to do good science
-“If you read junk, watch junk, and eat junk, you will be junk” - Rev Jessie Jackson
-Professional Development - graduate school
Ways of Knowing
The liberal arts
Grammar, logic, and rhetoric (the trivium)
Arithmetic, geometry, the theory of music, and astronomy (the quadrivium)
Epistemology: the theory of knowledge, especially the methods, validity, and scope of knowing. An understanding of what differentiates justified belief from opinion
Ways of knowing authority: authority
Basing our beliefs on what we are told by others
Examples? (parents, teachers, textbooks)
Authority brings stability and consistency and can be very beneficial, especially if knowledge gained is brand new
Problem: authorities can be wrong!
Ways of Knowing: Reason / Logical Argument
- Use of reason via conversation (discourse) to come to a consensus
- Uses the a priori method
a. Based on argument and logic, not direct experience
b. Denoting conclusions derived from premises or principles - Problems:
a. Our initial assumptions may be incorrect
b. By using reason/logic alone, we have no way to check the accuracy of our assumptions
c. Valid logical arguments can lead to opposite conclusions
Ways of Knowing: Empiricism / Direct Experience
- Process of learning via direct observation or experience
- problems:
a. Experiences are limited to our interpretations of them
b. Experiences can be influenced by social cognition bias - Confirmation bias
- Belief perseverance
- Availability heuristic
Attributes of Scientific Thinking in Psychology
- Determinism: is our behavior pre-determined?
- Are people like moving objects and can the behavior of humans be predicted by psychology in the same way that the movement of objects can be predicted by physics?
- Statistical or probabilistic determinism- events can be predicted but only as probabilities
- Objectivity - eliminating any bias from our own experimentation
- Other researchers should be able to verify our results through replication
- In order to make replication possible our publications need to be clear - especially the method section
- Scientific research in psychology must be based on what is observable
- Makes systematic observations
a. Less affected by bias than everyday observations - Produces public knowledge
a. Objectivity criterion - Agreement by two or more observers
b. Example → from introspection to behaviorism - Behaviorism (Watson)
- Sound scientific research in psychology is data-driven
- Here is my datum (singular)
- is my data (plural)
- Stadium, aquarium, colloquium, etcium
- I followed the Red Sox to several American League stadia
Scientific Conclusions (and therefore social conventions) are Subject to Revision
- The earth is flat
- The sun travels around the earth
- Airplanes can fly no faster than the speed sound
- Only men should have the right to vote
- Some races are inferior and should be made slaves to others (Eugenics and the Holocaust)
- Some races should be exterminated for the overall benefit of society
- War is an inevitable state in the relation between nations
Science as a Way of Knowing
- Produces tentative conclusions
a. findings subject to outcomes of future research - Asks answerable questions
a. Empirical questions (i.e answerable with data based on the use of valid scientific methods) - Develops theories that can be falsified
a. Falsification criterion
*scientists (i.e psychologists) are skeptical optimists!
We Study Empirical Questions
- While we may be attempting to answer questions about how people think, we conduct our studies by examining empirical questions
- We structure our research so our questions are empirical and can be answered through observation
- Not all questions (especially broad and important ones) can be answered empirically - but we try!
Social Problems that Need to be Addressed through Psychological Research
- Inequality and injustice in society
- Inequality and injustice in the world
- Migration and the plight of refugees
- Pollution, global warming, and climate change
- The high incarceration rate in the US
- Gun violence
- Eating disorders
- Impoverishment in all the forms that it takes
Psychology vs. Pseudoscience
- Read our Goodwin book…or see Ghost Busters
“- False science” - literally - Textbook examples → phrenology and graphology
- Compared to true science, pseudoscience
a. Associates itself with real science - Tries to appear legitimate
b. Relies heavily on anecdotal evidence - Ignores counter instances
- Results from effort justification
c. Sidesteps falsification - Avoids falsification by explaining away anomalies
d. Reduces complex phenomena to overly simplistic concepts
The Goals of Research in Psychology
- describe
a. Identify regularly occurring sequences of psychological events (e.g behaviors, thoughts, emotions, ect.) - predict
a. Psychological events follow certain “laws” that are regular and therefore predictable - Explain
a. Psychological events are explained in terms of their relationship to other factors
b. Causal explanations are ideal - apply
a. Science informs real-world applications of psychological events - Controlling behavior (?_…
a. B.F. Skinner Beyond Freedom and Dignity
Controlling Behavior?
- How about instead, influencing behavior
a. Learning - Education in schools and training on the job;
b. Early intervention
c. Rehabilitation, socialization, and training of convicted criminals (problems with recidivism):
d. Self-acceptance
e. Marketing
f. Avoiding risky or dangerous behaviors
g. Influencing people through persuasion instead of forcing them through threats
Scientific Thinking in Psychology
- Psychology is a science and adheres to the assumptions and goals of science
- Science distinguishes itself from pseudoscience by being systematic, empirical, data-driven, tentative, and falsifiable
- As psychological scientists, we strive to describe, predict, explain, and apply what we discover from our research
Overmier & Seligman Article (1967)
Dog Electric shock through the floor; learned helplessness
Author Name: Overmier & Seligman
Background: researchers wanted to see if learned helplessness would happen after being exposed to previous shock in different time periods
Hypothesis: After exposure, when shocked they would give up
Main Structure: 3 experiments
-initial shock exposure
- seeing if twitching helped with shock exposure (similar results)
- varying time in between shocks
Methods
- strapped dogs in so they couldn’t move
- shocked them from floor
- paralyzed some to stop twitching
IV: the pre trial exposure to shocks with the intensity and duration of and between shocks
DV: how the dogs reacted to the shock from the floor; if they laid down and accepted to shocks or if they escaped
Results: dogs experience learned “helplessness” where they believed there was nothing they could do to get out based on their past experiences.
Critiques:
- different settings between treatment and performance task
- differing apparatuses for shock treatment
no internal control group for experiment 2
- minimal shock intensity difference between “high intensity” condition and treatment condition
Developing the APA Code of Ethics
- Historical cases of ethically questionable research
a. Watson & Rayner (1920)- scaring little albert
b. McGraw (1941)- effects of repeated pinpricks
c. Dennis (1941)- raising children in isolation - First code → 1953
a. Hobbs committee
b. Critical incidents techniques - APA Ethical Principles of Psychologists and Code of Conduct
a. 2002 (2010 amendments)
APA Ethical Principles of Psychologists and Code of Conduct
- Guidelines for ethical behavior for the practice of research, clinical work, and teaching in psychology
- Applies to all of us in the field of psychology
- Code contains:
a. 5 general principles
b. 10 standards of practice
http://www.apa.org/ethics/code/index.aspx
5 General Principles APA
- Beneficence and Nonmaleficence
a. Constantly weigh costs & benefits; protect from harm; produce for greatest good - Fidelity and Responsibility
a. Be professional; constantly be aware of responsibility to society - Integrity
a. Be scrupulously honest - justice
a. Always treat people fairy - Respect for People’s Rights and Dignity
a. Safeguard individual rights; protect rights of privacy and confidentiality
10 Sections on Ethical Guidelines
Section 1: Resolving Ethical Issues
Section 2: Competence
Section 3: Human Relations
Section 4: Privacy and Confidentiality
Section 5: Advertising and Other Public Statements
Section 6: Record Keeping and Fees
Section 7: Education and Training
Section 8: Research and Publication
Section 9: Assessment
Section 10: Therapy
Ethical Guidelines for Research with Humans
Standard 8: Research and Publication
a. Several particular points, all of which fall under the General Principles. Some highlights:
- Identify potential risks
- Protect participants from physical and psychological harm
- Justify remaining risks
- Obtain informed consent
- Take care of participants after the study (debriefing)
Weighing benefits and costs: the IRB
- IRB = Institutional Review Board
- Determines whether the project meets ethical guidelines
a. Some research is exempt from review
b. Some research gets an expedited review
c. Some research requires a full review - Key factor: degree of risk to subjects
a. No risk (could be exempt)
b. Minimal risk (expedited)
c. At risk (full) - Issues: judging methodological adequacy, no appeal, anti-basic research, overly cautious
- Informed consent
a. Sufficient information provided to research participants to decide whether to participate - Historic examples of poor consent
a. Tuskegee syphilis study
b. Willowbrook hepatitis study
c. MK-ULTRA (CIA & LSD) - Deception in Research
a. Desire to have subjects act naturally
b. Milgram obedience study as an example - Cover story → effect of punishment on learning
- Real purpose → limits of obedience to authority
-No consent needed in some circumstances
a. Some survey, educational, archival, and observational research
Elements APA
- Study’s basic description
a. Enough information to decide whether to participate - How long participation will take
- May quit at any time
- Confidentiality and anonymity ensured
- Contact information given (researcher, IRB)
- Opportunity to obtain final results of the study
- Signatures
Conent with Special Populations
children
- Parental assent also needed
Children and other special groups (e.g prisoners)
- Special care to avoid feelings of coercion
Treating Participants Well
Debriefing
- Dehoaxing
- Desensitizing
- Confidentiality
- Participant crosstalk
a. Code allows partial debriefing followed by full report at completion of the study
Research ethics and the internet
- Problems with ensuring consent
- Problems with conducting effective debriefing
Ethical Guidelines for Research with Animals
- Animal rights
a. Not a new issue - Using animals in psychology research
a. Miller- aids both humans and animals - The APA Code for animal research
a. Justifying the study
–>Cost-benefit analysis
b. Caring for the animals
–>E.g expertise with species
c. Using animals for educational purposes
–>Minimize
Scientific Fraud
- Plagiarism
- “The presentation, with intent to deceive, or with disregard for proper scholarly procedures of a significant slope, of any information, ideas or phrasing of another as if they were one’s own without giving
appropriate credit to the original source (William and Mary Honor Code) - Safe Assign
- Falsifying Data
a. Cases: Diedrick Stapel, b. Stephen Breuning
Varying degrees (all unethical)
c. reasons
–>Range from individual weakness to societal moral standards
–> Publish or perish climate in academia
Festinger & Carlsmith (1959)
Cognitive Dissonance
Author Name: Festinger & Carlsmith
Background: To examine and measure levels of cognitive dissonance in relation to forced compliance
Hypothesis: the higher the reward the higher level of cognitive dissonance. Thought 1$ would be most positive
Main Structure
- The participants were introduced to the experimenter
-Hour long experiment
- Interview afterwards
Methods:
- groups completed 2 tasks (s0me told it was fun before hand)
- false debriefing: told that experiment information about it being fun affected performance as goal
- then proceed to interview where either given no money OR given money AND was tried to convince it was fun (paid to lie)
- Post -experiment questions
IV: quantity of reward
DV: level of cognitive dissonance
Results:
Unpaid Control
- (-0.45)
Paid $1
- (1.35)
- Rationalized due to no other justification
Paid $20
- (-0.5)
- Rationalized due to money
Critiques:
- very little diversity
- final interview was assumed to be truthful
- sense of self-importance may have skewed viewers
- more varying rewards
Varieties of psychological research
- The goals: basic versus applied research
- Basic
a. Designed to understand fundamental psychological phenomena
b. Example → stimulus factors affecting selective attention - Applied
a. Designed to shed light on the solution to real-world problems
b. Example → effect of cell phone use on driving
c. Example → how to motivate people to wear masks and get vaccinated for covid. - Basic (Theory)
a. Designed to understand fundamental psychological phenomena
b. Describing
c. Predicting
c. Explaining - applied
a. solving real world problems
b. Education
c. Poverty
d. Risky behavior
e. Job performance
The laboratory vs the Field
Lab
a. Focus on the independent variable
b. Measure the dependent variab;e
c. Controlling extraneous variables
d. More scientific
e. But is artificial
Field
a. More realistic
b. Real-world validity
c. Issue of ethics and privacy
d. Participants may self-select
Quantitative vs. Qualitative research
- Quantitative
a. Data collection
b. More “scientific” - qualitative
a. Anecdotal
b. Focus groups
c. Opinion
d. More “rich” human data - Operational Definitions
a. How will you know if you see it?
b. How can you operationalize behavior?
c. A strict and valid operational definition will add clarity and definition to research
d. It should be logical and understandable
e. It should be observable and measurable
Asking Empirical Questions
- Empirical questions
a. Answerable with data
b. Terms precisely defined - Operational definitions
a. Variables defined in terms of clearly specified set of operations
–> Hunger = 12 hours without food
–> Frustration = consequence of being blocked from a goal - Converging operations
a. Understanding increases as a studies with different operational definitions “converge” on the same result
Developing Research from Serendipity
- Serendipity: discovering something while looking for something else entirely, has been a source of numerous important events in the history of science
- real - world events
- Seize the moment, people have short memories
- How people react to recent events
- Kitty Genovese
- September 11, 2001
- Popular movies
- COVD19
- BLM
- January 6th assault on the Capital
parsimony principle
is basic to all science and tells us to choose the simplest scientific explanation that fits the evidence.
Developing Research from theory
- The nature of theory
a. Summarizes, organizes, explains, provides basis for predictions
–>Includes constructs → hypothetical factors involved in the attempt at explanation
–>E.g cognitive dissonance - The relationship between theory and research
a. hypotheses deducted from theory
b. Outcomes / data provide or fail to provide inductive support for theory
–> Theories are never “true” nor “false”
Developing Research from other Research
- Replication
a. Direct replication
–>A reproduction of the exact study procedures as the original study
b. Conceptual replication
–>A partial replication, with new features added to extend the original study’s findings - Ethics Box
a. Questionable research practices and replication remedies
Reviewing the Literature
Computerized database searches
- In psychology → PsychINFO
a. Most recent info → www.apa.org/psychinfo
- Search results
a. Advanced search option (use of multiple search terms)
b. Using truncated search terms to avoid being too narrow
c. Being strategy → trial and error, expand and contract
- Computerized database searches
a. Search results
–> Results lists begins with most recent research
–> Take note of source (e.g journal article, book, dissertation)
You may limit your search by date or source too
–> Read abstracts provided when you click on the title
Getting the Most out of reading journal articles
- Get as much as possible from the abstract
- Look for the general statement of the problem in the opening paragraph
- Look in the introduction for existing theories
- Near the end of the introduction look for the hypotheses
- In the method section look for who is being tested
- In the method section look for the procedure
- Understand the data in the results section
- Look in the discussion section for an explanation of the results and how the results answer the original question
Developing research from theory
10 steps of research
Attributes of good theories
- They should advance knowledge
- They should be subject to falsification
- They should be parsimonious
a. Minimum number of constructs
b. Minimum number of assumptions - They should solve real-world problems
- “There is nothing so useful as a good theory” Kurt Lewin (?)
Developing Research from other research
- Building programs of research
- Using findings of one study as beginning for another
- Explaining contradictions in publications
- Asking “what next”
- Applying to other settings
- Replication and extension
Reviewing the literature
- Using other people’s results as a beginning for our research
- Other people’s data bases
Dutton & Aron (1974)
Some Evidence of Heightened Attraction under high anxiety conditions
Author Name: Dutton & Aron
Background: Previous research suggests that sexual attraction occur more frequently when in states of strong emotion
Hypothesis: The aim of this study was to test the idea that “an attractive female is seen as more attractive by males who encounter her while they experience a strong emotion (fear) than by males not experiencing a strong emotion.”
Main Structure
- Experiment 1 tried to verify this link between emotion and sexual attraction in a nature
- Experiments 2 and 3 were conducted in order to clarify and test the validity of the results of the first experiment
Methods:
Experiment #1
- Men crossing a bridge were approached by an interviewer
- asked to write a story about a picture of a sad woman
- stories scored for sexual content
- interviewer offered number (difference in caller indicates attraction)
- asked to rate control bridge vs scary
Experiment #2
- repeated to make sure population sample was good
- verified result of #1
Experiment #3
- laboratory setting where they expected painful or no painful shock & woman in other room
IV: amount of stress/ anxiety
DV: indicated attractiveness
Results
- male subjects on scary bridge more attracted than control
- Higher anxiety reported when the subject anticipated a powerful shock
- Subjects who expected a high shock with a female confederate present showed less anxiety than subjects in the control (two male subjects run at the same time)
- ANOVA showed significant main effect for subjects expecting strong shock to attraction ratings
-TAT sexual imagery scores were found to be higher when both the subject and the female confederate were expecting a strong shock
- support hypothesis
Critiques
- definition of arousal was ambiguous
- type of anxiety (real & expecting)
- subjects not randomly selected
- consistency in interviewing should have been established
- diversity of participants (assumed to be heterosexual cisgender males)
- beauty is subjective
An introduction to Logic
Inductive reasoning
a. Going from specific to the general (based on research)
b. An inductive generalization
Deductive reasoning
a. Going from the general to the specific
b. A deductive argument (a proof)- proven
*look for premise and Logic
Essential Features of Experimental Research
Mill’s inductive logic
- Method of agreement
a. If X, then Y (sufficient → x is sufficient for Y)
- Method of difference
a. If not X, then not Y (necessity → X is necessary for Y)
- Together → X is necessary & sufficient for producing Y
a. Agreement :Analogous to experimental groups
b. Difference: Analogous to control group
A logical Argument (AKA a Syllogism)
- Premise → all birds lay eggs
- Premise → an eagle is a type of bird
- Conclusion → eagles lay eggs
If the premises are true and the logic is sound, then the conclusion must follow.
Is this an inductive generalization or a deductive argument?
What is/are the premises and what is/are the conclusions?
Because triangle A is congruent with Triangle B, and Triangle A is isosceles, it follows that triangle B is isosceles
deductive
Is this an inductive generalization or a deductive argument?
What is/are the premises and what is/are the conclusions?
The coffee, tea, and quiche at Aromas are all excellent. The likely conclusion is that entities on the menu are excellent.
inductive
Is this an inductive generalization or a deductive argument?
What is/are the premises and what is/are the conclusions?
Sherlock Holmes observed that since there were several watchdogs sleeping in the stable that night, and yet, though someone had been in and stolen Blaze, the horse, the dogs had not barked enough to arouse the two lads on the loft. Obviously, concluded Holmes, the thief was someone the dogs knew well.
deductive
Conditional Statements
If (antecedent) then (consequent) OR (consequent) and (antecedent)
A conditional statement is not in itself an argument, but it may serve as either the promise or conclusion (or both) of an argument
Modus Ponens (the way that affirms by affirming)
Affirming the antecedent (valid)
If P, then Q. P → therefore Q
ex.
If my hamster have birth then it must be female
My pet hamster has given birth
Therefore, It must be female
Invalid form of Modus Ponens
- the logical fallacy of affirming the consequent
- If P, then Q. Q → therefore P
- ex.
a. Students who participate in group study sessions do well on exams
b. Sam did well on the exam
c. Sam participated in a group study session
Modus Tollens (the way that denies by denying)
- denying the consequent: valid
- If P, then Q. Q → therefore P
- ex.
a. If i owned property in Manhattan today I would be rich
b. I am not rich
c. Therefore, I don’t own property in Manhattan
Invalid forms of Modus Tollens
- The logical fallacy of denying the the antecedent
- If P, then Q. P → Therefore Q
- ex.
a. If I owned property in manhattan today, I would be rich
b. I don;t own property in Manhattan today
c. Therefore I am not rich
Is this a valid argument?
Premise: if you water the grass, it will be wet
Premise you: watered the grass
Conclusion: it must be wet
Yes (affirming the antecedent)
Is this a valid argument?
Premise: if you water the grass, it will be wet
Premise: You didn’t water the grass
Conclusion: It must not be wet
No (denying the antecedent)
Is this a valid argument?
Premise: if you water the grass, it will be wet
Premise: the grass is wet
Conclusion: you must have watered it
No (affirming the consequent)
Is this a valid argument?
Premise: if you water the grass, it will be wet
Premise: the grass is not wet
Conclusion: you must not have watered it
No (denying the antecedent)
Is this a valid argument?
Premise: if you build it, they will come
Premise: you build it
Conclusion: they are here
No (affirming the consequent)
Is this a valid argument?
Premise: if you build it, they will come
Premise: you didn’t build it
Conclusion: they are not here
No (denying the antecedent)
Is this a valid argument?
Premise: if you build it they will come
Premise: they are here
Conclusion: you must have built it
No (affirming the consequent)
Is this a valid argument?
Premise: if you build it they will come
Premise: they are not here
Conclusion: you must not have built it
No (denying the antecedent)
Operant Conditioning
- Motivation: food deprivation to 90% (or 85%) of free-feeding weight
a. Monitor rats weight (determine what weight the rat should be to be at 90% of free feeding)
b. Rats need 2-3 food pellets to maintain weight
–> Remove food & leave rat with 1 pellet
–> Leave rat with 1 to 2 pellets depending on weight gain
c. There are other tiny pellets we use to train them (noise pellets)
d. Responsible for monitoring rats weight, attend to rat, and fill out log (check on rat once a day 8am-6pm) - Magazine training (in skinner box)
a. When i hear the click, I see the pellet, and I know it means food (rat pov) - Shaping: reinforcing successive approximations
a. We want our rat to learn to press the bar
b. Rats are curious
c. Reinforce successive approximations: as rat moves closer to the bar…we give him fun pellet; each time this happens raise the bar for how close the rat must get to bar (or if he presses it) - Fixed ratio rewards: 1:1
a. The rat presses the bar 1 time, it gets 1 pellet
*we want to see the rat has learned - 1 log per rat…we turn into Mark (log is graded)
Reinforcing schedules
- Fixed ratio: 1:1, 2:1, etc.
- Variable ratio: i.e. average (mean) of 2:1, 5:1, 10:1, etc. (letters, approval, gambling)
- Fixed interval: i.e. each 30 seconds if there has been 1 appropriate response during the time interval
- Variable interval: i.e. average mean of 30 seconds (the behavior will lasts the longest, the behavior is not rewarded every time so it will take longer for it to stop)
- Extinction. The response is extinguished, and the behavior becomes extinct. Which lasts longer?
- Controlling the professor?
Optional:
*You teach the last behavior first, once you have good behavior.. Then you teach it other things…
- Secondary reinforcer: something that leads to reinforcement or can be exchanged for a reinforced i.e money or gambling chips
- Discriminative stimuli: an environmental stimulus that signals potential changes in the relationship between the response and the consequence
- Chaining: a sequence of secondary reinforces or discriminative stimuli
- Spontaneous recovery: the behavior was extinct but is emitted again after a time delay
Types of reinforcement: giving or taking away
Handling your rat…we can name the rat hehe
- A short discussion of ethics and animal treatment
- You are responsible for your rat!
a. Weight, water, health - Two students per rat- wear gloves, carry in a box
a. Hold rat from bottom of stomach and top of back… only squeeze a little
b. Let rat sniff and pet it - Only class members with valid/ working ID may enter
- Your own flexible schedule 8:00 to 6:00 (a little later is OK)
- Use hand sanitizer upon entering and leaving
- Dump tray after use. Spray and wipe tables
- About one hour per day
- Locks are 1, then 2, then 5
- Keep a journal (one per rat) which Mark will review; graded on science of the journal
- Rat parade
a. Last day of lab we show what the rat can do
Rubin (1970) Article
Measurement of Romantic Love
Author Name: Zick Rubin
Background:
The focus of the current research was restricted to romantic love, which may be defined simply as love between unmarried opposite-sex peers, of the sort which could possibly lead to marriage
Hypothesis: An attempt to introduce and validate a social-psychological construct of romantic love
Main Structure
- Develop Love Scale
- Questionnaire study
- Laboratory Study
Methods
- love scale asking about physical attractions, predisposition to help, exclusiveness/ absorbtion (3 main)
- questionnaire study (not at all true –> def true)
- laboratory study: couples read study about couple and examined reactions (stronger couples participated in more mutual gazing
IV: severity of love
DV: amount of gazing
Results: A distinction between love and liking was found but there wasn’t a true definition found of romantic love (stronger correlation among men than women)
Critiques
- demographics
how love is defined
- cross over between love and liking
- questionable items included (unhealthy relationships)
- what is truly desirable to people
What to measure: developing measures form constructs and questions
- Operationalize behavior
- Does the behavior represent the constructs?
- Can we infer what a person is thinking from the behavior we observe?
Who to measure- sampling procedure
- Samples vs. populations
- The sample should be representative of the population
- But what is the population? Inductive reasoning
a. All Americans?
b. Gender?
c. SES?
d. All ages?
e. All people in all nations/ cultures worldwide?
Who to measure- Sampling Procedure
Probability Sampling
- Random sampling
a. Each member of a population has equal chance of being selected as member of sample
b. Sometimes use a random number generator to select the population - Stratified sampling
a. Proportions are important subgroups in a population are representative precisely in sample
b. 75% female, 25% male (2 strata) - Cluster sampling
a. Randomly selected a cluster of individuals all having some feature in common
b. Campus survey → sample first-year students who live on-campus
Who to measure- Sampling Procedure
Non Probability Sampling
- Does NOT provide representative samples, but are easier to do
- Convenience sampling
a. Select subjects who are available and convenient (e.g, introductory psychology “subject pool”)
b. Purposive sampling (e.g Milgram non-use of university students) - Quota sampling
a. Similar to stratified sampling, but non-random
b. Often done with online surveys
Evaluating measures: reliability and validity
reliability
a. Is it repeatable?
b. Does it give us (almost) the same result each time?
c. What is the measurement error
d. Reliability = repeatability, consistency
Reliability and Validity
- Validity
a. Does it appear what it is designed to measure - Face validity
a. Does it appear on the surface to be valid? - Criterion validity
a. Does it accurately forecast future behavior?
b. Is it a useful measure of behavior? - Construct validity
a. Is the construct being measured a valid construct?
b. Is this the best instrument for measuring it?
Four Scales of Measurement
- nominal
a. Categories - Ordinal
a. Categories with a true order - Interval
a. Equal fixed intervals between items but no true dero. You can add bu not divide - ratio
a. A true zero. You can divide or multiply
Nominal Scales
- Assigns numbers to events to classify them into one group or another
- Numbers are used as names (categorical)
How used:
a. Assign individuals to categories
b. Count the number of individuals falling into each category (reported as frequencies) - Example:
a. Verdict: 0 = not guilty, 1 = guilty
b. Sparrow, robin, owl, eagle, wren
Ordinal Scale
- Numbers are used to indicate rank order
a. How used:
–> Rank order (1st,2nd,3rd, ect.) individuals (or entities) - Examples:
a. Four applicants ranked for admission to grad school
b. Sports team standings
Interval Scales
- Scores indicate quantities
- Equal intervals between scores
- Score of zero → just a point on the continuum
a. a score of zero does not indicate - How used:
a. Calculate score from participants’ responses on a test - Example
a. Temperature, IQ scores, scores from personality tests
Ratio Scales
- With ratio scales we can add, subtract, multiply, divide
- A ratio scale has a true zero
- Examples:
a. The number of people are enrolled in this class
b. The amount of money in an account
c. The number of people who test positive for a disease
Descriptive vs. Inferential Statistics
- Descriptive stats: Summarize and provide measurements about a sample or population
- Inferential stats: Permit us to draw inferences/generalizations - from what we have observed and measured
Descriptive Stats: summarize data
- Mean, median, and mode
- Range, standard deviation, variance
- Descriptive stats: summarize data
- Measures of variability
a. How spread out or dispersed scores are in distribution
b. Range, standard deviation, variance, interquartile range
c. With outliers → interquartile range better than standard deviation - Visual displays of data
a. Histograms from frequency distributions
b. With graphs, carefully examine Y-axis to avoid being misled
Statistical Analysis
Visual displays of data
–> CNN example - beware the Y-axis
Inferential Statistics
- Permit us to make distinctions and draw inferences
- Inferential statistics
a. Inferring general conclusions about the population from sample data
b. Examples → t-tests, ANOVAs
Statistical Analysis
- Null hypothesis Significance Testing (NHST)
- Null hypothesis:
a. NO relationship (“no difference”) between - Alternative hypothesis
a. A relationship (a difference) between variables in population is expected, given our sample
–>A researcher’s predictions often specifies the direction of the relationship (e.g a positive correlation between variables)
Null hypothesis significance testing
2 possible outcomes
- Reject the null hypothesis (with some probability)
a. Conclude you found a significant relationship between variables
- Fail to reject the null hypothesis
a. Conclude you found no significant relationship between variables
Because you are testing a sample and making inferences about the population, your statistical decisions have a probability of being wrong!
- Possible errors
a. Type 1 → reject null hypothesis, but be wrong
b. Type 2 → fail to reject null hypothesis, but be wrong
Interpreting failures to reject null hypothesis
- Extreme caution: This does not mean it is “proven false”
- May be useful if the outcome is replicated
a. Example → questions a claim for the effectiveness of some new therapy; useful if studies consistently show lack of effect of therapy
Publication bias and the file drawer effect
The journal of non significant results
- Beyond null hypothesis significance testing
- Effect size
a. Emphasizes the size of difference between variables, not merely whether there is a difference or not
b. Useful for meta-analysis
- Confidence intervals
a. Range within which population mean likely to be found
- Power
a. Chance of rejecting a false null hypothesis
–>Sample size an important factor
Why is an Alpha Level .05 Intuitive?
- When we intuitively look for another explanation?
- Ex. coin flipping example (we guess probability and say there is either no way)
Bandura & Ross (1963)
Imitation of Film-Mediated Aggression Models Slides
Author Name :Bandura & Ross
Background: Even in 1963, it was believed that children were more likely to replicate things they saw in pictorial media.
Hypothesis: Children are more likely to imitate aggressive behaviors from real models than fictional models.
Main Structure
- Children are exposed to aggressive stimuli according to their group assignment.
- They are then placed in a different environment that induces frustration
- The subjects’ responses to this frustration are judged to see if they imitate the stimuli they were exposed to.
Methods
- Control group was not shown aggressive behavior.
- Cartoon group was shown an aggressive cartoon.
- Film group was shown an aggressive film model of adults.
- Real-Life groups was shown a live aggressive model of adults
- children were then taken into a room and told they couldn’t play with the best toys (frustration)
- types of aggression
IV: type of aggression shown
DV: amount of imitative aggression displayed
Results
- subjects who were shown aggressive behavior were more aggressive
- Sex of Model did affect the Imitation in some cases
- The film group was influenced to imitate the aggressive behaviors of the model more than the other groups
Critiques
- test was not life like
- there is no long term validity
- no diversity within subjects
Essential Features of Experimental Research
Mill’s inductive logic
- Method of agreement
a. If X, then Y (sufficiency → X is sufficient for Y)
- Method of difference
a. If not X, then not Y (necessity → X is necessary for Y)
- Together → X is necessary & sufficient for producing Y
a. Agreement
–> Analogous to experimental group
b. Difference
–> Analogous to control group
–> More on logic at the logic class
Types of Treatment Groups
- Experimental group: actually receives the treatment in question
- Control group: does not receive the treatment in question. May receive a placebo.
Variables
- Independent: we manipulate (sort of or usually) the IV
- Dependent: we measure the DV
- Extraneous or confounding: we control these
Types of independent Variables (aka treatment groups)
- Situational variables: manipulation of different features in the environment may occur
- Task variables: different tasks, different levels of complexity, different scenarios
- Instructional variables: manipulated by asking groups to perform in different ways
Experimental vs. Control Groups
- Experimental groups (given treatment
a. Research example 6 → given a golf ball and told it was a “lucky” ball - Control groups (treatment withheld)
a. Research example 6 → given a golf-ball and not told it was a “lucky” ball
Subject (participant) variables: who the person is as they come to the experiment:
Male vs female
Age groups
Education levels
Diagnosed healthy
Subject variables
- Already-existing attributes of subjects in a study
a. Examples → gender, age, personality characteristic - Anxiety example
a. As a manipulated variable → induce different degrees of anxiety in participants
b. As a subject variable → choose participants who have different degrees of their typical anxiety - Research example 7
a. Subject variable #1 → culture
–> European Americans and east Asians
b. Subject variable #2 → gender
–> Women and men
Using both manipulated subject and subject IVs
Bandura’s Bobo study
- Manipulated → type of exposure to violence
- Subject → gender
Controlling Extraneous Variables
- Extraneous variable: uncontrolled factors that are not of interest but might influence the behavior being studied
- We try to hold constant the extraneous variables
- If we fail to hold extraneous variables constant, they can confound (confuse) the experiment. These are confounding variables.
Dependent Variable
- We measure the DV
- It must be observable, operationalized, measurable
a. Problems:
–> Ceiling effects: task is too easy, all scores become very high, disguising any differences
–> Floor effects: task too difficult, all scores very low, disguising any differences
b. solution:
Task of moderate difficulty, determined through pilot testing
Validity
- Are we testing what we claim to be testing?
- Statistical validity: are we using proper statistical and drawing propper conclusions.
- Construct validity: do our IV and DV actually represent and operationalize the construct we are investigating
External Validity
can our inferences from this sample be generalized to other populations?
…to other settings?
…to other groups?
…to other times?
…to other cultures
Internal validity
Does my study actually answer the research question I proposed and designed to answer.
Internal validity studies…
- Have valid operational definitions
- Have valid measurements
- Have no confounds
Threats to Internal Validity
How do we prevent uncontrolled extraneous factors from reducing internal validity
Pre-post studies
History and maturation
Regression
Testing and instrument
Threats to internal validity: history and maturation
- History: events occur, things change, the effects of the Parkland, Florida shootings
- Maturation: did 5th grade make you smarter or was it a year of maturation
Threats to internal validity: regression
- Regression towards the mean
- The tendency to regress towards the average due to fewer low-probability event
- Did berating improve the students pilot’s performance? Did praise diminish it?
Testing to internal validity: testing and instrument
- Did taking the test over and over make you score better? Was it the treatment or the test?
- Are the instruments themselves equal? Is the first test easier/harder than the second? (This is one way to force improvement)
Participant Selection
- Did all members of the population have an equal likelihood of getting into the sample
- Polling people by phone
- Polling people in a store (Food Lion vs. Fresh Market vs. Farmer’s Market)
Attrition
Why did some people leave the population?
Failure
Lack of preparation
Sickness
Death
Darley & Latane (1968)
Bystander Intervention in Emergencies: Diffusion of Responsibility
Author Name: Darley & Latane
Background: Several years prior to the study, Kitty Genovese was attacked and fatally stabbed by a man on the streets of New York City within earshot of 38 individuals who were in their apartments. While it took the attacker over half an hour to kill her, not a single person bothered to even call for help.
Hypothesis: The more bystanders to an emergency, the less likely, or more slowly, any one bystander will intervene to provide aid.
Main Structure:
Condition 1: 2 person (Participant and Victim*)
Condition 2: 3 person (Participant, Voice, and Victim)
Condition 3: 6 person (Participant, 4 Voices, and Victim)
Methods:
Discussion, fake emergency, how long to ask for help
IV: The number of people the participants thought were in the discussion room
DV: Speed with which the participants reported the emergency
Results:
- The number of bystanders present had a significant effect on likelihood of participants reporting the emergency.
- Bystanders responded quicker to the emergency when there were less bystanders present.
Critiques
- real world application/ limited external validity (could not see)
- participant diversity
-
A very Simple Approach to Research (10 steps to research)
- Form a question in your mind
- A construct
- Why do people behave this way?
- How can we explain it?
- What causes this? - Form hypotheses
- Null hypothesis: no effect
a. There is no difference
b. Mean 1= mean 2
- Alternative hypothesis: effect
a. There is a difference
b. Mean 1 =(NOT) mean 2- Operationalize the question
- independent
a. We manipulate the IV - dependent
a. We measure the DV - extraneous variables (confounding variables)
a. We control the EV
- independent
- Determine your sample
- The population and the sample
- Representative
- random
- Run your experiment and make observations
- In some ways this is the easiest part of the process. But it can be very time consuming.
- Record your data
- Measurement
- We often realize afterward that we didn’t record something (i.e age, sex, gender, education, etc.)
- Analyze the data using statistics
- What is the likelihood we would see a difference this great due to chance alone
a. Does mean 1 = mean 2 or does
b. Mean 1 =(NOT) mean 2 - Do we reject the null, hypothesis, or do we fail to reject it?
- The .05 and .01 alpha level
- Type 1 error: finding a difference when in fact none exists
- Type 2 error: failing to find a difference when in fact it does exist
*Hypothesis Testing
- What is the likelihood we would see a difference this great due to chance alone
- Make inferences based on the analysis of your data
a. Make generalizations based on what you see
b. Apply this to broader issues - not just the lab - Form conclusions and answer your original question
a. Return to the original question posed in the beginning of the research and…
b. Pose additional questions for future research - Build your conclusions into a theory
a. See if it is possible to expand your conclusions into a broader and more systematic theory
- Operationalize the question
Objectives of writing a research report
Science: to move science forward
- To share with colleagues
- To collectively seek truth
Selfish: to get published and funded
- To enhance your stature as a scientists
- To enhance your own career
The style and how to get it
- There are rules of style…by carefully reading, you will see well-done style
- You will demonstrate your ability to confidently and effectively communicate rigorous science and make it look easy
- Your research paper should be written in the same style as the journal articles we are reading and discussing in class
The Sections of an APA-Style Paper
Title Page
- Your title:
a. Make it interesting, professional, informative
b. Questions are OK
- Your affiliation
William & Mary
Abstract
a. A summary or Precis
b. This will appear for online journal searches
c. Enticement to download it or read on
Introduction
- What is the point of the experiment
a. What problem am I solving?
b. What question are we answering?
- How is this experiment related to other research in the area?
- How will the method and design I am lead to a solution and understanding of the problem?
- Structure
a. Start at a mutually understood point (listen to politicians, comedians, others do this . . . start at a place that is familiar to the listener)
b. Known literature, a common situation, a social problem (systematic racism, gun violence, eating disorders, income inequality, education, crime)
c. Gaps or holes in the literature
d. Controversy (everyone pays attention to a fight)
e. Challenge earlier literature
f. The hook
g. Draw the reader down the path with you so they want to learn more
i. The existing literature . . .
j. The existing lit as a second point of departure
k. How your study takes the science further – your work, method, research yields additional insight
l. The journal editor will ask a recognized authority to review your article … so be nice (my own experience)
j. Pose hypotheses (not Ho or H1)
k. The introduction is the first act … set the stage, keep the readers interest, and prepare the reader for the rest of the story
Method (no s in Method)
- How you operationalized it
- The reader should be able to replicate
- As if the reader were there with you in the lab
- Participants . . how many, gender (not sex), age
- If animals, age, sex, weight etc.
- Apparatus…devices used, model
- Questionnaire, survey (appendix)
- Procedure
- How was/were the Independent Variable(s) administered?
- How was/were the Dependent Variable(s) recorded?
- Design: Completely randomized, within subjects, etc. (stats class)
- How many participants in each group?
- Descriptive name (not group 1, 2, etc.)
- Control Procedures
Results
- Factually report your results … (don’t discuss yet)
- Start with the most simple section comparison
- Move on to a series of more complicated points
- (the reader should have been prepared for this)
- Figures should be self-explanatory … use them in your presentations
- Should we present only significant differences?
- … but what if there is no difference (men/women) and you want to eliminate something as a possible explanation?
- Significant and marginally significant differences
- The journal of non-significant results
- Don’t write “significant results.” They are “significant differences.”
- Never write “insignificant”!
Discussion
-Now it’s time to discuss … refer back to the Introduction
-Answer the questions you posed earlier
-Accept/reject hypotheses
-The data show, suggest, call into question, permit us to conclude
-Reinterpret earlier research
-Avoid prove/disprove; demonstrate is pushing it
Conclusion
- Briefly restate findings per results/discussion
- Implications for implementing results of findings
- Areas of further research
- Other Considerations
Ryan Article (2002)
Caffeine Reduces Time of Day Effects on Memory Performance in Older Adults
Author Name: Ryan (2002)
Background:
- Previous research regarding memory in people over 65
- Previous research on caffeine has found that caffeine makes people feel more alert and feel like it is easier to pay attention
Hypothesis: Can decline in memory performance (in people over 65) from morning to afternoon be lessened by having caffeine?
Main Structure
- recruit people through news paper (and give survey) ONLY morning people
- tested at morning & at night
Methods
- testing participants at 8am and 4pm
- get caffeinated or decafe before each test
- given memory test (immediate, short , & long delay)
IV: caffeinated or not
DV: memory test performance
Results
- Results supported the hypothesis that having caffeine before taking memory tests would eliminate the otherwise present decline in performance on memory testing from morning to afternoon.
- Consistent with the hypothesis that performance changes depending on time-of-day are somehow related to physiological changes, as caffeine creates physiological changes
Critiques
- what makes someone a morning person
- decafe still has a little caffeine
- external validity because subject were screen for any issues
Between-Subjects Design
- Essential if the IV is a subject variable
- The need for equivalent groups
- Challenges and solutions for equivalent groups
a. Random assignment
b. Matching - Different sets of subjects in each level of an IV
- Comparison is between 2 different groups of subjects
- Necessary when
a. Subjects in each condition have to be naive
b. Subject variable (e.g. gender) is the IV - Main problem to solve → creating equivalent groups
Creating Equivalent groups (for between-subjects design)
Random assignment
- Each subject has equal chance if being assigned to any group in the study
- Spreads potential confounds equally through groups
- Blocked random assignment
a. Involves assigning a subject to each condition of the study before the condition is repeated.
matching
- Deliberate control over a potential confound
- Use when:
a. Small n per group might foil random assignment
b. Some matching variable correlates with DV
c. Measuring the matching variable is feasible
Within-subject Design
- Each subject does all treatments
- Sequence effects: controlled by random order
- Progressive effects: also controlled by random order
- Carryover effect: when one sequence may produce different result than another sequence
- Also called repeated-measure design
- Same subjects in every level of an IV
- Comparison is within the same group of subjects
a. Used when comparisons within the same individual are essential (e.g. perception studies) - Eliminated the possibility that differences between labels of the - IV could be due to individual differences
- Main problem to solve → order effects
a. Progressive effects- the effect is the same from trial to trial
b. Carry-over effects (harder to control)
–> Performance on or experience in Sequence A-B may affect performance (i.e. ‘carry-over’) on sequence B-A
Controlling Order (sequence) effects
Counterbalancing
a. Altering the order of the experimental conditions
Testing once per condition
a. Complete counterbalancing (n!) (ABC, BCA, CAB → ABC)
b. Partial counterbalancing (when n! Is too large)
Testing more than once per condition
a. Reverse counterbalancing ABCDDCBA
b. Block randomization (all one before any repeats)
*how many different seq. are the if there are 3 treatments (3 x 2 x 1) n-3, n-2, n-1
Controlling Problems in Developmental Research
Cross-sectional design
a. Between subjects design
b. Potential for cohort effects
–> Worse with large age differences
longitudinal design
a. Within-subjects design
b. Potential for attribution difficulties
Cohort sequential design
a. Combines cross-sectional and longitudinal
Problems with Biasing
- Experimenter bias
- Experimenter expectations can influence subject behavior
a. Blind and…double blind
b. Controlling for experimenter bias
c. Automating the procedure
d. Using a double-blind procedure
e. Research example 9: giving older adults caffeine in the afternoon - Participants Bias
a. Hawthorne effect
–> Effect of knowing one is in a study
b. Demand characteristics
c. Cues giving away true purpose and study’s hypothesis
d. “Good” subjects
–> Participants tend to be cooperative, to please the research
e. Evaluation apprehension
–> Participants tend to behave in ideal ways so as not to be evaluated negatively
f. Controlling participants bias with a manipulation check
Controlling for Participant Bias
- Effective deception
- Use of manipulation checks
- Field of research
Ethical Responsibilities of Participants
- Be responsible
a. Show up for scheduled appointments, or inform research of cancellation - Be cooperative
Behave professionally when participating in research - Listen Carefully
a. Ask questions if unsure of your rights or of what you are asked to for - Respect the researcher
a. Do not discuss study with others - Be actively involved with debriefing
a. Help the researcher understand your experience
Reynolds (1992) Article
Recognition of Expertise in Chess Players
Author Name: Reynolds (2002)
Background: Reynolds chose the medium of chess to study and answer this question due to it taking years or even decades to master the game, the study of expert games being important to improve one’s own play, and its well-established and highly reliable rating system
Hypothesis: What kinds of information facilitate the identification of expert performance
Main Structure:
- 15 chess players
- 6 positions
- give a confidence score (1-10)
Methods
- they looked at moves before the initial and were asked to rate it again
IV: The participant’s rating served as the independent variable
DV: The difference between the participant’s guess of the rating and the actual rating of the players who produced the positions shown, or error, was measured as a dependent variable
Results:
- The higher the participant’s rating, the lower their average rating estimation error
- Significantly greater error when given solely the initial position, error decreased significantly for all groups after the first and second half-move pairs were revealed. Decrease in error due to revealing the third pair of half-moves was not statistically significant
- All participants were significantly more accurate and confident when looking at positions produced by players close to them in skill level
- Additionally, when participants were particularly unsure about the rating of the game, they tended to guess Elo ratings different from their own
- self-reference heuristic
Critiques
- internal bias (assumption of self-reference heuristic)
- lack of diversity
- no control group
- baseline accuracy for measuring expertise
The Theory of Signal Detection
Hit, miss, correct nondetection, and false alarm
Noise Curve
For most any sensory modality – sight, smell, sound, etc. – there will be some background “noise” e.g. distracting background light, scents in the air, background sound, etc. The intensity of this background noise may vary and this variance will often follow the normal distribution. The most frequent levels of intensity will be in the midrange, with extremely high and low levels of intensity occurring less frequently.
- Noise can be anything we can sense
The Signal Plus Noise Curve
When a signal is present the strength of the signal is added to the level of the background noise, producing the signal plus noise curve.
- It displaces the curve to the right (adding signal displaces to the right by intensity of the signal)
Establishing the detection threshold
A Detection Threshold is the level of intensity above which the signal will be detected, but below which it will not be detected. The Detection Threshold may be deliberately set or may default to the minimum point at which a signal may be detected.
- We must be able to differentiate a. 4 things: Hit, miss, correct nondetection, and false alarm
The hit
When there actually is a signal and the intensity of the signal is above the Detection Threshold, the signal will be perceived. This is a Hit.
- To the right of the threshold AND in a place under the signal + noise curve
Two right responses: hit or correct nondetection (CND)
When there actually is no signal – only background noise – and the intensity is below the Detection Threshold, there will be no perception of a signal. This is Correct Nondetection.
False Alarm
When there actually is no signal – only background noise – but the intensity is above the Detection Threshold, there will be a false perception of a signal. This is a False Alarm.
- Above my threshold but it was just noise
A Miss
When there actually is a signal but the intensity of the signal is below the Detection Threshold, the signal will not be perceived. This is a Miss.
- There was a signal, and i missed it
Sometimes we are presented with an easy decision
Sometimes there can be a separation between the Noise Curve and the Signal Plus Noise Curve. This is the case where the signal is very clear/intense and there can be no mistaking it. Where do we position our threshold in a case like this?
- We will only get hit & correct detection, but NEVER a false alarm or miss
If the curves are separated place the threshold between them
If there is a space between the Noise Curve and the Signal Plus Noise Curve, place the threshold between them and get only……..Hits and Correct Nondetections.
Where do we set our threshold if the two curves significantly overlap?
If the curves overlap we must decide where to place our threshold. What happens if we raise or lower the threshold?
What happens if we raise the threshold?
We will get more misses and CNDs, fewer hits or false alarms.
What happens if we lower the threshold?
We will get more hits and false alarms, fewer misses and CNDs.
Where do we set the threshold on a fire or smoke detector? Why?
We set the threshold low because we don’t want to have any misses. We realize we will therefore have more false alarms but we are ready to accept that.
Where do we set the threshold in a criminal trial? What does that mean?
In a criminal trial we set the threshold high because we do not want a finding of guilty if the person may be innocent. We accept that there may therefore be findings of not guilty when the person actually committed the crime.
Signal Detection Theory is applied in “shoot/don’t shoot” situations
Examples of difficult shoot/don’t shoot decisions.
- The USS Stark was attacked by an Iranian fighter jet
- The USS Vincennes shoots down a civilian DC10
- The police officer in the dark alley
How does signal detection theory parallel hypothesis testing?
Does signal detection theory apply to understanding the communication between people?
Yes
