Social Psych Exam 1

Theory vs. Phenomenon

A theory is a systematic explanation of observed phenomena, while a phenomenon refers to an observable occurrence or event that is subject to study. Example: James Pennebaker trauma-writing phenomenon.

Inductive Theory Development

Specific observations to general theory. Example: Kitty Genovese was murdered and it led to the development of theories about the bystander effect.

Deductive Theory Development

Multiple observations to unifying theory. Different explanations for why they occur, but align with one theory. Example: Roy Baumeister Ego-depletion theory, where cognitive load determines memory capacity and ability to resist temptation, which was found in many different fields of study, such as psychology, nutrition, social psychology, etc.

Theory Evolution

Has to be testable/falsifiable. Example: Freud’s Id and Superego. It must fit data. It should generate research. Parsimony: Simplicity over Complexity is preferred (Occam’s Razor). Example: Clever Hans Horse

Design: Experimental vs. Correlational

Experimental: Manipulated IV, Measurable DV. Control, Random Assignment so you can assume causation, dispersing confounding variables. Standardization: procedure stays the same for all conditions except for manipulated variable.

Correlational: Systematic relations between variables -1 0 +1. No causation, no IV, control, RA. Path analysis: consider all paths at the same time, certain variables appear to be significant or important could be causal. Some variables are treated as causes, and others as effects. Example: Investment model in relationships. Paths of satisfaction, alternatives, investments all have to do with commitment. Researchers can see how much each variable in the prediction of the breakup accounts for the variance in the breakup as a measured variable.

Broad Approaches: Reductionism vs. Systems

Reductionism: explains complex behaviors and mental processes by breaking them down into simpler, often biological, components. Allows for certainty in causation. Ex. Discontinuity effects. Two groups interacting with each other are more likely to be competitive with each other than just two people. These researchers published many papers each about one independent variable and isolate one cause at a time. Build the variables brick by brick.

Systems oriented: Looking the entire complex system. How that complex system plays out in the prediction of an outcome variable. More correlational than causation. Ex. Caryl Rusbult’s lab; the investment model lab: the belief that there's a lot of things that might be affecting this ongoing close relationship. Another ex. five-year newly weds study, including many variables of a whole complex system.

Validity: Internal vs. External

Internal: The degree to which you are sure of the cause of the subject’s behavior. Links to experimental and reductionistic research, IV and RA → causation. Lower in correlational research. Ex. of internal validity in field experiment: Activating the part of the brain for aggressive thoughts. They had a research assistant alone in a pickup truck who would drive around within this small medium sized city. They had the truck come around and the procedure was this whenever there was just one car behind them with one person in that car, they would activate the procedure, which was when the light turned green, the truck didn't move. In the back window of the truck visible to the car behind them was a rack. It either had a contractor's level, or a rifle, that was randomly assigned. They manipulated independent variable, the rifle or contractors level, and measured a dependent variable, the aggressiveness of the response. There was more aggression when the rifle was shown. This part of the brain gets activated because it’s just in one’s visual sphere, and the phenomenon is called priming.

External: degree to which you are sure that a subject’s behavior is naturally occurring. Research in which subjects don’t know they’re part of a study have the most external validity. Study above was high in external validity. High in field experiments, observational studies. External validity suffers in lab experiments. Researchers often use deception to mitigate this. Subject is told the lab is about one thing when it’s about something else. Ex. subjects read a passage about architecture. Some have the vocabulary “old” in it. When they walk to turn it in, they see if the person walks more senilely.

Representation of Variables: Operationalization

How you choose to represent the variables in your study. Experiment: how am I going to measure IV or DV? Correlation: how will I measure those variables? Make sure you are measuring what you intend to measure so it reflects the theoretical concept. Ex. a study on the effects of stress on performance. How will the researcher make subjects feel autonomic nervous system arousal? Making sure that you are measuring that intended theoretical concept and theoretical hypothesis, not something else.

Laboratory Experiments

a research method conducted in a structured, controlled environment where scientists manipulate social variables to test cause-and-effect relationships

Pros: High in internal validity; IV, DV, RA, causation.

Cons: Low external validity.

What’s key is operationalization of the variable.

Observational Study

a research technique where scientists systematically watch, measure, and record human behavior in natural or controlled settings without changing any variables

Pros: High in external validity, cannot talk about causation.

Cons: technique to collect data can be difficult.

Field Experiment

Manipulating IV out in the real world, using a sample of people who don’t know they’re in a study. Ex. priming of aggression study in the truck. Maintaining external and internal validity takes a lot of effort and preparation. Ex. Priming in a wine store with different music. External validity, just people living their lives in the wine store, internal validity, IV, DV, causation.

Surveys

a standardized research tool used to collect self-reported data on people's thoughts, feelings, beliefs, and behaviors within social contexts.Interviews, free response, MCQ. Correlational data; measured variables apply, so it’s difficult to talk about causation.

Cross-sectional survey: one big survey of everyone at one time.

Pros: easy to administer and track, observe change over time. Easy to respond to.

Cons: No causation (path analysis can help). Conscious bias: people can lie, inflate information. Unconscious bias: People never use extreme values (1 or 10) or ONLY give extreme values. Passive, self-report. Not the same as actually measuring behavior. Ex. Exercise.

Experience Sampling

Give subjects a prompt and ask them to track/record whenever a certain thing occurs. Ex. Physical attractiveness linked to same-opposite sex individual-group interactions in college students. Students record every time they have a social interaction. Later, students rated those subjects based on physical attractiveness. Measures behavior in real time.

Pro: Important for getting behavior as it occurs in the real world relative to the prompt, over self reports.

Cons: Still self-reported, not actually measuring the behavior, correlational data.

Simulation Study

Used when it is difficult to get a sample in a specific situation or want to know what would happen in a situation.

Ex. Stanford Prison Experiment. Researcher Zimbardo recruited college juniors (all male, same socioeconomic status, homogeneous sample) in experiment. Some assigned to guards and others prisoners. Set up as a prison. Only rule: Guards cannot physically harm inmates. The guards had a power trip and mocked, dehumanized, and berated prisoners. Prisoners had mental health issues and breakdowns. Results showed what college-aged middle-class males imagined prison was like. Highly unethical.

Subject’s stereotypes of a situation. Correlational data.

Ex. Abusive relationships simulation. Computer program → preprogrammed responses about punishing or reinforcing. The player (subject) could stop at any time. Slow, increasing abuse left earlier. Scalloped pattern of abuse (some abuse, then reward, then abuse, then reward) played the game longer.

Archival Research

Collected data not for a psychology-reason, entirely different reason. Psychologist given access to use data to test theoretical hypothesis

Ex. study access to data regarding pledge drive, tapes of 10 days of how much money was given in any one hour period based on radio. Psychologists could make predictions regarding what is most effective to get listeners to donate i.e. matching gifts, physical gifts. Code for specific language i.e. “be a part of our family” or guilt-tripping language. Can test different hypotheses. Findings: “join our family of NPR station” worked the best, feelings of belonging; public health idea -> giving them a HOW to give mattered, explicitly giving strategies on how to donate like numbers or websites repeatedly

Psychologists have no say over treatment or data collection, only can analyze results

Pro: if you get permission, can do analyses on data in terms of psychological behavior. Interesting data, don’t have to worry that people know they are in a psychological experiment; high in external validity

Con: as a researcher, no impact on how data was collected. Have to go with the data as it’s presented to you. Can code on how you want to analyze the data, but cannot decide how deliberations go or how data was collected.

Another ex. Supreme court decision, majority vs minority opinion language. Testing for how political leanings interact with majority or minority opinions. Psychologists have access to these opinions. Finding: code for types of language used, political leaning doesn’t matter; majority opinions used language of “we are right/correct” more confidence than minority opinion

Quasi-Experimental

IV isn’t randomly assigned; IV is naturally occurring, compare IV to control group on a DV. Grouping that didn’t have any manipulation or RA. No assigned conditions. Assess change.

Pro: ability to study real-world causal relationships when true random assignment is impossible or unethical

Cons: could be a systematic difference not accounted for that explains a difference in the DV. 

Variable within a laboratory experiment

Common variable in QE: gender, ex. Helping behavior

Ex. size and scope of quality of people’s social network: depending on whether in a romantic relationship or not. Tracking changes over the semester. Data from the beginning of the semester, then at the beginning of each following month about romantic status and size/scope of social network. Can’t mandate/RA whether someone is in a romantic relationship/change. Finding: people who were single, then in relationship, social network shrank from time 1 to time 2. Oftentimes, you spend more time with that person at the expense of other friends; by time 4, people in a relationship had the biggest social network. People who are in a relationship then break up and shrink their social network as friends choose sides. CON Confounding variable ex. Big 5 personality characteristics people high in neuroticism are generally less liked in first impressions, but could be an important variable in romantic relationships and social networks.

Multitrait - Multimethod

Testing a theory one has, there is a reward in using multiple methods above. Theory is probably more true when proved in multiple methods. Multitrait: different types DV and IVs in the experiment ex. Multiple ways of getting aggression and frustration.

Potential Bias

Design/procedure problems; Demand characteristic biggest issue. Big problem if study has a clue in the study that tips off the subject to what is going on/expected behavior. If the person spots what the study is measuring, then their behavior is a reflection of what is expected or if they want to be different. Ex. Helping behavior study, if they know, they will act as if they would help. Threat to internal validity -> subject acts how they think they want. Threat to external validity -> not naturally occuring anymore.

Pretest the procedure many times to see if subjects can test what the study is about. Tweak the procedure to see if there is a lurking Demand Characteristic.

Experimenter Bias

Self-fulfilling prophecy: if I have an expectation, I will behave in a way to get you to act in that expectation ex. Element. Teachers were told 10 kids were the smartest, 10 kids were the dumbest (RA) teachers had expectations and taught accordingly: smart kids were more likely to be called on, positive reinforcement. Dumb kids were less likely to be called on, more negative reinforcement, less attention. Teachers' expectations affected their behavior, and the kids’ academic results reflected the assignment. Key component: randomly assigned

Comes in to play with research assistants interacting with the subjects

Ex. The research professor had first semester grad students be in his experiment (they didn’t know) to test and interact with subjects. He said there’s a phenomenon of undergrads accurately rating other undergrads’ intelligence

Procedure: one subject at a time, show the subject a feature and tell the subject to rate how smart they think the person is. For half of grad students, cards that had A feature was smart vs. B was not smart. The other half, the opposite way. Findings: grad students found statistical evidence that their one side were smarter than the other. The phenomenon is not real, but the undergrads were biased in asking subjects to rate intelligence, something in their body language to get the subjects to do the rating with their expectation. Blind is the best. Sometimes it is impossible, but you can have standardized interactions with subjects and research assistants. Have a script.

Participant Bias

When subjects know they're in an experiment/study, you’re aware of evaluation, which tends to make people apprehensive. Threat to internal and external validity. Dealing with it: make sure subjects know that their identity is not tied to anything within the study. Code numbers used only. Anyone analyzing doesn’t know the specific subjects’ answers. Another way is quasi-therapy, put subjects at ease and it's for the interest of science; not about that specific person. Subjects that don’t know they’re in the study don’t have to worry about this apprehension.

Issues of Ethics

Pre-Milgram study: researchers did not standardize ethics; a lot behaved ethically, but a lot of others definitely didn’t. Ex. John Watson’s Little Albert study, creating a phobia of a white rat and other white small, furry animals. Ethics up to individual researchers.

Milgram Study

In 1961, during Nazi trials, a psychologist named Stanley Milgram asked a question: how far can an « ordinary » person go when an authority figure orders them to do something morally unacceptable?

A volunteer was placed in front of a machine that delivered electric shocks. He was instructed to punish a « learner » (actually an actor) every time he gave a wrong answer. As the test progressed, the shocks increased: 75V… 150V… 300V… all the way up to 450V, marked « Danger: Severe Shock. »

In Milgram’s experiment, participants believed they were administering stronger and stronger electric shocks to the learner. Despite the screams, the pleas, and eventually complete silence, 65% of participants went all the way to the maximum 450-volt shock, simply because a man in a lab coat calmly told them to “please” continue.

Another psychologist realized this seemed very traumatic, tracked down og participants and interviewed them 9 months after. Those people were messed up, had intrusive thoughts, affected their life, and had nightmares. This report triggered a widespread response.

Post-Milgram Study

Panel of experts came together to create Belmont Report for institutional review boards

3 fundamental concepts of ethics when using human subjects in any field: informed consent (voluntary, warning about what the study was about, minors have to have a guardian who can consent),

risk vs. benefits(potential emotional states, the procedure, benefits: knowledge that will be gained),

debriefing (how are subjects going to have follow-up)

Social Cognition

How we think the world around us (especially the social world, ourselves, other people)

Schemas: organized knowledge structures that influence perception, memory, and behavior. Representations of concepts in our mind, such that some things are closely tied together (i.e. colors, food, red things, etc) about a concept closely, distantly, and not all linked to other things

Our brains are set up to take on language and representation, information sorted in our brains

Common schemas, or idiosyncratic schemas only related to an individual

Schema Generation/Development

• Experience: as we have experience with an object and the object gets labeled, we take that in. Just by interacting with an object, a schema forms. More individual.

• Operant conditioning (learning theory): notion that some schema generation representation of an object comes from external rewards or punishment; as we have response to the way we think, the information organized changes. We try language and get corrected (grammar, definitions). 

• Modeling: Occurs often in childhood. Imitation. Kids want to imitate what they see around them, not about reinforcement/punishment. Acquiring schemas occurs through this, and how we think about an object.

• Social comparison: when one doesn’t have a good rep of something or behavior or feeling, they are motivated to look to other people as a guide, forming a schema, even emotional states. Ex. taking a pill -> two experimental conditions: taking epinephrine. Subjects are told that they are taking it, and other subjects not told. Two subjects, one is a confederate. For half of the time, the confederate starts to act angry about the questionnaire. In the other half, confeds are in a euphoric mood. Then, have the two go to different rooms. Then another questionnaire asking what emotions they were feeling when filling out the first personality survey. Control group, informed epinephrine: don’t feel that much. Not told about epinephrine: match the emotion of the confederate. We learn a lot about schemas by being motivated to learn what other people think about things. Others have a dramatic impact on how schemas are formed. When we don’t have a schema, social comparison becomes important. How other people affect our behavior.

Modeling vs. Social Comparison

SC has a motivational component, motivated to make what other ppl think or feel because we actively do not know what to think or do. Modeling is simply imitation, not what someone else thinks about the situation. Modeling has a target.

Schema Structural Characteristics

• Dynamic yet stable. When a topic is new to us, not a lot of info, our schemas have a lot of room to change. Over time, they become stable and we value consistency. It takes a lot of effort to change a schema. We rely on schemas that are fully formed. Avoid trying to change schemas. Example: how kids see different dog breeds - resistant to change, but eventually able to change schemas. 

• Central nodes: types of schemas - if this is node activated, other schemas are activated ex. Halo effect physically attractive people are assumed to have other positive attributes. There is research that in a hospital, after a baby was born, time spent with the baby is based on attractiveness. Not universal, but quite common. Might have central nodes that are idiosyncratic to you. 

• Fuzzy boundaries: there will be overlap over located schemas; fuzziness on where one schema ends and one begins. Ex. love and lust share overlap. An fMRI machine measured what neurons were firing in certain people who said they were in love with a person, vs very attractive photo of someone - what fired in the brain for love vs lust.

Evaluation Characteristics

Seek meaning: why did someone do this? Why is this causing a certain emotion? We rely on schemas to fill in the blanks

Categorize quickly: this is what the situation is, using schemas to label. We don’t want to spend a lot of energy; we want to use schemas. However, there are times when you want to use mental energy.

Seek consistency: consistent with the schema activated, ignore things that are inconsistent with the schema. Confirmation bias, which leads to not spending lots of cognitive energy. Ex. Study where undergrads were gonna interview other undergrads. The person who was interviewing was led to believe that the person is high in introversion/extroversion. Let the interview process occur, and expectations shaped the questions and perception of the interviewee. Interviewers shaped their perception but also created that reality in a self-fulfilling prophecy. People independent of the study also fulfilled this after watching the interviews.

Types of Schemas

Person: schema about a person, complexity based on how well you know the person, personality, behavior, interactions, judgements. Ex. If you had to text one of your parents for money, which one would you ask?

Concept: schemas we have for objects, broad category for language/nouns.

Self: schemas about oneself. Tend to be very complex. All your interactions, experiences, unexpressed thoughts, oftentimes role dependent. When we think about work self vs. social self, etc. Many times we highlight what makes us different or unique

Group: schemas about certain groups; stereotypes. Sweeping generalizations about certain groups that activate schemas based on some information. Sometimes it has to do with confirmation bias.

Event (script): schemas for situations that illustrate dynamic vs. stable schemas. Some level of anxiety and nervousness that makes us likely to act in terms of social comparison. If we have information, we know what is going to happen.

Controlled vs. Automatic Processing

Controlled processing: performance/thought concentration

• Requires cognitive effort

• Concentrated behavior

• Single task oriented

• Ex. first learning to drive (gas vs. break, mirrors)

Put into controlled when: 

• High motivation

• Difficult task

• Novel task

Automatic: performance/though with little awareness

• requires less effort

• Often automatic behavior

• Multi-task

• Ex. drive car and talk on cell, sing with radio

Put into automatic when:

• Low motivation

• routine/easy task (brushing teeth)

• Tired

• distracted

Need for cognition: some people want to be in concentrated behavior

And vice-versa

There’s also people in between

Schemas: Priming

• schema activation

  • Covert priming: the person’s inability to realize it’s occurring; subconscious level

  • Example: visual perception study. Hitting certain keys based on color and shape as fast as possible. In milliseconds, they flash words about “helping”. Control: they flash neutral words. They get a ticket to go down the hall to finish the study. (Deception, high in external validity). People were more likely to help if they were covertly primed with “helping” words. People weren’t even aware that the priming occurred. But, if they are activated to being focused, such as having leaky pens that fall, the subjects wouldn’t help as there would be a ramification. Thus, they were moved to concentrated processing.

  • Overt priming: schema activation happens on a conscious level or awareness, but not focused on. 

  • Example: wine store music ethnicity. Prime activated the schema without knowing, but they could have been aware if they were concentrated.

Schemas: Perception

• unconscious mental frameworks that organize, filter, and interpret incoming information based on past experiences and cultural norms

• Ex. The Hannah study. Subjects in the study signed up for a memory study. They watch a video then recall as much of the video as they can. The back of the video is the same with a girl named Hannah. She is given an oral exam by a teacher, and she gets easy questions wrong and hard questions right (mixed performance). For one half of the subjects, she comes out of a high economic status house. For the other, she comes out of a low economic status house. Everything is the same in terms of the girl and the school. People who saw her come out of the nice house were more likely to remember when she got hard questions right, and people who saw her being poor remembered her getting easy questions wrong. Paid attention to schema consistent information; confirmation bias. Schema influenced their perception of Hannah. Perception AND memory study.

Schemas: Memory

• (encoding and recall)

  • Ex. The waitress-librarian study. A woman in a film is getting ready to leave her apartment to go out. In pre-testing, they identified stereotypical things of either waitressing or librarian and put those in the video. Overt activation and focus. Another IV: if you got that schema (label of waitress or librarian) before or after the video. Wait 10 minutes then try to remember everything from the video. People who got “waitress” were much more likely to remember the waitress things and vice versa for the librarian. Confirmation bias. People who got schema before the video could recall a lot more of the video, skewed in either waitress or librarian. But having context improved their ability to recall information because things made sense, especially the schema-consistent information. 

  • Elizabeth Loftus eyewitness testimonies; half subjects asked how fast the car was going when it “hit” the tree, others were asked how fast it was going when it “smashed” the tree. “Smashed” imagined more damage. Schema activated at the time trying to recall something. 

  • Memory often changes based on what schemas are activated, and how we store the memory. The memory is very faulty. 

  • Loftus found that you can change memory based on information given when prompted. Memories can also be created. Example: memories from parents, fake memories of being lost in a department store.

Schemas: Behavior

• Behavioral Confirmation: not only does the schema affect the person who has the schema activated, but also the person who they are interacting with’s behavior

• Ex. if you expect someone to be an extrovert, they will “act” like an extrovert based on what the person does

• Study: social interaction study over the phone with another subject, not face to face interaction. In every pair, there are real subjects. Guy and girl call each other. The guy gave answers to a survey the girl answered. They manipulate the headshot; half get an attractive female and the other get a less attractive female. Men were more engaged with the attractive female. Affected the women’s behavior. Recorded conversations and sent to other subjects to describe how people acted. The guy’s expectation elicited the behavior of the women. Our expectations of someone might affect how the person we interact with acts.

• Ex. interviewer asking questions based on established assumptions about personality.