PSYCH 290 - Midterm 1

Chapters 1-7, Psychology of Personality at the University of Michigan

What is the psychological triad?

How people think (thoughts), feel (emotions), and behave (behaviors). Often overlaps, and conflicts with each other. For example, a generous person not donating to a homeless person doesn’t make them not generous. Awareness of the triad conflict leads to resolving it, with brings it back to the “balance.”

Personality

An individual’s characteristic patterns of thought, emotion, and behavior. Must be consistent over time and across situations in different environments.

Basic Approach to Personality Psych

Limiting the amount of information you look at — simply searching for the strongest and most repeating patterns. Contains all types of approaches we look at.

Every human being is…

Like ALL others (human nature), like SOME others (group similiarities), like NO others (individual differences).

Trait Approach

How do people DIFFER from one another? What are the fundemental dimensions of personality? How psychological traits and differences in these traits may be conceptualized, measured, and followed over time.

Openness

Open to new experiences, intellectual, high amount of creativity/imagination.

Conscientiousness

Organized, self-motivated, driven, cleanliness, hard-working

Extraversion

Sociable, outgoing.

Agreeableness

Get along with others, caring, emphathetic.

Neuroticism / Emotional Stability

Anxiety OR calm, mood swings.

Biological Approach

How is personality influenced and expressed by biological processes? Ex. Phineas Gage rod through Prefrontal Cortex, lacked impulse control & emotion regulation after, change in personality.

Psychoanalytic Approach

What are the mechanisms that drive personality? What is the role of the unconscious?

Explicit vs Implicit Attitudes

Conscious attitudes that are easily reportable and endorsed, vs involuntary, uncontrollable, and unconcious.

Cognitive Approach

How do cognitive proceses and subjective experiences shape personality? How does personality shape cognitive processes?

Positive Illusions

When people overestimate their true abilities. Is it helpful or harmful? Adaptive or maladaptive? (Cognitive approach would seek to answer that question)

Cultural Approach

How do social and cultural contexts influence personality? Culture differences (i.e. Individualism vs Collectivism), family (i.e. birth order).

Personality Disorder Approach

What aspects of personality are adaptive vs maladaptive? What are the clinical manifestations of personality?

Social Learning Theory

Suggests inference should be drawn from looking at how observation and self evalutation determine how behaviors are formed. We can look at others to learn.

ONE BIG THEORY.. and its problems

Historical goal to create one big theory to psychology of personality… but it wouldn’t really work. If one does everything, it won’t do anything particularly well.

• A toaster is worthless if you want coffee and music.

• Personality psych uses many approaches individually, rather than all at once.

Funder’s 1st Law

Advantages and disadvantages are so interconnected that they may be inseperable. Great strenghts are usually great weaknesses, and vice versa depending on the context.

What is the goal of personality psychology?

To explain the WHOLE PERSON and how people function in their daily lives.

Funder’s 2nd Law

There are no perfect indicators of personality. There are only ambigious clues.

Funder’s 3rd Law

Something beats nothing. Gather as much evidence as you can, even if results can occasionally be misleading. The more info, the better.

B-Data (Behavioral)

See what the person does in a real-world scenario. Explain natural behavior.

• Advantages: Wide range of contexts, we can view different scenarios (addressing a part of personality directly), and the apperance of objectivity (appears to be unbiased or natural)

• Disadvantages: Extremely expensive (best method would be to use a private investigator) & subjective results (researches have to use own judgement to determine & interpret what an action really means)

• Examples: Diary entries, naturalistic observations, experiments, phsyiological data (ex. blood pressure, skin moisture), social media activity

What is a work-around to make behavioral observation data a little cheaper?

Participants self-report the ACTIONS of their day without any input in regards to interpretation or bias in any way. This avoids being S-data because they are only reporting on their actions, not opinions on how they felt or behaved on those actions.

The Delayed Gratification (Marshmallow Test) Behavioral Obsevations Example

Would kids wait to eat the marshmallow to get more marshmallows in the future, or do they eat it now and not get any more? Found to be able to predict higher SAT scores and greater academic and social competence, tolerance of frustration (L-Data as well).

• However, findings were weaker when demographics were considered (family backgound and home environment, early cognitive ability). Even weaekr associations for behavioral outcomes.

L-Data (Life-Outcomes)

Looking at archival information, already objective and existing info to determine where they progressed in their life.

• Advantages: Objective and verificable. Researchers do NOT need to touch it, strong relevance psychologically

• Disadvantages: Multidetermination (behavior is caused by multiple factors working together, so hard to pinpoint a source)

• Examples: Criminal records, SAT score, marriage, career outcomes, living spaces

I-Data (Informant Report)

Data are provided by others, judgements made by “informants” who know you very well who can speak to your personality and behaviors across sitations.

• Advantages: Large amount of information, real world basis, potentially less biased (no fish in water effect), some I data is true by definition (if they are likeable, others will speak highly of them)

• Disadvantages: They don’t see you when the curtains fall down, limited behavioral information. Subject to error and bias.

• Examples: Questionaire filled out by friends, family members, colleagues

S-Data (Self-Report)

Information provided directly from the subject. Often from a rating scale (just like I data) and you have access to the psychological triad directly.

• Advantages: Large amount of information, you know your own true thoughts, feelings, and intentions, simple/easy, some data is true by definition (if you have high self esteem, you will speak well of yourself)

• Disadvantages: Subject to bias, error, too simple and too easy — maybe data is too surface level?

• Examples: Questionaire filled out by the subject.

Behavioral Observations of how happy people spend their time, association with happiness.

Sitting alone or engaging in small talk has a negative association with happiness, whereas talking to others and especially substantive conversations have a positive association.

Projective Tests

Looking at an ambigious, meaningless image — your interpretation reveals inner thoughts, feelings, and experiences. Always yields B-DATA!!!!

• Example: Rorschach Inkblot Test, TAT

• Because Inkblot Test is so ambigious, there is NO reliability or validity

Thematic Apperception Test (TAT)

Asks individuals to tell a story from a series of drawings/images of people and ambigious events that are designed to reveal needs and implicit motives.

• Once again, interpreting what this means is quite subjective and can be inaccurate.

Ambigious Situations Task & Implicit Racial Bias

Evolution of the TAT where researchers measured the same situation but with reversed roles of skin colors. This test was found to be effective measures of implicit racial bias, both reliable and valid.

• Image in which black girl was standing and white was down, participants said she pushed her.

• Image in which black girl was down, and white was standing, participants said she fell.

Objective Tests

Results can be detected at a glance. Think yes/no, true/false, numeric scale. Less open to interpretation. Results should be similar (in regards to the type of answer) with different people. ALWAYS YIELDS S-DATA!!

• MMPI = Yes/No questions

Rational Method

Each question must be worded to mean the same thing, whoever answers the questions must have a good understanding of the questions, must self-report accurately, and all questions must be valid indicators of what the test is trying to measure.

• All must be met to measure an attribute correctly.

Example - A researcher develops a test for anxiety by selecting questions about excessive worry, nervousness, and physical tension because these behaviors directly reflect anxiety.

Factor Analytic Method

Identifies groups of things to have something in common (the commonality IS the “factor”). Asks a set of questions to find a pattern in their respones, and determine the factor. Identifies patterns of correlations among responses and groups related items into underlying personality dimensions.

Example - A psychologist gives thousands of people a personality questionnaire and uses statistical analysis to identify clusters of related responses, such as sociability, energy, and assertiveness.

Empirical Method

Seperating people into already opposite/different groups, THEN administering the test. Use patterns of responses to make inferences. Retains items based on their demonstrated ability to distinguish groups or predict outcomes, even when their content seems unrelated.

Example - A psychologist gives a large group of people with a history of rule-breaking and a group without such a history hundreds of questionnaire items. The psychologist keeps items that reliably distinguish the two groups, even when the questions seem unrelated to rule-breaking.

Reliability

Results need to be consistent. If a question asks something that differs day-to-day, it won’t be reliable. For example, “are you happy RIGHT NOW?” can differ, making it not reliable.

• Low precision, state of the participant and the experimenter, variation in the environment can all undermine reliability

Test-Retest Reliability

If you take the same exact personality test 6 months apart, the results should be very similar.

Internal Consistency

Are the methods/items being tested for one thing correlated with each other? In other words, is there conistency within all the questions that ask for a single thing?

Inter-Rater Reliability

Are multiple raters giving similar assessment?

Validity

How well does a test measure what its trying to measure?

Face Validity

Does it look like it measures it? Very shallow. Does it smell like, look like, seem like?

Construct Validity

Requires extensive research and review with different dimensions, items approved by experts.

True or False: Something can be reliable, not valid.

TRUE! A question like “what is your birthday” will be accurate and the same every time, but it isn’t exactly measure.. anything.

True or False: Reliability is implied with validity.

TRUE! If a question can measure what it’s supposed to measure, then the results are gonna be the exact same every time you ask the same person, therefore, the question is reliable as well as valid.

Case Study

Looking at a single, indepth case of an individual or very uniqe group or event. Descriptive and valuable for GENERATING a hypothesis, but you cannot TEST a hypothesis because it is not generalizable.

• Ex. Phineas Gage

• Advantages: Potentially lots of info, can examine unusual situations or people, and can generate hypothesis

• Disadvantages: Not suitable for hypothesis testing or generalizable

Correlational

Identify naturally occuring relationships with 2 variables.

• Ex: Ppl. with Prefrontal Cortex damage tended to have more difficulty recognizing self-concious emotions and made inappropriate disclosures

• Advantages: Can examine naturally occuring relationships, some variables can’t be manipulated by other factors

• Disadvantages: Difficult to determine the direction of the causality. Subject to the third variable time

Experimental

Test hypothesis by manipulating variables. Independent (what we change), dependent (how we measure change), and control group. Can determine casuality because variable manipulation and control group. Subject to the third variable problem.

• However, some variables (like culture, family dynamics) can’t be manipulated

• Can create certain unnatural contexts, raising a question if something is actually unrealistic?

• Advantages: Can provide powerful test of casuality

• Disadvantages: Are we manipulating what we think we are, unnatural contexts created, some variables can’t be manipulated.

Longitudinal

Repeated observations of some variable over time.

Replication

Same study, different participants, different labs.

Stability

Repeating study is best way to evaluate stability of results.

Publication Bias

Tendency of scientific journals to only publish studies with strong results.

Replication Crisis

Recent phenomenon in science where researchers find they cannot repeat or reproduce the same results of published experiments. Raises concerns about the reliability of previous studies.

Open Science

A global movement to make research significantly more transparent. Sharing data, findings, methodology, and code with each other.

External Validity

Real world relevance. The highest is case study (because it looks at an event that has already physically happened), and the lowest is experimental (because we are simulating events).

Internal Validity

More experimental control. The most is experiment (because we get to manipulate and control everything) and the least is case study (because we can’t change anything).

Measurement Error

Occurs when an individual’s scores and results flucuate from day to day. Can’t use a current state to measure a trait. = Low Reliability

Generalizable

Measure is valid across contexts, populations, and implementation. Generalizability assums reliability and validity, and can be implemented across contexts and populations.

Threats to Generalizability (WEIRD)

W - Western

E - Educated

I - Industrialized

R - Rich

D - Democratic

Samples

*Gender, religion, race are also included

How can we improve replicability?

1. Provide more detail about methods, results, and hypotheses

2. Report all data, measures, conditions, exclusions

3. Use larger samples (especially for small effects)

4. Pre-register hypotheses before data collection

5. Conduct replication studies

6. Publish non-significant (null) findings

Essentially, open science practices allow us to replicate results.

Unreliable measures cannot be valid. True or False?

True! Validity assumes reliability.

Reliable measures assumes validity. True or False?

False. Reliable measures aren’t always valid. For example, what is your birthday? Is reliable because results will be consistent, but it fails to measure… anything really.

Rise of Situationism & Related Studies

1960’s featured a rise on the emphasis of the power of the situation to shape behavior.

• Asch Conformity Study where participants were shown lines and asked to compare them in length, whereas the other people (actors) deliberately gave the wrong answer, and the real participant often voted with the group’s incorrect answer to align with the group, despite the real answer being obvious.

• Milgram Obedience Study where participants were asked to shock “learners” (actors) who got answers wrong in increasing voltages, with some lethal, and participants often administered the shock anyway in response to the authority’s direction.

• Zimbardo Stanford Prison Study where college student participants were placed into a prisoner or officer role, and simulated a prison for multiple days. They had to cut it short, as the officers used brutal force and it became a dangerous environment.

Walter Mischel’s (1968) Argument

Personality doesn’t predict behavior (low r=.3 correlations), people act differently across situations and that our intitutions about personality are flawed.

Personality’s Response to Michel

1. Personality traits predict behavior OVER TIME. Extraversion is a good predictior of number of parties attended during a semester, but a weak predictor of party attedance on any given night.

2. Broad traits predict broad behaviors, narrow traits predict narrow behaviors. Conscientiousness may predict overall academic performance, but punctuality predicts if you’ll show up to class on time.

3. People are consistent relative to others. Rank-order consistency.

4. Personality traits explain AS MUCH varience as situations do (r=.3-.4)

5. But..we can still do better cause we can conduct more research outside the lab

Aggregation

Averaging across time or a set of behaviors enhances the reliability of the measurements.

P-Level

In Null Hypothesis Testing (NHST), shows the probablity that your observed results (or larger) occur, assuming the null is true. .05 is typically the cutoff.

Significance Testing

Test to see how likely results were found by chance using the “p-level” (significance level).

Correlation Coefficient

Determines the strength and direction of an association.

• R=-1 is strong negative

• R=0 is no correlation

• R=1 is strong positive

Type 1 Error

Thinking one variable has an effect on another when it really doesn’t.

Type 2 Error

Thinking one variable doesn’t have an effect on another, when it really does.

Effect Size

A number to describe the extent of a relationship between two variables.

P-Hacking

Analyzing data in various ways until one finds the desired result.

Personality vs Situational Factors

Traits, genetic dispositions, free will vs Obedience to authority, peer pressure, social roles.

Trait Approach

Individuals differ in patterns of thought, feeling, and behavior.

• Relatively consistent across situations

• Emphasis on personality traits

Situationist Approach

Behavior is too inconsistent across situations to be characterized by traits

• Situational/Context-based

Fundamental Attribution Error

Overemphasize someone's personality traits while ignoring external situational factors. Attributing a behavior only to personality traits, rather than the situation.

• Flaw of trait approach

• Ex: Assuming someone is an meanie head for cutting me off in traffic, and not thinking that there may be a reason for them cutting me off outside of their trait.

Person-Environment Transactions

Both personality AND situations are important predictors of behavior.

1. Personality and situations influence each other.

2. Personalities are drawn to certain situations.

3. Personalities can change situations.

Weak Situation

More variation in behavior, personality is more influential.

• There is a low situational constraint, meaning you can be a lot more free to be you.

• Examples include parties, dorm room, lounge, park.

Strong Situation

Less variation behavior, situation more influential.

• There is a high situational constraint, as there is a social expectation for how act in these environments.

• Examples include church, job, interview, class.

Manipulation to the Stanford Prison Experiment

They originally put out two different advertisements: only one version mentioned prison life. 4x more volunteers in the non-prison study.

• Participants who volunteered for the prison study were lower in empathy and altrusim (selfless)

• However, they were higher in aggressiveness, narcissism, social dominance (belief in group-based hierarchies), and authoritarianism (expecting submission to authority)

Interactionism

Individuals and situations interact to shape behavior together.

• Short Term: Situation best predicts behavior in specific situations.

• Long Term: Personality traits better predict patterns of behavior across relationships, work settings, economic decisions, and overall life situations.

• Trait Combination: People with different personality traits may choose different situations.

• Situation Combination: Situations are affected by the personalities of the people who inhabit them.

Concensus

Do others agree?

• Inter-rater agreement (reliability) — multiple people make judgements about the same person and come to similar conclusions.

Accuracy

Does our “target” agree and do our judgements PREDICT behavior?

• Predictive Validity

Convergent Validity

Many different types of information CONVERGE and lead to the same conclusion.

Dorm Room Study (Consensus & Accuracy)

• Undergrad observers rating students’ personalities only based on their personal living space. (Using Big 5)

• Accuracy defined as correlation between judges, close friends, and self report by target.

• OPENNESS has highest consensus AND accuracy, followed by conscientiousness for consensus.

• Emotional Stability (Neuroticism) has the lowest concensus but second highest accuracy.

Moderator

A variable that affects the relationship between two variables.

• Moderators of Accuracy: Judge, Target, Trait, Information

The Judge (Moderators of Accuracy in Personality Judgement)

Person doing the judgement. Must be emotionally stable, less biased, less shallow. Generally positive and higher in emotional intelligence.

• To become a better judge, invest in relationships (get more meaningful)!

• Women are generally better judges.

The Target (Moderators of Accuracy in Personality Judgement)

Person being judged. Dependent on consistent behavior across situations, and how transparent vs conceling of internal experiences they are.

• To become a better target, be more open and consistent

The Trait (Moderators of Accuracy in Personality Judgement)

Trait that is being judged for. How easy is it to judge the trait? Big difference between openness (creativity!) and neuroticism (clearly u anxious bro).

• Visibility of the trait is key as well, which is also affected by the context of where the trait is judged.

The Information (Moderators of Accuracy in Personality Judgement)

Characteristics the judgement is based on (quality and quantity). The more information, the better. (Funder’s Third Law!)

• Quality of the information (relevance) is also crucial.

Self-Fulfilling Prophecy / Expectancies

What you expect of others impacts how you will interact with them.. which in turn, affects how they interact with you. Leads them to behave how you wanted them to.

• Example: A teacher believes a student is exceptionally smart, which motivates the teacher to give the student extra attention, causing the student to perform better as a result. The original expectation becomes true, partly because of the teacher’s behavior.

Opportunities

Personality judgements can affect reputation, which can affect opportunities.

Realistic Accuracy Model

Four stage process necessary to achieve accurate personality judgements.

1. RELEVANCE: The person being judged must do something RELEVANT/informative about the trait being judged.

2. AVALIABILITY: The information must be AVALIABLE to a judge.

3. DETECTION: The judge must DETECT this information.

4. UTILIZIATION: The judge must UTILIZE this information correctly.

Example - The Devil Wears Prada

1. What relevant cues are there for Miranda to judge her fashion sense?

2. Are the cues to Andy’s fashion sense present in this situation?

3. Did Miranda notice the cue?

4. Did Miranda utilize the cues to judge Andy

How is accuracy & concensus affected by the amount of information avaliable?

Concensus slightly increases with amount of info avaliable, wheras accuracy dramatically increases with the amount of info avaliable.

“Thin Slice” Judgements

Generally, more information is better. However, brief impressions can sometimes be accurate

• Many traits can be judged fairly accurately in only a few seconds.

Thin Slice Judgement Study #1 — Small Groups

250+ psych students placed in small groups, where members has zero acquaintance (didn’t know each other and were not allowed to talk)

• Sat in silence for a few minutes, then rated each other on personality traits similar to the big 5

• Hair, clothing, and physical cues gave it away.

• High consensus and accuracy with sociable (extraversion) and responsible (conscientiousness)

Thin Slice Judgement Study #2 — GSI Muted Video

13 GSI’s videotaped while teaching, and compiled 3 randomly selected 10 sec SILENT clips to show to 9 unfamiliar students.

• The students made personality judgements, which were compared to answer with end of semester evaluations by actual students.

• Found a correlation of r=0.76 between stranger’s ratings and the actual students, crazy consensus.

Consensus/Accuracy meets Reliability/Validity

Low reliability (consensus) limits validity (accuracy). But a reliable measure is not necessarily valid.

• Judges can agree.. but they could still be wrong!!