Chapter 2: Research Methods

Manipulating Brain Activity

TMS

How to we determine causality then?

•In order to get around some of the issues we discussed with correlation we need to have more control over how the study is run •Using an EXPERIMENTAL DESIGN would allow us to directly test a hypothesis in a controlled way and find support for one hypothesis vs. another •This helps us solve issues with correlation such as the third variable problem

How do we gather data in psychology?

•We can observe and record people's behaviors, with and without interference •We can give them tests and observe their performance/behavior •We can measure their physiological reaction to stimuli, such as recording brain activity

How do we test claims?

1. Find a question 2. State Hypothesis 3. Gather Data 4. Analyze Data 5. Draw Conclusions

Testability

A scientific hypothesis needs to be testable/measurable

independent variable *IN EXAM*

An independent variable is what causes a change in the dependent variable. Example: Does texting cause more car accidents? Independent variable: Amount of texting Dependent variable: Amount of accidents Does texting (independent) cause more car accidents (dependent)

Variable

An object, event, idea, feeling, time period, etc. that you are trying to measure. There are two types of variables: Independent and Dependent

Negative Correlation

As the amount of beers consumed increases, the grade decreases. Negative correlation = When one variables increases the other variable decreases.

Positive Correlation

As the amount of hours studied increases, the grade increases. Positive correlation = When two variables increase or decrease together.

Standardized tests

Assess people's *knowledge and ability to reason* EG: SAT/IQ tests

Reaction time

Assessing how long it takes someone to perform a task

Directionality Problem

Difficult to know the direction of the correlation. Is less sleep increasing stress, or is more stress leading to less sleep?

Serendipity

Finding something valuable that you were not intending

Correlation

How do we examine whether two things are related to each other? For example, what if we wanted to know whether the amount of hours you study is related to your grade on a test? As the amount of hours studied increases, the grade also increases. These two variables are correlated

Eye-tracking

How long people look at a given stimuli and what patterns their eyes make

Memory Tests

How many items can someone hold in memory over a given period of time

Experimental Design

Hypothesis: Hot weather causes an increase in shark attacks Experimental Group: People on the beach in hot weather + Ice Cream Control Group: People on the beach in AVERAGE temperature weather + Ice Cream

Third variable problem is an example of poor *(internal validity)*

INTERNAL VALIDITY is the degree to which the effects observed in a study are due to the independent variable and not confounds (hot weather was a confound) Hot Weather Ice Cream Shark Attacks? Other things that could effect internal validity include: • *BIASES* in how you assign participants to groups • *CHANGES* in participants over time that are not accounted for, such as participants getting better the more times they do your study

Independent vs Dependent variable

Independent -> Dependent BUT Independent NOT <- Dependent Notice the inverse is weird: Do car accidents (dependent) cause more texting (independent)?

Ethics in Psychology:

Informed Concent Institution Review Board Privacy Access to Data

Case Studies

Intensive examination of an unusual person or organization. •Common for studies of patients who have an impairment that is unique, such as inability to form new memories

Falsifiability

It should be possible to test the hypothesis and potentially prove it to be false

Correlation Does Not Equal Causation

Just because two variables are correlated, it does not mean that one variable is causing a change in the other variable For example, ice cream consumption and shark attacks are correlated, but it does not mean that ice cream consumption causes more shark attacks

Validity (Construct Validity)

Let's say I wanted to know if stress caused people to get less sleep. •I measure sleep by asking them to fill out a questionnaire indicating how much sleep they've gotten over the past week. •I then measure stress by asking them how much food they've consumed over the past week Problem with this study? *My stress measure has poor construct validity. It does not accurately measure what it is supposed to*

How does science begin?

Need a good question

Behavioral Measures of Cognition

Reaction time, Memory Tests, Standardized tests, Eye-tracking

EEG

Records electrical activity from the brain that reaches the scalp of the head

fMRI

Records what regions of the brain are active during a given task

Why use correlations?

Some research questions require correlational studies for ethical reasons •For example, it would be unethical to make people talk on the phone while driving in order to test whether talking while driving causes more accidents •Instead we could correlate existing data on the amount of car accidents to the amount of time people talk on the phone while driving (SAFER TO USE EXISTING DATA)- easier and safer Correlational studies are easy to conduct and allow us to gain information on possible relationships •However, we need to do more work to determine what exactly is causing the relationship to occur

Behavioral Coding

Systematic observation and assessment of people's behavior in various situations. (OBSERVING PEOPLE'S BEHAVIOR FROM AFAR) Example: You would like to know if talking on cell phones causes impairment while walking

TMS

Use a very fast and powerful magnetic field to disrupt brain activity

Self-Report

Uses surveys and questionnaires to gather data on past behavior, personal attitudes, beliefs, etc. This information can then be used to answer scientific questions, such as "Does a person's attitude and beliefs affect how successful their romantic relationships are?"

How do we control for the possibility that it might be ice cream that is causing the attacks?

We can make sure that both groups eat the *SAME AMOUNT OF ICE CREAM*

Problem with Ebbinghaus' memory timeline study? (External Validity)

What if Ebbinghaus had a much better memory than the average person? What if other people have a different forgetting curve? *Because we don't know if his findings generalize to other people, this study has poor external validity*

Recording Brain Activity

fMRI & EEG