Popular in PSYC4201W
verified elite notetaker
Popular in Psychlogy
This 5 page Class Notes was uploaded by Rachel Onefater on Saturday January 23, 2016. The Class Notes belongs to 76884 at George Washington University taught by Dr. George Howe in Spring 2016. Since its upload, it has received 51 views. For similar materials see PSYC4201W in Psychlogy at George Washington University.
Reviews for PSYC 4201W
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 01/23/16
Causation and Psychological Science: Many Terms we use are “infused” with causation: Influence Impact Teach Shape Mold Trigger Elicit Drive: what drives someone to act or feel the way they do Change:verb! Convince Excite Inhibit Causal: What things that happen are likely to increase risk later on? What factors can we do to resolve the stress? Do our interventions help prevent or resolve? Simple Physical Causation: → Schultz et. al (2007) studied preschoolers Set up a box so they can change how gears influence each other, and can set it so shaft and gear are turning together (pictures on slide). a. Strategy for Determining Cause : flip switch and you see that both are turning, and you flip switch and see if “B”turns, and if it does or doesn't, you learn something new! A.K.A. You see if one has an effect on the other and which gear the switch is attached to, or if the switch attaches to both Gear “A” and Gear “B” b. Results: After watching the interviewer go through the strategy, children should identify correct casual pictures, → children could also say what would happen at each stage of the strategy. → So early on, we use experience to construct causal ideas and use causal ideas to predict what will happen. *NOTE: Science develops from our experiences! Causal Reasoning ● We can infer(warrant→ may or may not be correct) that A influence B: ● When changes in A are followed by changes in B, given that the absence of change in A would have been followed by absence of change in B, all other plausible causes are equal 1. Changes in A a. In the example from today’s reading: “A” occurs as a specific event at a particular time and place, but in Psychology, we are interested in general kinds of types of “A’s” i. Example: A particular angry face in a pictures is one of a set of socially threatening events. We think about blame, rejection, shunning in society/in their culture, verbal attack. → So are we intered in change in more abstract conditions such as threat referred to as CONSTRUCTS *IMPORTANT: The cause will depend on what we are interested in. 2. bserving Change in A Two Methods a. Create the change by doing something (manipulation) b. Observe naturally occurring instances of change Two Elements of research design a. Experiments that manipulate stimuli, contexts b. Longitudinal studies that observe context over time and identify new events or patterns of change 1. Observing Changes in B Two Methods → Observing something more than once, and calculating some index of change over that time period a. Did some type of event occurs(e.g. Person responded correctly or not) b. Did some characteristic change? (e.g person became more anxious by some amount) → Observing Something later difference between groups who started out the same place (e.g. Williams study from Tuesday) ● Changes in A are Followed by Changes in B ○ Causation is asymmetric effect follows cause ○ We need to observe change in B that occur after changes in A → Observing for temporal priority Experiments: 1. Builtin we always observe effects after the manipulation 2. Timing of Observation depends on how long we think the effects last a. Lab studies: seconds to minutes b. Treatment or prevention trials: months to years to decades Longitudinal observation studies: 1. Often requires a minimum of three observation occasions a. Cause changes from T1 to T2 b. Effect changes from T2 to T3 Contiguity? (def.) Close to each other in space and time → So the two events have to be close to each other in space and time? Not necessarily: we can allow for causal chains, ormediation David Hume → “We reason about causation by associating constantly conjoined events” Repeatedly Conjoined Events → A Stable association between A and B ver multiple instances → In psychological science, what kind of instances? a. Multiple occasions for the same person? b. The same type of occasion for multiple people? WithinPerson Association: → Example: in the dot probe task, you are exposed to a threat stimulus 50 times. We divide your response into fast or not fast a. Which of the following patterns would reflect a stable association? BetweenPerson Association: → Example: 70 people participated the dot probe task after going through the verbal worry instructions We use each person's average speed to determine whether they are fast or not fast with the threat cues,. a. Which of the following patterns would reflect a stable association between worry and speed. Adding a “What if” i.e. If “A” had not happened , then “B “ Would not have happened → more recent work(Paul Holland, Don Rubin, Judea Pearl) adds another condition: a. “When, if X had not occurred, Y would not have occurred” b. Called the counterfactual:what would have happened if things had been different. c. Many instances in literature: “what would have happened if the South had won the Civil War?” Rubin Causal Model : → If for one person we define: → Y(X=1) as the outcome when X was present → Y(X=0) as the outcome if X had not been present a. The the causal effect for that individual is defined by b. Y(1)Y(0) c. If that value is zero, there is no causation(i..e Y is the same regardless of whether X occurred or not) *If it hadn’t happened, we would not have had a change, but you can;t do this becuase the counterfactual DOES NOT EXIST. The central problem of inferring causation from observational: We cannot observe both X and notX at exactly the same time. → But we can come up with a way to estimate the counterfactual? a. Find an appropriate conditions: “counterfactual carpentry” In exp. studies, the substituted counter actual condition is often called control or comparison group. Both experimental and control conditions need to be chosen carefully to allow causal reference Example : Experimental constion→ presence of threatening stimulus (i.e. ANGRY FACE) Control Condition→ absence of threatening stimulus: which one ? (i.e. NONANGRY FACE) 1. Experimental condition : involvement in yogabased stress management program 2. Control Condition : not being involved in stress management program: Which one? → no program at all → Vigorous exercise program of the same length → Program of gentle stretching of the same length → Stress management program that focuses on selfaffirming messages Answer: It depends on what you want to study Observing Over Multiple instance: → Allows us to observe both conditions (factual and substitute and counterfactual) Causal inference depends on whether outcomes differ across conditions. Ex: in the dot probe task, you are exposed to a threat stimulus and a nonthreat stimulus,but most causes have only partial effects, so we then compare probability: 68% vs. 58% This leads to s light change in our causal reasoning: We can infer that A influence B: → When changes in A are followed by Change sin B, given that absence of change in A would have been followed by absence of or different pattern of change in B, all other plausible options being equal. Observing other multiple instances : → Another reason to include multiple observations (or participants) in studies: → For example Under certain circumstance , comparing groups can be reflected in causation → Counterfactual Assumption: Groups equivalent on all other plausible causes a. Make sure the groups do not differ from each other
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'