Lecture 22: Critical Thinking About Psychological Theories

Question 1

What are 3 reasons why we need theory

Answer 1

1. Mapping out assumptions; observations are based on theories
2. Research techniques influence reasoning about psychology
3. Connecting research findings; integrating findings from different researchers

Question 2

What are 6 forms of pseudoreasoning and explain them

Answer 2

1. Circular reasoning; almost literal reformation of experimental findings, no additional reason that explains the results
2. Equivocation; one-word explanations, the same word describes several underlying processes —> without further explanation you can predict opposite results
3. Confirming the consequent; a fallacy where one assumes that if a certain consequence is true, the antecedent that caused it must also be true, disregarding other possible explanations
4. Straw target; bringing down the position of your opponent to be easily defeatable
5. False dilemma; make it look like there’s only two options when in reality there’s more alternatives
6. Common practice; doing something because everyone always does it that way

Question 3

What are 3 types of circular reasoning

Answer 3

1. Repeat the premise
   The Bible states that god exists
   The Bible is the word of god
   God exists
2. Premise presupposes the truth of the conclusion
   The Bible states that god exists
   The Bible is the word of god
   God exists
3. Premise is logically/semantically equal to conclusion
   Jan reacts more quickly to aggression-related words
   Aggression-related words are more accessible in the long-term memory of Jan

Question 4

What are the 10 criteria for evaluating theories

Answer 4

1. Descriptive adequacy
2. Precision and interpretability
3. Coherence and Consistency
4. Prediction and Falsifiability
5. Postdiction and Explanation
6. Parsimony
7. Originality
8. Breadth
9. Usability
10. Rationality

Question 5

What is overfitting and what are generalizability and complexity

Answer 5

The better the model fits the current data, the more likely it is that is does not fit the new data so well
Generalizability is the ability of a model to predict new data —> overfitting decreases the generalizability of the model
Complexity is the degree to which the model is susceptible to overfitting

Question 6

What are 3 pitfalls of model selection

Answer 6

1. Irrelevant specification problem = bridging the gap between informal verbal descriptions of theories and formal implementations, which can lead to unintended discrepanties between theories and their various formal counterparts
2. Bononi paradox = when models become more realistic and complete (which means they can predict outcomes better) they also become less understandable and less transparent (we cannot find theory to explain the findings anymore)
3. Identification problem = for any behavior there may exist many different models all of which are equally capable of reproducing and explaining the behavior

Question 7

What are 3 advantages of formal models

Answer 7

1. Can be used to know how well a model fits the data without NHST
2. Allows to say exactly how closely a model approximates the data
3. Gives additional info about the relationships between variables