Self-Control and the Nature of Science

Question 1

benefits of self-control according to marshmallow study

Answer 1

• 4-year-olds who delay gratification for longer have better outcomes 10+ years later:
• • More socially skilled (parent’s ratings)
• • Better in school (parent’s ratings)
• • Higher SAT scores

Question 2

benefits of self-control according to other research

Answer 2

Self-control appears to be perhaps the best predictor of university grades (better than IQ or standardized test scores)

Question 3

why self-control is hard

Answer 3

• Present looms large, future is a distant consequence that almost seems like it’ll affect someone else
• • We are heavily influenced by what’s in the present
• • We perceive our “future selves” almost like a different person

Question 4

self-control strategies

Answer 4

• Bringing distant rewards/costs into the present (ex. Wearing tight jeans to potato chip eating contest)
• Distance self from current temptation (ex. Kids covering their eyes so they won’t see marshmallow)

Question 5

self-control as muscle

Answer 5

You can exhaust it -> if you’ve just used a lot of it, it might be hard to use more of it (ex. If you’ve just resisted eating cookies, it may be hard to use self-control to force yourself to work on a difficult puzzle)

Question 6

ego-depletion

Answer 6

• coincides with the “self-control as a muscle” perspective
• doing one task that requires self-control impairs performance on a subsequent task that also requires self-control
• However, not all studies showing ego depletion could replicate, and in a giant replication using 20+ labs, no effect was found

Question 7

replicability crisis

Answer 7

• well-known effects don’t seem to hold up to large-scale replications
• This is an issue for all of science, not just psychology
• How did we get here?
• • Publication bias
• • Small samples
• Norms are rapidly changing to try to prevent this (ie. Larger samples, pre-registration)

Question 8

p-hacking

Answer 8

analyzing the same dataset a bunch of ways and only disclosing the significant results

Question 9

what to watch out for when reading articles (potential indicators of p-hacking)

Answer 9

• Small samples (ie. <20 people per condition)
• P values close to .05 (ex. .04)
• Random covariates (ie. Controlling for age)

Question 10

False-Positive study (Simmons): what did they do?

Answer 10

• had people listen to music to see if it changed their age (deliberately impossible study)
• controlled for random measures
• used researcher degrees of freedom to find a false positive result, demonstrating how easy it is
• put forth recommendations to avoid this problem

Question 11

False-Positive study (Simmons): what do they recommend to reduce problem?

Answer 11

• decide when to terminate data collection before study begins (and report this)
• use at least 20 observations
• list all variables collected in a study
• report all experimental conditions, including failed manipulations
• report results of any eliminated measures had they been included
• if analysis includes covariates, authors must also report results of analysis without covariates

Question 12

False-Positive study (Simmons): what do they NOT recommend to solve the problem?

Answer 12

• changing alpha levels
• using Bayesian statistics
• doing conceptual replications
• posting all materials and data

Question 13

what is the culprit of false positives?

Answer 13

• researcher degrees of freedom (the decisions researchers make when conducting a study)
• often self-serving, especially in ambiguous cases (ie. what are the outliers in this study?
• Likely feel that the best decisions would be ones that end up leading to significance