Concepts (page 28)
- Representativeness
Representativeness is judging how probable something is by how much it resembles your mental picture of the category, or how much a sample resembles the process that produced it. Resemblance is fast…
- Research Misconduct
Research misconduct is the fabrication or falsification of data, or the serious manipulation of analysis and reporting, established by an institution rather than suspected by a reader. It is a far…
- Response Bias
A shift in how readily somebody says yes, as distinct from a change in how well they can tell two things apart. Becoming more sceptical of everything and becoming better at sorting true from false…
- Response Styles Theory
Response styles theory says that what decides how long a low mood lasts is not the mood itself but what you do once it arrives. Sit with it and turn it over, and it lasts longer. Do something…
- Retail Forecasting
Retail forecasting is the business of predicting how shopping will change. Checked against official statistics, the forecasts of the last decade were mostly right about direction and wrong about…
- Reverse Causation
The arrow may run the other way. A measure that predicts an outcome well may do so because the outcome, or something upstream of both, produces the measure.
- Reverse Dominance Hierarchy
Reverse dominance hierarchy is Christopher Boehm's account of forager egalitarianism: not the absence of a hierarchy but an inverted one, in which the rest of the group collectively dominates…
- Reverse Inference
Reverse inference is reading a mental state off brain activity: seeing a region light up and concluding that a particular feeling or thought is happening. Because most brain regions do many things…
- Reward Prediction Error
Reward prediction error is the gap between the reward expected and the reward received, and it is the signal that midbrain dopamine neurons actually carry. They do not fire to reward as such. Better…
- Rewards of the Self
Rewards of the self is the third of the three variable-reward types in Nir Eyal's hook model, set out in Hooked in 2014. The payoff is mastery, competency and completion sought for their own sake…
- Rewards of the Tribe
Rewards of the tribe is one of the three variable-reward types in Nir Eyal's hook model, from Hooked in 2014. The payoff is social validation and connectedness, the sense that your tribe rewards you…
- Risk of Bias
A formal judgment about whether a study's design could have produced its result by some route other than the thing being tested. Not an accusation, and not a verdict on the finding. A statement…