Key Terms Flashcards

Question

Why do we need rationality?

Answer 1

1. We need to be rational scientists 2. We need the assumption of rationality to study and understand people. Example: Neanderthals developed highly complex methods of shaping stones – In particular, they developed techniques to create many flakes that were sharp on one side and blunt on the other side, and had the right size to be held in one’s hand – Such flakes are ideal for cutting meat and cleaning animal skins ● Conclusion: – Neanderthals used these flakes to cut meat and clean animal skins ● This conclusion presupposes that Neanderthals were rational

Answer 2

Based on means/end rationality Acknowledges the problem that agents often have more than one goal. Core idea: we can set up a calculus of benefits and costs. (A Utility Function) Agents behave rationally when they maximize their expected utility

Answer 3

- Agents have perfect information about their goals and the probabilities of outcomes of their actions. – Agents have unlimited computing power. – In multi-agent contexts: agents have perfect information about the goals of other actors.

Answer 4

1: no room for irrational action 2: highly unrealistic ● Agents often have incomplete information, they face uncertainty, and they may not even know their own preferences very well.

Answer 5

Will tax breaks get Dutch people to install solar panels on their roofs? ● Define utility in terms of money. ● Calculate the relevant expected costs and benefits: – BENEFITS: having to play less for energy; selling energy; tax breaks – COSTS: purchase, installation, maintenance, annual depreciation ● Predict how big a tax break will make people purchase solar panels. BUT– Even in simple cases like the solar panels, there is uncertainty about electricity prices in the future; uncertainty about whether I’ll need my money for something else; a lack of time and energy to find out about the expected utility; uncertainty about innovation in solar panel technology; a vague sense that perhaps these panels lead to pollution; ideological reasons for or against solar panels; etc.

Answer 6

The study of rationality in situations of practical restrictions Gives us forms of choosing that are – Cheap and easy to implement – Give results that are good enough Mostly useful as an empirical research programme. ● We have to establish which heuristics people actually use in what circumstances. ● And then we can claim to understand why they made these decisions, and predict their future decisions.

Answer 7

Context-dependent rule of thumb for reasoning and decision making (not necessarily optimal)

Answer 8

In some cases, a very fast and fairly successful choice strategy is simply to choose what you recognise. ● Examples: – Which German city is larger, Hamburg or Augsburg? – Who won Wimbledon in 2003, Roger Federer of Mark Philippoussis? Not flawless, of course! – What is the capital of the state of New York, New York City or Albany? We use this all the time in the supermarket, unless we have good reasons to spend extra time.

Answer 9

In choosing between two options, base your choice on a single dimension of difference, ignoring everything else. ● Examples: – I care about climate change, so I choose the political party that does most against climate change, ignoring all else. – Offered three hotels in Benidorm, I decide to just check which is closer to the beach. – Seeing two paths in the woods, I choose the one where I see most mushrooms. ● Clearly suboptimal if you want to optimise, but makes for fast decision making. ● Can be extended to take in more than one dimension.

Answer 10

Some actual human decision making may be more rigorous than simple heuristics suggest. It’s also not always clear how agents select heuristics. When do I know that a particular heuristic will be good enough? And aren’t we actually optimising, except that our utility also involves the limited resources we have for decision making?

Answer 11

Western thinking has long made a sharp distinction between rationality and emotion – Emotion is antagonistic to rationality – The best decisions are made by un-emotional people (Often linked to sexist narratives: men as rational, women as emotional) ● Cognitivist views of emotions became popular (again) from the 1980s. Suggest that emotions have cognitive content and can be rational: – Being afraid of a tiger is rational, being afraid of a mouse is not. – Feeling disgust of rotting carcasses is rational, feeling disgust of tulips is not

Answer 12

– Motivational kick: emotions put us into motion, to not just have goals but also pursue them – Cognitive short cut: emotions enable us to quickly identify salient factors – Access to values ● It’s hard to do justice to human decision making without taking emotions into account.

Answer 13

Good for: – Reverse causal inference – reasoning from effects to causes – Reconstructing complex causal pathways – Generation hypotheses Weaknesses: – Danger of selection bias – Difficult to generalise findings

Answer 14

Good for: – Forward causal inference – reasoning from causes to effects – Testing generalisations about relationships between variables Weaknesses: – Poor basis for reconstructing complex causal pathways – Usually not a great basis for hypothesis generation

Answer 15

A qualitative method The intensive study of a single case with the purpose to shed light on a larger class of cases A researcher may choose cases with various characteristics: – Typical case: highly representative example – Extreme case: may reveal causal relations most strongly – Exceptional case: may reveal causes of accidents – Deviant case: anomaly, may help pinpoint limitations of a theory

Answer 16

– Degree to which a study’s findings are warranted for the case studied

Answer 17

– Degree to which a study’s conclusions are generalisable

Answer 18

Measures taken to enhance the warrant of findings (internal validity) tend to limit their generalisability (external validity) ● For instance, looking into a lot of detail – And vice versa

Answer 19

A relation of similarity or isomorphism between two domains of phenomena By extension: the cognitive process of transferring conceptual resources from one domain to another

Answer 20

- Discover similarities and identities between domains - Generalise over and unify distinct domains - Create and extend scientific concepts

Answer 21

source domain - Domain of phenomena that supplies the terms of the analogy: typically more familiar to us target domain - Domain of phenomena to which we apply the analogy: typically less familiar to us mapping - Set of correspondences between terms in the source and terms on the target domain relations - Claims that, under the mapping, hold for both the source and the target domain

Answer 22

Organic Analogy “Society is analogous to an organism” Classical social science understood societies by analogy with organisms – Emile Durkheim (1858 – 1917), French sociologist – Functionalism: interpreted society in terms of the functions of its constituent elements Mapping from domain of organisms to domain of societies – Living organism -> Society – Parts, e.g. organs -> Components, e.g. institutions – Functions of organs -> Functions of institutions – Ill health -> Loss of social equilibrium Claims that hold in both domains: – The whole [organism / society] is greater than the sum of its parts [organs / institutions] – Parts [organs / institutions] work together for the benefit of the whole [organism / society] – To understand a part [organ / institution] is to show how to relate it to other parts and what role it plays in the continued existence of the whole [organism / society]

Answer 23

● Positive analogy – We know already that these relations hold in both the source and the target domains ● Negative analogy – We know already that these relations hold in one domain but not in the other ● Neutral analogy – We know that these relations hold in the source domain, but we do not know if they also hold in the target domain