Ludwig: Decision making Flashcards
1) What is decision-making?
Decision-making refers to any action taken from a number of options where the outcome is uncertain, relying on the imperfect information available.
1) How do reflexive decisions differ from rational decisions?
Rational decisions are carefully thought through, such as choosing which university to attend, whereas reflexive decisions are made instinctively, such as swerving to avoid a pedestrian.
1) What do behavioural economists study in risky choice paradigms?
Behavioural economists study how people compare options based on their relative rewards and risks, which provides insights into the information shaping choices and when people accept gambles.
1) How do mathematicians quantify the risks and rewards of options?
Mathematicians calculate the expected value (EV) of options by multiplying the probability of an outcome by the value of that outcome.
1) Why doesn’t expected value always explain real-life decisions?
A5: Expected value doesn’t always apply because people’s risk preferences and subjective evaluations of value, such as utility, also influence decisions.
1) What is the difference between risk and uncertainty?
Risk involves options with known outcome probabilities, whereas uncertainty involves unknown probabilities.
1) What is utility, and how does it differ from monetary value?
Utility is the subjective value people assign to an outcome, which varies between individuals. For example, £100 is more valuable to someone with limited wealth than to a billionaire like Elon Musk.
1)
What does a utility curve illustrate?
A utility curve shows how the subjective value (utility) of something levels off as additional units have less impact, such as earning extra money or achieving higher grades.
1) Why doesn’t expected utility theory always apply in real-life decisions?
Expected utility theory isn’t always upheld because real-life decisions can be influenced by factors like emotions, biases, or the impracticality of quantifying subjective values and risks.
1) How can expected utility (EU) help weigh decisions?
Expected utility combines the risk and subjective value of outcomes, allowing individuals to rationally compare options. For example, an option with a higher EU may outweigh one with a higher EV if the associated risk feels too great.
2) Why is expected utility (EU) theory considered both appealing and flawed?
EU theory is appealing because it is tidy, quantitative, testable, and provides a structured framework for decision-making under risk. However, it is flawed because it cannot accurately predict human choices, which are influenced by cognitive biases and violations like the common ratio and framing effects. This indicates the need for greater psychological realism in decision-making models.
2) How do risk preferences vary based on probability and whether the outcome is a loss or a gain?
Low probability gain: Risk-seeking
Low probability loss: Risk-averse
High probability gain: Risk-averse
High probability loss: Risk-seeking
These patterns reflect the differing psychological impacts of gains and losses under varying probabilities.
2) What is prospect theory, and how does it address the shortcomings of EU theory?
Prospect theory is a modified version of EU theory that incorporates psychological realism. Its main tenets are:
Reference Points: Prospects are evaluated relative to context-specific reference points, which explain framing effects.
Asymmetric Utility Functions: Gains and losses are evaluated differently, with losses being steeper and convex, reflecting why losses loom larger than gains.
Probability Distortion: Probabilities are not perceived accurately, with decision weights explaining why events like rare disasters seem overestimated while common risks are underestimated.
2) How does prospect theory explain the common ratio effect?
The common ratio effect is explained through probability distortion. For example, the psychological distance between 0.8 (likely) and 1 (certain) feels much larger than the distance between 0.2 and 0.25. This illustrates that people do not treat probabilities as objective values but rather distort them in decision-making.
2) How does prospect theory account for framing effects?
Framing effects are explained through the concept of reference points. The context or wording of a question influences whether outcomes are perceived as gains or losses relative to the reference point, thereby affecting decision-making.
2) How does cultural context influence risk preferences?
Risk preferences may not be universal, as evidence suggests that different cultures exhibit varying risk-seeking behaviors. This highlights the role of cultural factors in shaping attitudes toward risk.
3)- - Why do (or should) we think of perceptual decision tasks as a form of decision making?
because we make a choice about what we percieve based on the information we have. choosing from a number of options with uncertain consequences
3)- Describe the classic random dot motion paradigm. Why is this task suited to study perceptual decision making? What are the key behavioural benchmarks in this paradigm?
pp views multiple dots moving in different, often random directions CONTROLLED BY EXPERIMENTER. pp are asked to make a perceptual decision about the direction the dots are moving as a whole. accuracy is improved when the directional movement is more coherent ( most dots moving in a set direction), reaction time is reduced when there is more directional coherence. the coherence threshold for getting the correct answer lessens the longer the viewing distance.
3)- What makes the ( random dot motion) paradigm well suited for neuroscientific investigation?
because it measures eye movement and we understand the mechanisms/ action patterns that underpin eye movement
3)- Describe the evidence accumulation to threshold model (at an abstract level, without reference to neural implementation).
specific neurons for each direction, they become activated/ fire when they percieve motion in their set direction. a different system basically tallies the scores for each direction ( known as temporal integration because lets use big clever words hoh hoh hoh). evidence becomes clearer over time because its more able to gather information
3) - How is this model ( evidence accumulation to threshold) able to account for the behavioural benchmarks from the random dot motion paradigm?
the more sensory evidence you accumulate the more informed you are to make the correct decision, increasing accuracy ( explaining why higher viewing time requires a smaller coherence threshold) and why it takes longer to decide when the evidence is mixed and why the more coherent the motion, the easier it is to integrate and therefore the more accurate the choices
3) - For the classic random dot motion task, characterise the neural circuitry that implements this model
1) sensory input begins in area MT
2) from there, we have the sensorimotor area LT, which generates decisions about eye movements
3) which feeds into the basal ganglia which is thought to set the decision thresholds
4) superior colliliculus- motor unit centred on the action execution e.g. looking towards the target
3) - How does the model account for the speed-accuracy trade-off?
-You accumulate visual information over time
- There is an action threshold- you have to accumulate x amount of information in order to justify an action ( e.g. moving your eyes towards a target)
- the more coherent the visual information, the quicker you will reach threshold
- BUT if you want accuracy, then the threshold should be set higher as having a low threshold runs the risk of a false action due to noise
- SO you have to take time to generate MORE visual information in order to reach threshold
- In short, you can do it quickly or you can do it right
4) Describe the evidence that LIP neurons can encode expected value of actions (eye movements).
platt and glimcher (1999) recorded LIP neurons and found that neural firing rates were modulated by the probability that the movement encoded would be carried out and the magnitude of reward associated with the movement encoded by the neuron// meaning the neurons were sensitive to both probability and reward ( components of EV)
