Why are people's judgements outperformed by linear models, including models of their own judgement?

Why are people's judgements outperformed by linear models, including models of their own judgement?

There are several reason for this:

• People can be inconsistent
• they (people) can mistakenly rely on irrelevant information or look at wrong variables as well do addition or multplication errors
• when relying on relevant information, they can weigh the information in the wrong way
• when presented with more information, people can easily identify single cases as exeptions to the rule
• people who are in close contact with particular domains may see a skewed sample of events
• people can underestimate situational factors on other people's behavior- fundamental attribution error
• people may obtain incorrect relationship between cues and criterion
• people are influenced by their past personal experiences as well as fatigue or boredom
• people can misunderstand the task
• people can "follow their feelings" based on the idea, that every case is unique and statistics do not apply
• people have limited information processing capacity
• linear models and models of people's own judgments are consistent and always follow the same rule or a formulae

Research has shown, that people may actually make judgements using fast and frugal heuristics, meaning they often base their judgments on a single cue ( a cue found from memory that enables a judgement to be made)

  It is strange to think about what actually happens in our mids when we make decisions or what we base our judgments on. To most people, like me, we know we have to make a decision and we will do it based on many things like feelings, instinct and considering various variables (for example would the outcome of the decision make us happy?), but we do not really think about what exactly happens in our brains when we do that. Do we? Or maybe it's just me...

Prospect theory

  Prospect theory... ( a small summary)

...consists of two phases:

1. Editing stage. At this stage the decision maker simplifies the problem for the purpose of evaluation and choice. It has 4 sub-stages: coding- based on gains and losses, combination- associated problems, probabilities and outcomes, segregation- separate risky components from non-risky ones, and cancellation- dismissing common values.
2. Evaluation stage has two functions: value function- first gains have a really high value, but the subsequent gains which are also good have less and less value. For example a first drink- a bottle of cold water after half a day at the beach seems priceless, whereas the subsequent drinks are still good, but not the same value.The same works for losses, meaning that the first loss has the highest value- this reminds me of heart-break, the first is always the worst, the subsequent ones are always less and less painful. To compare losses and gains: losses are worse than the gains of the same value; weighting function- the likelihood of things occurring. Sensitivity to changes is probability. Small probabilities have an undue value. For example according to this, it is not very wise to play the national lottery and pay a pound for that, because the £1 ticket's value is actually much less than £1.

correctly plotted graphs...

   These are the correctly plotted graphs from last week:

 

The graphs are not as similar as first thought. It seems, that in the certainty equivalence task i am more risk-taking than in the probability equivalence task. I do not really know how to explain this, but there is something about percentages, that i feel i need a really high probablity percentage of winning a great deal in order to be willing to take the risk of losing the amount i would have safely gotten if i would not have taken the lottery. With monetary amounts it is not the same: somehow by looking at the amounts i feel more willing to take the risk to win a great deal rather than less, because i have nothing at the moment anyway. It is strange though, that £ or %  marks make a difference in strategy choice.

certainty equivalence vs. probability equivalence

   The following graphs show the assessment of my utility function using certainty- and probability equivalence exercises.

    As we can see, both of the graphs look roughly the same and show that as the monetary value increases, utility value does too. The similarity of the graphs indicate that a same strategy is used for both exercises, whether it's deciding on monetary value or the utility value- assessing me as risk aversive peron.