Decision making - summary of chapter 10 of Cognitive Psychology by Gilhooly, K & Lyddy, F, M

Cognitive Psychology
Chapter 10
Decision making

Decision making is the cognitive process of choosing between alternative possible actions.

Normative approaches: attempt to establish ideal ways of deciding that will give the best decision possible.
Descriptive approaches: aim toe describe how decisions are actually taken as against who they should be made.

Decision problems differ in a number of ways:

• Risk: there is a probability that one of the options could lead to negative outcomes for the decision maker.
• Riskless: choices where the outcomes of the choices are known with certainty.
• Object variability
• Single-attribute: decision problem involve alternatives that vary in only one dimension.
• Multi-attitude: decision task in which the alternatives vary in many dimensions or aspects.

Expected value: the long-term average value of a repeated decision which is determined by the probability and size of the outcome.

Expected value theory

People should act to maximize the expected value of choices.

The expected value of a risky choice is the average result you would get if you repeated the actions many times over.

The expected value approach is an optimal way to deal with risky decisions in which we can put a money value on the possible outcomes and can say exactly what the probabilities of the possible outcomes are.

But the expected value approach does not fit people’s behavior in real life.

Risk aversion: avoiding risky choices even when a higher expected value than riskless alternatives.
Risk seeking: a preference for risky choices even when riskless alternatives of higher value are available.

Utility: the subjective value of an option
Subjective probability: how likely a person believes an outcome to be irrespective of the objective probability.

Utility and prospect theory

Prospect theory
Prospect theory: a decision theory stressing relative gains and losses.
Deals with how people choose among gambles and importantly extended the utility plot into the area of losses.
Decisions about monetary gambles are about gains and losses relative to one’s current wealth.
Losses of any kind are weighted disproportionately to gains of the same amount.
Loss aversion: there is a greater dislike of losing utility than liking for gaining the same degree of utility.
Endowment effect: a tendency to over-value a possessed object and to require more money to sell it than to buy it in the first place.
Status quo bias: a tendency to prefer the current state of affairs.

Subjective probability and prospect theory

Peoples perception of probabilities systematically depart form objective values.
Objective values are transformed into subjective probabilities → decision weights.
People tend to overweight small probabilities and underweight large probabilities.

Framing and prospect theory

Framing effects arise when irrelevant features of a situation affect the decisions that are made.
Ways in which the alternatives are presented (or framed) in a decision problem, in terms of gains or losses, will strongly influence the choices made.

Invariance: the principle that choices between alternatives should not be affected by how the options are described.

Positive frame → risk aversion
Negative frame → risk seeking

Making probability judgments

Two major heuristics:

• Availability
  Judging frequency or probability of events by how easy it is to bring the events to mind.
  Extension rule: if the extension of a set, A, includes the extension of a set, B, then the probability of A must be greater than or equal to the probability of B.
  But people don’t bet so in real live.
• Representativeness
  Judging frequency or probability of an event or object by tho representative or typical it is of its category.
  An assessment of the degree to which an example fits our idea of a typical member of the category.
  Representative or typical instances tend to be judged more likely to occur than unrepresentative instances. This can lead to error.

Conjunction fallacy: the mistaken belief that the conjunction of two events (A and B) is more likely than either one A or B.

Base rates

Base rate of an event is the overall probability of the event in a population.
There is a tendency to ignore base rates.

Posing problems in terms of frequency may evoke quite concrete representations of the situations described and hence provide guidance for interferences and checks on erroneous inferences.

The affect heuristic

Affect heuristic: involves substituting feelings (positive or negative) for target attributes in decision problems.
A common way of quickly making decisions.
People tend to form emotional responses to the initial component to which they were exposed and the emotional responses affect the second presented component.
An overall affective response to each item to guide assessment of risks ad benefits such tat positive affective response leads to a low assessment of risk and high assessment of benefits and the opposite for negative affective responses.

Somatic markers: emotional responses

Inability to use affective cues or somatic markers in decision making have a negative effect on quality of decisions.

Decision processes for multi-attibrute alternatives

Multi-attribute utility theory

MAUT: Mulit-attrubute utility theory
Suggest that the decision maker should:

• Identify the relevant dimensions or attributes
• Decide on weights to be assigned to the attributes
• Obtain a total utility for each object by summing the weighted attribute values
• Choose the object with the highest weighted total

Problems with MAUT:

• The decision maker can not be certain what the relevant dimensions are
• The attribute weightings and scorings may suffer from unreliability over time.

Elimination by aspects

EBA
The chooser would first select an attribute and eliminate all options that did not meet some criterion level on that attribute. All the choices who do not meet the criterion are ruled out.
Then a second criterion may be selected and so one.
If the chooser continues in this way to eliminate alternatives, sooner or later, only one option will e left and so the decision will effectively be made.

Problem:

• Very different choices can arise depending on the order in which aspects are used to eliminate alternatives

Satisficing

People are generally content to set a minimum acceptable level which will satisfy them but be short of the maximum.

Testing mulit-attribute choice models

People use EBA as well as MAUT as well as satisficing.
Time pressure can shift preferred decision strategies.

When choices are difficult, people adapt a more demanding strategy to help ensure a good decision. With easier tasks less effortful strategies are seen t o suffice and tend to be used.

Studies indicate that no single decision strategy is always used in choosing between multi-attribute alternatives.
It seems strategies are adopted that tend to compromise between minimizing cognitive load and maximizing the utility of the outcome chosen.
Generally, cognitive load in decision making could be minimized by simply choosing at random but the results would tend to be very poor.
Quality of decision making would be maximized by assessing all alternatives on all relevant dimensions, but the information processing required would be very demanding.

Two-system approaches to decision making

The differences between the intuitive and more reflective forms of decision making have been highlighted recently in two-system approaches of thinking and decision making.
Two-system view is that there are two modes of thought, system 1 and system 2.

• System 1: rapid intuitive thinking. Automatic, fast effortless and emotional.
• System 2: slow deliberate thinking. Permits abstract reasoning

Fast and frugal heuristics: the adaptive toolbox

Fast and frugal heuristics are simple and fast to execute as they require little effort. Involve system 1. Together the heuristics form an adaptive toolbox, as the heuristics are generally valid for the real life situations in which they have developed.

• Recognition heuristic
  Recognizably is correlated with the criterion attribute for the groups of people being tested. Too much knowledge of the domain can make the heuristic unworkable.

Simple decision trees
Only a few key attributes are considered in a priority order so that if the first attribute has the critical value a decision is made otherwise the next most important value is considered.

Heuristics and consequentialism

Consequentialism: assuming that decisions are made on the basis of the consequences expected to follow each of the choices available.

Examples for fast-and-frugal heuristics

Omission bias

People tend to state that they would feel more responsible for deaths caused by their action than by their omission.
People have a tendency to judge consequences less bad when the consequences are due to omissions rather than commissions.

Punishment

Many people feel that retribution is the main function of punishment and are little swayed by the consequences of punishment.

Resistance to coerced reform

Although people will often endorse certain reforms desirable consequences for society or even the planet, the same people will often say that they would not vote for the consequentially desirable reforms.

Naturalistic decision making

In natural situations the decision maker may not be explicitly presented with options to decide among but rather has to generate one ore more possible actions.

Critical incident analyses: gaining information about naturalistic decision making by analyzing detailed recalls of recent important decisions.

Recognition primed decision: expert knowledge based decision making in which cues in the situation are recognized as indicating particular actions.
Most common.
Typically, in many critical situations only a single actions was mentally generated and it was then chosen to be executed.

Naturalistic decision making and important real-life choices.

Overall, it seems that many real-life decisions by experts do not actually involve conscious decision making between alternatives.
Time-critical situations: take-the-first-option
Immediate responses required: fast-and-frugal
Important and time pressure is low: effort-full MAUT.

Neuroeconomics: neuroscience approaches to decision making

Two key ideas from studies of human decision making:

• The notions of utility
• The role of dual systems in reaching decisions

Neuroeconomics: the study op neural processes underlying economic decisions.

When alternatives differ widely, it seems natural to suppose that decisions between such alternatives must be made by assessing the alternatives on a common scale of subjective value or utility.
Decisions would favor options that activate reward systems most strongly.

Recordings from dopamine neurons in the orbitofrontal cortex, and posterior cingulate cortex have shown neural responses that relate directly with reward size in primates and similar results have also been reported in human studies.
Activity in dopamine neurons is linked to reward size and so much activity is linked to choices as choices follow reward.

Short-run impatience is driven by the limbic system which reflects System 1 activity and responds impulsively to immediate rewards.
Choices of the delayed rewards are governed by the lateral prefrontal cortex, which reflects System 2 activity.

The aging brain and financial decision making.

Good financial choices require: an understanding of how financial systems work and the mental acuity to find and choose the best option.
Experience brings improvement but after a point, the accumulation of experience starts to get overwhelmed by decline of cognitive function.

Affective processes are also involved in decision making and it has been found that older people generally feel more optimistic than young people do, and are more likely to focus on the potential upsides of a situation.
This tendency changes the decisions older people make.

When expecting a loss, younger adults report being more upset and showed higher blood flow in the insula, a part of the brain implicated in negative emotions.
As the loss increases, so did negative feelings and insula activation.
The older adults did not feel as bad as younger adults did and showed less activation in the insula.

Whereas looking on the bright side is emotionally beneficial, it has drawbacks in financial decision making, when it is important to consider possible losses.

The psychology of financial decision making and economic crises

When it came to making decisions under uncertainty, people tend to be more influenced by perceived risk than by objective risk.