Editing (or framing) risky prospects

If we say that a terrorist action like bombing or armed assault is a risky prospect, by this we mean that the action will result in an outcome, w, that occurs with some probability, pr On the basis of the possible outcomes and their probabilities the decision-maker can form an expectation of what might happen but there is always a chance that the actual outcome will be different. The higher this chance of divergence between expected and actual outcomes, the riskier the prospect. For example, a bombing might be expected to attract some amount of attention in the print media, say 100 column inches.10 The actual amount, however, might be anywhere from 85 to 115 column inches. This makes bombing a risky prospect. Consider another attack method. Armed assault might be expected to attract 150 column inches of coverage. However, this attack method might be riskier. The actual amount of coverage could be anywhere between 50 and 250 column inches. The terrorist group would have to decide whether it wants to take a chance on the riskier attack method that has a higher expected outcome or whether it wants to play it safer, with narrower limits on both the up and downsides.

The terrorist group’s editing process will have certain implications for the way the group comes to view the alternatives and, ultimately, implications for the particular action or attack method that will be chosen. Kahneman & Tversky (1979, p.274) break the editing phase into four distinct operations:

  • 1. Coding outcomes as gains or losses. Imagine that the terrorist group is interested, among other things, in the publicity that its action will receive and that a particular action is expected to attract 100 column inches of coverage. In orthodox economic theory, this would automatically be treated as a ‘gain’. But what if a rival group had recently been accorded far more coverage? If this has become the terrorist group’s reference point in assessing its possible future actions, 100 column inches may actually be perceived as a ‘loss’. Whether outcomes are coded as gains or losses depends on the terrorist group’s reference point. Prospect theory allows for positively valued outcomes to be treated as losses if they are less than the reference point.
  • 2. Simplification by combination. Suppose that the terrorist group is aware of a correlation between fatalities inflicted and media attention such that 10 fatalities is associated with 100 column inches of coverage. Now suppose that bombing has a 20 percent chance of inflicting 10 fatalities and a 20 percent chance of generating 100 column inches of coverage from fatality counts other than 10. The terrorist group will perceive this as a 40 percent chance of receiving 100 column inches of coverage.
  • 3. Separating the risky and risk-free parts of the prospect. Kahneman andTversky call this segregation. For example, armed assault may be expected to inflict 20 fatalities with a likelihood of 0.40 and 5 fatalities with a likelihood of 0.60.This can be broken into two parts: (1) 5 fatalities for sure and (2) 15 fatalities with a likelihood of 0.40.
  • 4. Cancelling out common elements.When two different prospects share common features, the decision-maker cancels them out. For example, if there is a multistage process involved in each prospect and the first stage is common to both, it is ignored. If an additional payoff of the same magnitude is added to both prospects, it is also ignored. The terrorist group might see the initial stages of a campaign as being shared similarly by each of the possible attack methods that are available to it. This common feature of each attack method will therefore be ignored when the terrorist group is assessing and ranking the alternative attack methods. In a period of deteriorating security, for example, each terrorist attack method may inflict 10 more fatalities than before. This will be ignored in assessing the alternative attack methods because it is common to all of them.

Usually, editing results in a simpler representation of the choice problem. In fact, this is one of the key features of human decision-making. People tend to turn difficult problems into simpler representations. Recognising that there might be such a thing as an editing phase helps us to avoid the conclusion that a terrorist group does not have the capacity to think about and weigh up payoffs and probabilities for all of its potential actions before choosing what it determines to be the best one. In fact, it is the very complexity of the problem that makes editing necessary and editing is the very thing that introduces certain systematic errors into the decision-making process. We cannot simply sweep away concepts such as expected payoffs and probabilities because we think that terrorist groups do not or could not ‘think like that’ because it’s too ‘hard’. On the contrary, prospect theory leads us to expect that they will think in a certain way because the decision-making task is ‘hard’.

The important question is, how simple can simple get? Research into framing or editing has led to the exploration of alternative frameworks, one of which is called fuzzy trace theory (Reyna & Brainerd 1991,1995). According to fuzzy trace theory, simple can get very simple. In this theory, it is only the ‘gist’ of the situation that matters. Consider a terrorist group editing the possible outcomes of three alternative attack methods. The editing might be as simple as the terrorist group concluding that two attack methods have some chance of‘some fatalities’ while the third attack method has some chance of‘zero fatalities’. The ‘some’ will be favoured over the ‘none’ (Kiihberger & Tanner 2010, pp.317-318) and the third attack method will be edited out of further consideration. While more research is required, the answer to our question is that editing might boil down different prospects to very simple terms. Having so simplified the range of alternatives that will be considered, the decision-maker then moves on to evaluate the remaining options.

 
Source
< Prev   CONTENTS   Source   Next >