Making decisions under uncertainy #3

Posted on September 17, 2011


As noted in the previous post decision makers with very high confidence  (C) in their initial preference  (e.g,. political and religious extremists, people in love etc. ) ignore subsequent information or evidence at odds with that preference.

Ef = f[C, Ei + (1 – C)g(E1, E2, . . . , Ej . . . , En)

Confidence level


C = 0.9 to 1 True deductive believer—rejects (–) evidence
C = 0.7 to 0.8 Strong deductive believer—discounts (–) evidence
C = 0.4 to 0.6 Reasonable person who becomes overwhelmed by information overload and conflicting evidence; unproductive because of inability to settle on a given hypothesis (Ei), or evidence-sorting ritual (g), long enough to go to press
C = 0.2 to 0.3 Strong inductive believer—sophisticated (g’s)
C = 0.1 to 0 True inductive believer—dogmatic (g) rituals

Next we consider people at the other extreme, people who lack any confidence in (C = 0.1 to 0) or have no opinion regarding any initial preference but do have a trusted method, ritual or model  (g) for selecting what to endorse in the subsequent flow of information, opinion or evidence. We call such decision makers true technicians or disciples.

true technicians or disciples

At the other extreme from the true believer, the example  at the bottom of the table, we encounter the decision maker with little or no confidence (0 to 0.1) in the particular hypothesis. Therefore, the weight (1 – C) will flow to the particular cherry-picking or editing mechanisms they use (g) which, in turn, will strongly influence the final estimate (Ef). Such decision makers may be pure technologists, dependant on their particular data-processing hardware or software (Ej).  Some opinion pollsters are perhaps the best example of pure (g) data processors, they have no loyalty to any particular hypothesis or data – will sell their polling or  cherry-picking  services to the highest bidder.We don’t view them as scientists, but rather as technicians. These decision makers have a trusted rule (g) for cherry picking or selecting what to endorse (Ej) from the subsequent flow information – for time being or moment. As noted they may be an opinion pollster who relies on a polling ritual, or they may  be a disciple whose decision rule (g) is to blindly follow  a trusted or charismatic other (Ej), a  particular leader, boss, wife, husband, friend, group or party. In brief rather than investing their confidence (C) in an initial preference (Ei), they invest it in trusted others who tell them what to believe or decide, for the time being, and when to change it – a follow the leader decision-maker. Notice whether they rely on a particular ritual (opinion polling) or on a particular expert (leader, editor, etc.)  they have found a clear or  affordable way of managing or reducing uncertainty.

Next we consider the procrastinator or flip=flopper – the person in the middle of the table who divides their confidence 50-50 – half attached to a given decision and the other half assigned to cherry picking from the flow of subsequent information or evidence. They procrastinate, flip-flopping between their initial decision and their endless cherry-picking from the subsequent news – perhaps focusing on the last thing they hear. But notice if that option conflicts with their initial opinion they’re in a bind having half their confidence attached to the initial opinion and half to the conflicting latest news.

Posted in: Sciencing