Heuristics: Decision-making Shortcuts
What are you more afraid of: boarding a plane, or getting into a car? For many people, flying comes with nervousness or trepidation. Such concerns don’t reflect realistic concerns. Car crashes claim more lives each year by orders of magnitude. But they do reflect something else: a cognitive trap to which almost everyone is susceptible.
Making decisions can be difficult. To help us along, our minds use a number of cognitive shortcuts. These shortcuts, called heuristics, allow us to make more rapid decisions with minimal calculations. Unfortunately, while generally efficient, heuristics can also lead us astray.
Psychologists Daniel Kahneman and Amos Tversky first coined the availability heuristic. It describes a phenomenon in which someone confuses “easy to recall” with “likely to occur.” It’s easy to see how this could be helpful–when it works. However, it is also easy to see how this can fail.
The availability heuristic is based on how “available” an example is one’s memory–the ease with which something comes to mind. Sometimes the most easily recalled example is indeed the most likely on. For example, hoofbeats in the distance are more likely to be horses than zebras. But there are many other explanations that might also be at play. Movies and TV dramatize hijackings and conspiracies. They can consequently be easy to recall. But are they actually more common than mundane events like suicides or car crashes? No. The hijackings, conspiracies and plane crashes become sensationalist front-page news because they are so uncommon. Yet that public exposure tricks us into exaggerating how common such events really are.
Fighting Heuristics with Probability
Rootclaim’s methodology includes several components that help avoid the disadvantages of heuristics. Firstly, the “Starting Point” section asks how likely each possible hypothesis is initially, before looking at evidence specific to the case. Researching concrete data, like percentages of deaths or plane crashes, helps avoid skewing conclusions towards what’s easily remembered. In the Serbian Lottery analysis for example, lottery fraud instances over a 15 year period are tabulated and averaged. This calculation shows the general likelihood of lottery fraud.
Secondly, for each individual piece of evidence, the question asked is “how likely is this piece of evidence given each hypothesis?” The likelihood under each hypothesis is then assessed, again on the basis of relevant data. Examining each piece of evidence separately forces one to actually consider the particulars of the case, not just its generalities.
Instead of a person trying to balance multiple inputs and reverting to heuristics, Rootclaim’s algorithm then calculates the likelihoods.
Each part of this process (the starting point, the individual assessment of the evidence, and the final, probability-based calculation) reduces the opportunity for mental heuristics to torpedo the analysis.
Join Rootclaim here, and contribute to an existing analysis or suggest a new topic.