Rossmo's Probable Possible
In case you missed it probability and problem solving are all the rage in lit lately. Nassim Taleb's Fooled by Randomness and Black Swan examine speculation, luck and skill in the world of financial markets. Len Fisher's Rock, Paper Scissors looks at game theory (particularly the Nash Equilibrium or Prisoner's Dilemma) and how this can make us better negotiators. In Outliers Malcolm Gladwell reexamines a common statistical phenomenon in which rare occurrences that are interesting, but not appropriately significant are often cast off from making logical assumptions.
I have not read Outliers, am half-through Rock, Paper Scissors, and quickly abandoned Fooled By Randomness. Taleb appears to be under the impression that the way his mind works is significantly unique to justify a meandering writing style. Maybe this stuff works at cocktail parties, but it's dead-dull in print. Narrative contrivances such as the yin and yang "Nero Tulip" and" John The High Yield Trader" don't help. I could read Goofus and Gallant and gain this all-too-familiar homespun insight.
Still I am a sucker for writing on probability theory, which is why I am so thoroughly enjoying Kim Rossmo's Criminal Investigative Failures. You might assume that since I contributed a chapter to the book that I have already read it. Not so. Though I had advanced access to the text online, I preferred to wait and read it in print (you know, like a real book? Remember those?).
I'm about 90 pages in and it's great, and infinitely applicable to problem solving in everyday life. Observing the process by which humans engage in decision making through the prism of police investigations only enriches the experience, allowing you to better extrapolate the process to your own, common experience.
And Rossmo takes on some pretty big fish, including Special Agent for the FBI, John Douglas and attorney Alan Derschowitz, exposing how even so-called experts are prone to bias and errors of probability. In the case of John Douglas, the experience is particularly satisfying as I now consider him in his writing to be pretty much an investigative fanboy and self-aggrandizing blowhard. Rossmo's conclusion to the book (yes, I skipped ahead and read all the Rossmo bits in sequence) presents a paragraph retelling of the infamous Duke Lacrosse Scandal; as eloquent and concise a rendering of the pitfalls of subjective reasoning as I have ever read.
The book is introduced with two chapters by Rossmo previously published in the FBI's Crime Bulletin. The first chapter focuses on cognitive biases such as errors in perception and intuition, while chapter 2 concentrates on errors of statistical probability, or rather how statistics can be misused, misinterpreted and abused in the hands of laymen. What follows are a number of chapters of case studies and researched theories by other authors which illustrate the problems of bias and human error in the criminal investigative process. Throughout, Rossmo steadily makes his case that heuristics -the cognitive shorthand we engage in to rationalize the world - is a double edged sword; we desperately need it to interpret reality, but it can be our worst enemy in working toward a rational decision making process.
I suspect some chapters were included to play devil's advocate to Rossmo's thesis and promote discussion, as is the case with Snook and Cullen's Bounded Rationality and Criminal Investigations: Has Tunnel Vision Been Wrongly Convicted? Snook and Cullen argue that the technique of tunnel visioning to draw conclusions has been severely maligned in recent years, when in fact it is a natural tendency of focused determination, a vital shortcut process to making rapid decisions. Police forces do not have the time or resources to engage in a pure rational decision making process, and instead must rely on cognitive tools such as tunnel visioning to make decisions and conclusions.
Snook and Cullen's argument relies on a key erroneous assumption. Their theory is grounded in the findings of Canada's Federal-Provincial-Territorial Work Group's (FPT) 2004 Report On The Prevention Of Miscarriages Of Justice, in which the FTP made a series of three policy recommendations, including that "police officers should avoid tunnel vision". In fact the third recommendation concerning tunnel vision was largely directed at Crown prosecutors, not law enforcement. Indeed the entire report was written largely by legal professionals for the benefit of the legal profession, and is largely concerned with the wrongful conviction of innocent victims. In this context, the use of tunnel vision is not a cognitive tool, but a calculated legal strategy in which evidence that supports a case is disproportionately weighted by a legal prosecuting team, while evidence that would exonerate the accused is suppressed. Though no less egregious, this is not bias, but a deliberate tactic taken by the legal system in a win-at-all-cost strategy of criminal prosecution. If the aim is to change the process of tunnel visioning in the judicial-legal system, then the place to start is to stop misidentifying it as a flaw in the criminal investigative process. Better to show lawyers the costs to the justice system of wrongful conviction in order to promote the reform of harmful legal processes. No one would argue that tunnel-visioning is misused technique when it is practiced at the exclusion of other rational decision making methods. This is sound sense not just for investigators and lawyers, but for me when I next decide to weigh the benefits of purchasing a Carolina Hurricanes season ticket package.
Despite this problem, I very much enjoyed Snook and Cullen's chapter because it got me thinking about the human decision making process. Communication, motivation, organizational design and structure, judgement and decision making... these are foundations of management in any organization, and they should be studied and pondered seriously. And on one thing I can agree with the authors: On the basis of Ecological Rationality we can assume Mark Messier is the higher point scorer, not just because he endorses Lays Potato Chips, but because Madge would never sleep with Eric Cairns.
In the weeks to come I hope to post some more thoughts on this book. At $60 it is an expensive investment, but I urge everyone to read it, maybe start by convincing your local library to purchase a copy.
posted by John Allore @ 1/02/2009 01:03:00 PM
0 comments
0 Comments:
Post a Comment
<< Home