Sunday, May 11, 2014

Logic in Your Plotline - Investigatory Thought Choices with Richard McEachin






Fiona -
Today our guest is Richard McEachin

Richard B. McEachin has been an investigator for 40 years. Since 1993, his firm, McEachin & Associates Ltd., has provided research, training, and consulting to private investigators, law firms, due diligence analysts, journalists, public relations professionals, and government. Since 2006, the author has published The Confidential Resource , which is one of the top 10 Private Investigation blogs . He has also appeared in The Toronto Star, the Financial Post, Profit Magazine, and appeared on Canada's most-watched current affairs and documentary program, W-5. He is the author of Sources & Methods for Investigative Internet Research

Richard, you've been consuming a lot of chicken soup this week to help you with your illness. Wouldn't it be nice if we had such a time-tested remedy for investigations that are suffering?

Richard  - 
That's for sure

Fiona

What kinds of things do you see going wrong that would thwart an investigation from a successful conclusion?

Richard -
Most of the problems I see today relate to a lack of understanding of the sources & methods being employed and not knowing all the sources & methods available to the investigator.

Fiona - 
So today we are talking about crime investigation - sometimes the trail goes cold. What are some options an investigator has at that point?

Richard -
Well, fixing an investigation is about the details, logic, and the human factors that caused it to go off the rails.

Fiona -
Let's start there. What kinds of issues might cause a problem to begin with? These will be very interesting to writers because they act as plot twist material.

Richard -
Let's look at some of the human factors. Detectives are people too. All the things that influence a factory worker or witness also have an influence, to a larger or smaller degree, on the Detective. For example, bias. Bias isn't an indication of any evil in this context. If something in the evidence paints a picture of who the culprit might be, then the detective may start believing this and start molding the investigation around that premise.

Fiona - 
Can investigators train to avoid bias?

Richard - 
Yes, just like LEO's (Law Enforcement Officers) can train not to overreact to verbal taunts.

Fiona -
Bias then is a big issue

Richard -
Bias in this context is most often 'pattern matching' or contextual bias. Something in the evidence starts to point you in a certain direction, and you get tunnel vision and don't recognize other possible avenues of investigation or other possible offenders. 
SVG files
 (Photo credit: Wikipedia)


Contextual bias appears in the 2009 National Academies of Science report on forensic science. In testing fingerprint examiners, they found that supplying the examiner with too much contextual information influenced his identification of the person associated with a sample fingerprint.

The report is here: http://www.nap.edu/openbook.php?record_id=12589&page=1

Strengthening Forensic Science in the United States: A Path Forward
www.nap.edu

Fiona - 
Excellent. Okay, bias. And what else?

Richard - 
The other two things I see often are failures in logic and failures to understand the strengths and weaknesses of the underlying science. I also see a lot of failures to understand the strengths and weaknesses of data sources, but I think that is related to both of the above.

Fiona - 
Great. So our team has dug themselves into a hole of poor logic, bias, and misunderstanding of forensic sciences. How can we get them out?

Richard -
Once you are in a hole, stop digging. I start by
sifting thru the pile of dirt instead.
Question mark
 (Photo credit: Wikipedia)
I look at how the team presented their case. I look at all the reports and supporting documentation. How they present this material is very telling. It will tell me a lot about their skill level, particularly about handling the details that have a large impact. The presentation of the 'product' of the investigation has a large impact on how the case proceeds.

This material will usually tell me if they used top-down or bottom-up logic. Top-down (deductive) logic is more prone to produce a problem as its conclusion is a certainty. This is not a small consideration when someone's future is at stake.

Fiona - 
I think this is a very interesting point. Can you give me an example that you can take us through using top-down logic and then take
us through again using bottom up logic?

Richard -
Deductive logic requires that every premise is always true or the conclusion is wrong. The conclusion may be logical but wrong because one premise in the argument was not correct throughout the closed domain of the argument.

Fiona - 
So let's make something up - you have a murder scene, at least the inspector assumes it is a murder scene because everyone hated the victim and everyone threatened to kill the victim. When they found the guy's body, it was badly decomposed. But the investigators are running on the assumption that someone must have killed the horrible guy - something like that?

Richard -
Yes. essentially, you have a dead guy. He was hated. Therefore, he was murdered.

Fiona - 

Quick deductive reasoning tutorial -

1. The investigator creates a hypothesis
2. The investigator looks for data that would support his hypothesis
3. When the investigators gathers enough supporting data, they will draw a conclusion.


Quick Inductive reasoning tutorial

1. Start with a small observation or question
2. Works towards a theory by looking at related data/issues (more exploratory than deductive)


Video Quick Study (1:32) Inductive v. deductive reasoning in a nutshell.
Video Quick Study (1:16) Sherlock Holmes deducing correctly enough that he gets a glass of wine thrown in
                               his face.
Video Quick Study (4:26) Monty Python uses deductive reasoning to discover if a woman is a witch or not

Richard - 
Amazon Link
Bottom-up is Inductive logic. The conclusion of the argument is probable based upon evidence that is not certain and cannot form a premise that is always true.

Fiona - 
But on the inductive it would be - I have a dead guy I believe he was murdered and I will now search for the clues to prove that I'm right?

Richard - 
Yup. You got it.

Fiona - 
Okay good - and you prefer the second. The inductive reasoning? If this is correct can you tell me why?

Richard - 
Yes. The closed domain required by deductive logic is very hard to maintain. An example would be GSR testing. Your suspect might have fired the fatal shot. His clothes and hands test positive for GSR. You got the guy--wrong. Many things that are not GSR will test positive. In 2006 the FBI stopped testing for it, in part because all there labs were found to be contaminated by GSR and GSR-like material

Fiona - 
The labs were covered in gunshot residue, and it was contaminating the evidence? Wow. I'm thinking that probably there's GSR on all of my clothes... well, not my bathing suit. Can you wash it off? Suddenly I feel contaminated. Maybe even a little OCD. Also, what might give a false positive?

Richard - 
You can wash it off. Soap & water for the OCD in you. An auto mechanic would probably test positive but someone who fired a .22 rimfire might not as much .22 RF does not test positive in the chemical test.

Fiona - 
Things like that make it very difficult for investigators but help put kinks and twists into a story line. Thank you so much for coming on today. A ThrillWriting traditional question - Can you tell us about your favorite scar?

Richard - 
Scar: nobody notices the good work but they sure will bitch about the bad work

Fiona - 
LOL You are very cryptic. Richard, my great thanks to you for coming in and helping us out today. 


You can catch up with Richard at: his website and blog



My best wishes to you. 

See this article in action in my novella: MINE


Thank you so much for stopping by. And thank you for your support. When you buy my books, you make it possible for me to continue to bring you helpful articles and keep ThrillWriting free and accessible to all.



4 comments:

  1. Thanks, as always, Fiona. Helpful in putting together my current book.

    ReplyDelete
  2. Thanks, Dennis.
    I found this one particularly helpful to my project, as well.
    Happy plotting.

    Cheers,
    Fiona

    ReplyDelete
  3. Thanks Dennis for the advice. This really makes thing clear in light of a mystery novel I am currently reading.
    Prescott fry
    Twitter: @kingfry1

    ReplyDelete
  4. Thanks Dennis for the advice. This really makes sense in light of a mystery novel I am currently reading.
    fry
    Twiitter: @kingfry1

    ReplyDelete