The Differential Diagnosis

....aaaaand we're back folks!

My sincerest apologies for being gone for so long. School got a little hectic and this got away from me. Now that summer's officially begun, I thought I'd begin anew with a fresh perspective on the science/art I've been discussing thus far.

One of my favorite shows of late has been House M.D.. For those of you unaware, Dr. Gregory House is the medical version of Sherlock Holmes. He uses the rules of logic in combination with his astute observations to diagnose patients who have other doctors stumped. This has led me to another interesting way in which to view deductive exercises.

For the foreseeable future, I'll be attempting to cast a medical light on my deductions. The reason for this is incredibly simple - the act of deducing someone's personality, their recent whereabouts, their career, etc. can all be boiled down to a differential diagnosis.

Here's an example to illustrate my point:

Elimination:
Let's say you see a man in a suit walking into a hotel. Our job is to deduce his career. If we are viewing this exercise as a differential diagnosis, it may first be helpful to get an idea of what he is not, thereby excluding several incorrect guesses before they're made. "Construction worker," for example, while still potentially possible, is definitely not high on the list. We can begin to set aside careers like school principal and restaurant manager  - jobs which do not require much movement outside their place of business. Keep in mind that we do not yet have enough information to completely exclude those options, although we can label them as "less-than-likely."

Symptoms:
Now we need to build a list of other observations (symptoms). Closer inspection reveals that the man is wearing cufflinks. This is sufficient to eliminate most if not all of the employees at the hotel - a uniform would not include cufflinks. This means that if the man works for the hotel, he is probably upper management or some other salary position. Furthermore, the man is wearing a bluetooth device suggesting that he is frequently on the phone. Finally, you notice that the briefcase he is carrying is a bit deeper than average. While you're not sure what to do with information, you believe it is a clue to determining the contents.  

Patient History:
It is at this point we can borrow from the medical line of thinking and get a patient history - did the man come from his car? Which door did he enter? Was he greeted by anyone? If so, how? If the man's car is out on the hotel driveway, he's probably a guest using the valet (upper management would have parked in a more permanent location). If the man was greeted by the doorman with a cheerful, "Welcome, first time guest!" Then we have our answer right there.

Tests:
Let's assume that the doorman ignored the man entirely, and his car is nowhere to be seen. He could still be a guest at the hotel, which would leave quite a bit of wiggle-room in our diagnosis. Let's run a test. One of the most important things to remember about tests (medically and deductively) is that you should first form a hypothesis for testing - you need to have an idea of what you want to find out. In this case, we want to determine whether the man works at the hotel. An easy way to do that is ask him a question a guest would probably not know the answer to. (Ex: "Excuse me, how many rooms does this hotel have?") If the man answers immediately, we can infer that he works at the hotel, and is responsible for running the hotel, or perhaps in a non-customer service department (accounting).

Time:
Let's assume that our test was negative - the man has no idea how many rooms the hotel has. This means that it's very likely he's a guest. It is at this point we can use one of the most valuable tools to both deductionists and doctors alike - time. In many cases, the best thing to do is wait for new observations (symptoms) to present themselves. Let's say that after a few seconds, another man walks into the hotel and you observe a meeting between your subject and this new man. They greet each other with a handshake and our subject tells his associate his room number and hands him a business card. Now we can be sure that he's a guest. Furthermore, you can begin to form the strong hypothesis that he's here on business. Now all that's left to do is determine what business. Finally, as the men sit down at a nearby table, you see the subject reach into his briefcase and hand the other man one of many small rectangular cardboard boxes inside. The man turns it over, and you see the label - it's some sort of medication.

The Whiteboard:
There's nothing wrong with writing your ideas down. Even if you choose not to, you should be trying to maintain two separate lists.t  One list contains the information you've gathered (symptoms, strange behaviors, environmental information which might be relevant). The other list is your "top 5." This is a constantly changing list of possible explanations for the observations you've made. In a medical scenario, this would be a list of the most likely diseases causing the symptoms. In this example, this would be a list of professions which your subject might have. The list does not have to contain exactly 5 options (in fact, the more the merrier). The important thing is that you're updating it with every new piece of information you get.

Putting it all together, we have a man who carries medication in his larger-than-average briefcase, may travel occasionally for work, is frequently on the phone, meets his clients one-on-one, and by the state of his cufflinks and suit, appears to be doing well for himself.

One very likely diagnosis is that the man is a pharmaceutical representative - someone whose job is to introduce prescription medications to doctors or even hospital administrators by handing out samples. The other man could be a doctor who may eventually prescribe this medication. A test for this hypothesis my simply be to ask him if he is indeed a drug rep, or any other question which, if answered correctly, would confirm your choice.

I hope you enjoyed this post, and there will be many more. For the most part I will post with the same regularity which I had prior to my break. As always, feel free to submit photos of people and places for deduction exercises.