Blink: The Power of Thinking Without Thinking

Have just finished Blink by Malcom Gladwell (an appropriately quick read), and thought I'd jot down a few notes of summary:

** our intuition is smarter than we give it credit for, and we can make surprisingly accurate judgments based on comparatively 'little' information

**adding more information can actually mess up our judgment by obscuring what's really important

**unless we're experts, we are awful at actually explaining/understanding what informs our preferences and judgments (prime us to be 'logical' about taste tests, for instance, and we'll rank it completely differently -- and 'incorrectly,' in terms of quality)

**context is hugely important in our decisions; too little time, too much pressure, cultural and racial and gender prejudices all mess with our ability to judge properly

A lot of interesting anecdotes to back this up -- one I found particularly interesting in the context of librarianship had to do with doctors diagnosing possible heart-attacks in a poor Chicago district hospital. Gladwell described how a scientist created a 'decision chart' with a few very limited factors (ECG, systolic blood pressure, fluid in lungs) -- essentially an algorithm for diagnosis -- and found that it increased doctor accuracy by 70 percent.

Medical professionals were incredulous, arguing that the model didn't take enough of the patient situation into account, it was far too limited...more information is always better, right? But Gladwell argues that more information in this case was actually detrimental to the doctors' ability to properly diagnose the symptoms; data about the patient's background, other conditions, exercise levels, etc. were all adding up to 'white noise' that interfered with their ability to process the relevant factors.

So, as an 'information provider' at a medical reference library, it occurs to me that swamping down our patrons and clients with data is not only unhelpful but actually dangerous - if Gladwell is right, information provision needs to be tailored to providing the 'best' data for a given diagnosis situation, not 'all data you could possibly be interested in.' In a vast oversimplification: Information Overload Kills.