Showing posts with label classification. Show all posts
Showing posts with label classification. Show all posts

Saturday, August 3, 2024

The Map Is Not The Territory

 

I ran into a quote this week that I must have read and forgotten from the past – because it was referenced in Bateson’s Steps to an Ecology of the Mind.  That was a book I read back in the hippie era after seeing it referenced in the Whole Earth Catalogue.  It happens at a time when I was writing about the usual philosophical rhetoric used to criticize psychiatry.  The circular logic argument I have encountered frequently by philosophers seeking to either destroy the profession or portray psychiatrists as unthinking buffoons.  That quote was “A map is not the territory” and it is attributed to Alfred Korzybski.

When I saw it – I associated immediately to the map I know the best and that is Hwy US2 running across northern Wisconsin between Minnesota and Upper Michigan.  I have travelled that road hundreds of times.  In fact, in 1988 I drove it over 200 times that year to keep a small inpatient psychiatric unit open. Maps these days are much better than they used to be.  For the old road maps to have the same scale and sufficient detail meant a large size that had to be folded and refolded to get it back into the glove compartment.  The above map is a clip from Google Maps and it can be scaled down to the individual house level and from there a street view that is regularly updated.

Thinking about old maps and new maps it is easy to see Korzybski’s argument. Driving US 2 late at night it is common to encounter characteristics of the territory that are not listed or even included in your GPS updates. The territory at night is much different than the territory during the day.  A major difference is deer on the highway.  There are the occasional deer crossing signs but I have suddenly found myself driving among a herd of 30 or 40 deer running next to my car and alongside the road.  The Google camera cars fail to update the video information fast enough to account for social and cultural changes that happen in the small towns along the way.  Am I going to encounter a large influx of out-of-staters for the Blueberry Festival in Iron River or the Strawberry Festival in Bayfield?  Is that small general store still there or is it finally gone? Is the posted or suggested speed limit accurate or do I have to correct for the weather?  

In the era of climate change even modern maps have uncertainty.  Highway 2 has been washed out and under water – both events that have never happened at any other point in my lifetime.  Using modern GPS guidance – I ended up on what appeared to be a dirt wagon trail that eventually got me back to Minnesota.  Every inch of that terrain looked like it had been seen by very few people in the last 50 years and no Google camera cars.  Most people unconsciously adapt to the terrain on the drive home – that sunken manhole cover or pothole to avoid.  We automatically adjust to the hazards even though they are not indicated on any map.  

Korzybski’s argument is basically twofold. First – no matter how far you drill down with a map – even a much-detailed map you will not find what you are experiencing – what your perceptions tell you is there. The map after all is an abstraction by someone and that is not a perfect representation of geography but also not your reality.  From consciousness science - your reality or experience of it is not my reality.   From information theory – the human brain is acquiring much more information going forward than you can get from one derived across a series of finite dimensions and time.  Second – this has clear implications for the ideas of subjectivity and objectivity.  In medicine we construct clinical trials – with exclusion and inclusion criteria that eliminate large real populations and at this point cannot account for the heterogeneity in the remaining research subjects. That does not preclude progress but it should introduce humility into the eventual results. No matter how broad or narrow those selection criteria are – they are only an approximation of the real population who will be treated.

Lest these connections be seen as speculative – here is what map makers and geographers have to say about the situation.  Basic geographic data is a space-time location. In addition, there is other relational data that contextualizes a location.  Data and relationships are discussed in terms of model and how the model is a simplified representation of reality but not reality itself.  A good example was John Snow’s map of cholera during the 1854 epidemic in London and how he used that to determine the source and isolate it. Cartographers are aware of these relational loops to space-time location as well as the limitations that are due to the large number of contextual features.  The map cannot account for them all.  

What does it say about philosophy and rhetoric applied to psychiatry?   

It says a lot about classification systems.  Much research today is preoccupied with ideal classifications.  The DSM for example is criticized for not being a perfect diagnostic system when in fact (like all medical classifications) it is a crude system with additional landmarks.  The graphic below illustrates the problem and how the assumptions made for the diagram on the left do not reflect the reality of the diagram on the right. That diagram is more complex – but not nearly as complex as the real clinical situation. After all – if the clinical situation was accurately reflected in the diagram on the left everyone with schizophrenia would be the same.  Psychiatrists would not have to concern themselves with a developmental history, a social history and life narrative, a medical history, and a family history.  They would not have to consider critical psychological events in a person’s life and putting all that together in a formulation about what is unique about that person.  The territory of that person would include supportive people and important contacts. Like the map of Highway 2 – the DSM gets us into the ballpark but it is not specific about what we will find. 

Korzybski has been described as an independent scholar.  He is credited with inventing the field general semantics.  There is a research institute founded on his ideas. There are not a lot of scholars taking his work forward.  There is an excellent online biography of Korzybski that describes the controversies associated with his writings and varying degrees of acceptance.  Interestingly he wrote about psychiatry and in his biography, there was apparently a group of psychiatrists interested in his work.  He referenced “neuropsychiatry” as a field that had generally been ignored by the rest of medicine.

 

Irrespective of the complexity and controversy of general semantics – I am still focused on the map is not the territory concept for several reasons.  First it reflects what is going on in the DSM classification system.  Second, it describes limitations of any classification system and how that abstraction differs from reality. That is probably the reason that medical diagnostic systems die hard, especially after decades or centuries of the same observations.  Is there any reason to suspect a dimensional or sub phenotyping system would be any better?  Probably not at least until very detailed observations can be made.  A classic paper (4) suggested that hundreds of true/false questions identified psychological traits and that this was an actuarial method superior to clinical judgment.  Despite that alleged superiority many of the methods suggested in that review like the Minnesota Multiphasic Inventory or MMPI have fallen out of use and are no longer used for screening purposes or making diagnoses.  Machine learning and artificial intelligence can produce these results faster and on a larger database but continue to have limited applications.   Third, it reflects expert opinion by at least one of the top theorists in the field (5).  Fourth it reflects good clinical practice that includes a formulation with additional commentary on psychopathology, associated observations and theories. 

At the minimum I hope that you find Korybski’s observation as interesting as I do.  I probably will not read his voluminous works – but I am always aware of the fact that no matter what classification system you are using it is always an abstraction with various degrees of precision.  Further it is an abstraction by one person or a group of people.  The way the DSM (and all of medicine) is structured the precision of both the diagnosis and treatment of a particular patient depends on what occurs during the encounter and the experiences and abstractions of that physician.   

George Dawson, MD, DFAPA

 

Supplementary: Doing research for this post, I encountered another quote that expresses a similar idea:  "The menu is not the meal".  Alan Watts is credited with that quote. 


References:

1:  Korzybski: A Biography (Free Online Edition) Copyright © 2014 (2011) by Bruce I. Kodish.  See chapter 30 for Korzybski’s contact with psychiatry including Harry Stack Sullivan and William Alanson White:  https://korzybskifiles.blogspot.com/2014/06/korzybski-biography-free-online-edition.html?spref=tw

2: Doerr E. General Semantics. Science. 1958 Jul 18;128(3316):156.

3: Gardener M. General Semantics. Science. 1958 Jul 18;128(3316):156.

4:  Dawes RM, Faust D, Meehl PE. Clinical versus actuarial judgment. Science. 1989 Mar 31;243(4899):1668-74

5: Kendler KS. The Phenomenology of Major Depression and the Representativeness and Nature of DSM Criteria. Am J Psychiatry. 2016 Aug 1;173(8):771-80. doi: 10.1176/appi.ajp.2016.15121509. Epub 2016 May 3. PMID: 27138588.

.