If there is anything that this horrible tragedy can teach us, it’s that a mental model is a precious, precious commodity

Back in the mid 1800s, the mathematics community was wracked by debate over non-Euclidean geometry. Can it exist? What does its imply about existence as we know it? How do you even have a sensible conversation about it? The fervent partisanship of a divided geekdom presaged today’s tabs-versus-spaces debate.


Pied Piper is rumored to be working on middle-out indentation

To give a sense of how intense was the debate was, consider Alice in Wonderland. Lewis Carroll designed Wonderland as an cautionary tale of the existential dangers to our very conception of reality of a mathematics unbounded from Euclidean oversight (with the Cheshire Cat symbolizing imaginary numbers which vanish into thin air), much the way The Matrix was a cautionary tale of AI unbounded from human oversight (which adds a whole layer of meta to all those white rabbit references).

What brought the debate decisively to an end was Beltrami-Klein’s constructing a real-world model of non-Euclidean geometry: on the hyperbolic side as a pseudosphere, on the elliptical side as great circles. Once an abstract concept can be overlaid onto a tangible model, its feasibility lies beyond doubt.

The power of models works in the other direction, too. The earliest pioneers of microeconomics developed the model for perfect competition by looking at the salient attributes of an existing market – agriculture – and abstracting out the most salient features.

But what makes models really powerful is that they’re not just passive reflections of the idea, but active tools that can produce new insights and spur action.

The book Freakonomics was basically about constructing new economic theories to explain real-world observations, while conversely, Nudge was about anticipating (and steering) behavior based on attempt to anticipate (and steer) behavior based on theoretical models on information and motivation.

OK, but so what?

So in the first post, I set out a goal of defining “spirituality” with the requirement that it avoid any undefined metaphysical terms.

And in the second post, I developed that definition using a model of the mind as a Google Map, consisting of features both objective and subjective, where “science” vs. “spirituality” is defined as the cultivation of the former vs. latter types of features.


The payoff in this post: putting that model of mind-as-Google-Map to work as an actual tool in improving one’s mental health.

Case in point: CBT identifies 10 key types of cognitive distortions or negative automated thoughts (NATs). Having social anxiety disorder, my #1 NAT is catastrophizing. Basically, blowing perceived dangers out of proportion, leading to paralysis, passivity, and stress in real time. Which in turn can lead to rumination, regret, and anger.


and anger leads to terrible, terrible acting


The form of that “danger” can vary from introducing bugs if I migrate a partially-tested code change to awkwardness if I approach a woman without any idea what to say.

In our mind-as-Google-Map model, an analogous situation would be seeing red on the road ahead to the edge of the map – which may be only a few blocks – but assuming it extends forever and will hopelessly derail my changes of ever reaching my destination.

Getting back to the committing code / chatting up scenario, the corrective is to take a step back and put the threat into proper proportion. Running through the worst-case scenario to establish a lower bound is helpful: “if the system breaks, you back out the change and replay any interrupted transactions” or “if she doesn’t return interest she’ll eventually go her own way and never see you again which would have happened anyway“.


if only

None of which is terribly surprising, but if you have anxiety issues like me, it’s REALLY hard to remember to do in real-time. And the real-time aspect is important, because the optimal course of action inevitably becomes apparent in hindsight, always it’s too late, and quite often at my next therapy session.


Which is where the advantage of the mind-as-Google-Map model comes in. Because the analogous corrective operation in the map is, of course, zooming out.

buttonsBut here’s the kicker: those little “+” and “-” zoom buttons are permanent features of the map. Which means the very act of modeling your problem in terms of the map simultaneously models your solution.

And that raises the odds that you’ll remember to take a step back, put the risk into proper proportion, and act accordingly, at that moment. The Google Maps isn’t quite putting the solution directly into your hands, but at least politely coughing “ahem” and tipping its head in the general vicinity.

As an exercise with my therapist last week, I went through the full list of 10 cognitive distortions defined by David Burns and modeled the solution for each in terms of the mind-as-Google-Map.

  1. catastrophizing (above)
    • distortion: overreacting to a perceived danger
    • problem model: red congestion that extends length of road to edge of screen
    • solution prompt: zoom out button
  2. Screen Shot 2016-05-31 at 11.51.43 PMmind reading
    • distortion: assuming others think as you do (or you as they)
    • problem model: projecting subjective map features (position, destination, route) onto other’s screen
    • solution prompt: start/end and path selection screen
  3. black-and-white thinking:
    • distortion: seeing events in all-or-nothing terms (usually nothing)
    • problem model: losing the path or having a bad path in navigation mode
    • solution prompt: auto recalculation of a new path
  4. overgeneralization:
    • distortion: concluding negative outcome based on a limited sample size
    • problem model: taking same journey repeatedly, expecting same results
    • solution prompt: refresh button. Google dynamically recalculates results each time from scratch, and results often differ from one minute to the next
  5. mental filter:
    • distortion: dwelling on negatives without seeing positives
    • problem model: red spots of traffic congestion
    • solution prompt: green spots of low congestion
  6. discounting positives: really the same as mental filter
  7. should statements:
    • distortion: judging what “should” be instead of observing what is
    • problem model: not having a destination, or path to destination, or alternate paths, or projecting yours onto others
    • solution model: see Mind Reading solution prompt
  8.  labeling:
    • distortion: applying normative labels rather than neutral descriptions
    • problem model: running late
    • solution model: just provides ETA
  9.  blame:
    • distortion: taking or assigning undue level of responsibility – focus on perception instead of substance
    • solution prompt: updates accompanied by suggested alternatives
  10. emotional reasoning:
    • distortion: reasoning from how you feel at a given moment
    • didn’t get around to modeling this one – perhaps feeling lost between recalculation? pulling over to side of road while recalculating? or setting out on fresh journey before first getting bearing?


At some point I might go a step further in developing this model to correlate:

  • mindfulness => navigation mode vs. bird’s eye view
  • fear => stepping out of accepted boundaries private property, unmapped areas, dirt roads, buildings, switching from car to foot
  • having values, priorities, and goals => compass, path, orientation


but at this point I’m really tired and really gotta get back to work.


And since that’s a kinda weak note to end this 3 post sequence on, here’s this post’s title reference (jump to 1:41):




This entry was posted in Uncategorized. Bookmark the permalink.