In the early Summer of 2012, my wife and two daughters travelled to New York to visit family. Work kept me in town, but I was able to get over to Berkeley for a two-day, 9-5 go workshop hosted by Janice Kim. Janice is a professional player born and raised in America. It is because of this, I believe, that she is an excellent teacher of Americans.
All attendees were given the opportunity to submit three games for Janice to use to develop the material for the workshop. I was somewhat surprised to find one of my games at the top of the stack we were handed as the workshop started. I was delighted, however, at the thorough analysis she proceeded to give. A link to the complete analysis is provided below.
This is arguably the biggest theme of the workshop. We may be playing well, identifying good points, finding sharp moves, yet still lose the game. It is natural to ask, "What went wrong?" We may even review our game and find moves which look like reasons why we lost. "Ah, I made that bad move and it cost me the game." So we use those situations to teach ourselves how to improve. "Next time in this situation, I'll play differently." And in this way we try to learn. But we often fail to get stronger. Why? One reason could be that we failed to play correctly *earlier*, and were already in a bad situation.
When we play a bad moves, our opponent may not demonstrate its badness. When we play good moves, we may screw up the continuation, and thereby achieve a bad result. In both situations, we are learning the exact opposite of the true value of the move. This is what Janice means when she says, "there is no cause and effect in go." We literally have no way to judge the value of moves based on the subsequent results.
In fact, this is why we must take some things on faith. If you can help it, you should not give yourself the bad shape shown in the diagram. It seems like a reasonable way for white to help his other stone, and white may even find it to be effective many times. But these experiences do not make it a good move. On the contrary, it is bad shape. After , black actually gains more than white achieved with the kosumi (diagonal move).
Go, like everything it seems, is subject to a kind of 80/20 rule. It's as if 20% of the moves affect 80% of the outcome. Many moves are straightforward, not controversial. But some moves are pivotal. These pivotal moves occur at times when a great deal can be lost or gained depending on the strategy applied. If we could just know when these moments occur, we could slow down and consider the whole board carefully to determine the right course of action.
We are often blinded by the fact that we are thinking mainly of the moves we have made, the objectives we hold, the opportunities we foresee. There is go proverb which says something like, "the observer is three stones stronger." Why should this be so? Perhaps the observer is not so invested in the stones already played, and would be more willing to make a large strategic sacrifice. But of course, different observers have different experiences, and naturally bring different perspectives. But even beyond that, an observer has a different focal point.
The player is concerned with the moves, which moves will bring out the benefits seen in the position. The observer, however, gives more consideration to the reasons why the player played there. This frees the observer to operate at a different level. The observer, free of the player's "realities", can often see more than the player can.
So perhaps if we can become our own observers, we can gain three stones in strength. This is an area of research in learning theory known as dual-loop learning. People often have mental models they use to make decisions, and when asked to learn from a situation, they often simply recite those models. The idea behind dual-loop learning is to be critical of the models themselves rather than merely their external effects. Perhaps a quote will help:
"For Argyris and Schön, learning involves the detection and correction of error. Where something goes wrong, it is suggested, an initial port of call for many people is to look for another strategy that will address and work within the governing variables. In other words, goals, values, plans and rules are operationalized rather than questioned. According to Argyris and Schön, this is single-loop learning. An alternative response is to question the governing variables themselves, to subject them to critical scrutiny. This they describe as double-loop learning. Such learning may then lead to an alteration in the governing variables and, thus, a shift in the way in which strategies and consequences are framed." 
In order to help become observers of our own behavior, we need to think differently than we normally do. Sometimes taking an extreme viewpoint can help to break out of a limited way of seeing things, providing useful insights. To try to achieve this different thinking, we used role playing as described under WearManyHats.
Janice spent the first half of the first day reviewing my game. I have recorded as much as I can recall of her commentary on the game. Please take a look at this Kirk/Janice Kim Commented My Game 2012.
The game we reviewed Sunday afternoon was host to "the biggest point we've seen all weekend." Kirk/The Biggest Point Ever Seen.
After lunch the second day, a commonly seen life and death position came up, so we spent some time studying it. It is quite interesting how a subtle difference in connection can have a significant effect on the available strategies elsewhere. Take a look at JaniceKimsLifeAndDeathStudy
The end of the second day, Janice shared her, sure-fire way to play well in the opening. It's all about the brown M&M... MandMsBrown
Janice did give a few pieces of direct advice: Janice Kim's Guidelines For Better Play
When teaching, you should avoid saying that specific moves are good or bad. Is black 2 a good move? In many situations it is the appropriate response to a (sente) move like 1. But consider the diagram. If white continues to take these "sente" points, black will surely lose. The appropriate strategy here is for black to respond to 1 at 3, adopting a strategy of mutual destruction.
Instead of passing judgement on specific moves, review the thought process. You may ask, "What was your objective in playing this move?" And give suggestions on why that may or may not be an appropriate objective. You may ask, "Does this move accomplish that objective?" And then provide examples of other moves that might better fit the objective.
"If it's awkward to save, give it up!"
"Stones don't need friends."
"No one lives in sente."
"It's always ok to just ask." Meaning, you don't have to follow up your move right away. Similar to the proverb, "tenuki is always an option."
"Lost stones are an endgame goldmine."
"Rather than think about how to kill stones, I like to think about how I can effectively get my stones killed."
It is often critical to know the balance of territory to know the appropriate strategy to adopt. Consider reading Cho Chikun's PositionalJudgmentHighSpeedGameAnalysis to improve your counting skills.
"When something dies, there is a natural desire to immediately get it off the board in exchange for something, but it is better to just leave it." The opportunities lost by committing to a particular use of their aji are greater than what may be gained from waiting until the position changes, and a different way of using them becomes more effective.
"Some moves are their own punishment." A non-joseki move does not imply there is some significant, immediate gain to be made. When your opponent makes what you recognize as a bad move, it may often be best to just tenuki. The bad move is a burden and will show itself as such in due time.
 Smith, M. K. (2001) 'Chris Argyris: theories of action, double-loop learning and organizational learning', the encyclopedia of informal education, www.infed.org/thinkers/argyris.htm. Last update: May 29, 2012