Hi Todd Thanks for your continuing thought experimentation! I'd say that using metaphor-as-reflection potentially combines some of the qualities of the meta-model and reframing. You're helping them understand/clarify the structure/form of the map they've created, which may also lead them to moving up a level 1. You are confirming your understanding of what they're presenting (or giving them a chance to clarify), which in itself can be quite beneficial - sometimes all we need is for someone to hear and understand us! This is similar to paraphrasing your understanding of the content back to the client - mentioned in JG/CB's Verbal Package (the New Code refinement/update of the Meta-Model). 2. There's a kind of dissociation. If the patient has a phobia of fish and you use a metaphor of grains of sand, you help them to work with the structure/form of the problem, without the direct emotional associations. I think this is one of the major benefits of the Symbolic Modelling approach (from the "Metaphors in Mind" book you mentioned in another post). 3. Further to 2, there's a possibility to elicit different, more positive associations. If they love football and you use a football metaphor, you're associating a more resourceful state to the structure of the problem. It might be even more ethical/effective to ask them to create their own metaphor? "If this was a football match, what would the situation be?" 4. As mentioned previously, you're offering a second description - a new perspective. 5. You may move them implicitly towards a process that will help to resolve the problem. For example, if you tell them the story of the six blind men feeling the elephant and each describing it as a different animal, you both describe the form of their problem (their decription is made up of small unconnected chunks) and imply a process that may help solve it (combine the chunks, or somehow get a bigger picture) Perhaps the "resonance" you mentioned is a function of the combination of these factors, and especially of 1. and 3. - how accurately the metaphor reflects the form of their problem, and how much it associates a more resourceful state to that form? Process is content if you move up a level... and so on ad infinitum. But, in a therapeutic context we have a start-point which is the content that the client is presenting to us as their problem - that becomes our baseline, our Level 1 (for the duration of the session). We can work on that level to help the client examine/clarify their map (using for example, meta-model, or metaphor-as-reflection), which will sometimes be enough to allow the client to solve their own problem. Or we can move up to Level 2 (relative to the client's Level 1) to work with the processes used to create content on Level 1. I think an important point is that the type of hierarchy of levels we use is of the same type as Bateson's Levels of Learning - i.e. the criteria defining a higher level is that it consists of the set of processes which can be used to produce the things at the lower level (in our case, the processes used to create the client's current map of the problem-context and those which could be used to create different maps). I'm kinda thinking that the New Code games give a third option - to move down a level (Level 0?), closer to the original stimuli from which the client created their problemmatic map, associating a highly-resourceful state (with a sprinkle of know-nothingness inside) to them, so that next time the client encounters those stimuli they will create a new map (or perhaps, operate without a map!). But that's a "thought in progress" and I'm not sure if it fits with the logical levels model! Cheers, Jon