For each object, you need to decouple its function from its form. McCaffrey (2012) shows a highly effective technique for doing so. As you break an object into its parts, ask yourself two questions. "Can I subdivide the current part further?" If yes, do so. "Does my current description imply a use?" If yes, create a more generic description involving its shape and material. For example, initially I divide a candle into its parts: wick and wax. The word 'wick' implies a use: burning to emit light. So, describe it more generically as a string. This brings to mind using the wick to tie things together (once I extract it from the wax). Since 'string' implies a use, I describe it more generically: interwoven fibrous strands. This brings to mind that I could use the wick to make a wig for my hamster. Since "interwoven fibrous strands" does not imply a use, I can stop working on wick and start working on wax. People trained in this technique solved 67% more problems that suffered from functional fixedness than a control group. This techniques systematically strips away all the layers of associated uses from an object and its parts.
