As quoted from Wikipedia,
The word chunking comes from a famous 1956 paper by George A. Miller, The Magical Number Seven, Plus or Minus Two : Some Limits on our Capacity for Processing Information.
In NLP we have adopted the term “chunking” as a process of moving to details and specifics as “chunking down” versus “chunking up” which is a process of moving to abstractions and conceptualizations.
For instance, a starting concept like a “car” when chunked up will go into a higher abstraction called “transport” and can also chunk up toward “convenience”. If chunked down, “car” can be chunked to “brand” like BMW or Toyota. Subsequent chunking down can go into parts of the car.
The use of chunking is primarily for the purpose of getting different individuals to think at the same “chunk size”, because in mismatching chunks, we will end up breaking rapport due to a mismatch of our level of focus.