The Design of Everyday Things: Knowledge in the Head and in the World

This chapter begins with the shocking realization that people's knowledge and behavior are not always equivalent - for example, a typist can type with speed and accuracy without being able to arrange the keys on a keyboard. The author attributes the following reasons for this phenomenon: information is in the world and combined with knowledge to produce behavior, precision is not required - the correct choice need only be differentiated from the others, natural constraints provide limits, and cultural constraints guidance. Furthermore, people possess two kinds of knowledge: knowledge of (explicit memory, declarative knowledge) and knowledge how (implicit memory, procedural knowledge).
Memory is divided into short-term memory (STM) and long-term memory (LTM). In computing terms, STM in analogous to RAM and LTM a hard disk. STM is limited to 5-9 segments - these segments need not be individual characters, a technique called chunking can be used (ex. a phone number is typically 7 digits in 3 chunks) and STM is very volatile. LTM contains information that takes longer to retrieve but isn't as easily forgotten. The author categorizes memory: memory for arbitrary things (rote learning), memory for meaningful relationships, and memory through explanation. Memorizing arbitrary things is difficult because of a lack of clues - no context. Memorization based on relationships is significantly easier, giving constraints and structure to limitless possibilities. The best form of memorization involves understanding. This allows a person to reconstruct the knowledge trying to be remembered using procedural memory. This is why mental models are so valuable.
In addition to memory, knowledge also exists in the world. One such form of knowledge is reminders, which consist of a signal and a message. Another way the world can communicate is with natural mappings, such as with burners on a range. There are tradeoffs associated with how knowledge is stored. Memory is requires learning and is not readily retrievable, but is very efficient and doesn't rely on clues. Knowledge in the world doesn't require as much overhead, but is dependent on the environment.