In chapter three, Norman attempts to answer the question: What causes the discrepancy between the precision of behavior and the imprecision of knowledge?​​​​​​​
His thesis for answering this question is that knowledge is distributed between the HEAD and the WORLD: He postulates that human behavior is guided by the combination of internal (in the head) and external (in the world) knowledge. We can therefore minimize the amount of knowledge we need to have at any given time by restricting our choices or copying the behavior of others. 
In design, we must therefore be sure to provide sufficient cues (knowledge in the world) to facilitate the navigation of a new product, even without instruction.
Most of the time, great precision is not required for day-to-day actions, and approximate knowledge is enough for most behavior to “distinguish appropriate choices from all others”. Additionally, natural constraints usually drastically narrow down possible actions, as most objects are limited in their possible interactions with other objects. Furthermore, cultural constraints (in the head) restrict actions even further, thereby further reducing the need for knowledge in the head. 
Knowledge in the world
Norman explains that the need for us to learn is reduced when we can put some of the burden of remembering into the world. As we only need sufficient knowledge to get our tasks done, oftentimes we can make do with rough approximations and we usually structure our environment to provide a lot of external cues in our environmentSignifiers, physical constraints, and natural mapping can act as perceivable cues to help us navigate our tasks without too much knowledge in the head. 
However, when something in the environment suddenly changes, the cues don’t make sense anymore which can lead to chaos.
A specifically helpful way of simplifying the number of things we need to remember is constraints → External constraints control and limit our options, thereby reducing the burden on our memory.
Norman points out that, while technology helps us remember many things, some big design issues still remain, such as when it comes to security issues, specifically the choice of passwords. As soon as a password is complex enough to not be easily guessed, it is also usually too difficult to memorize. 
The safest method of encryption is to require multiple identifiers: something you have (knowledge in the world) + something you know  (knowledge in the head), but still, security poses major design issues – to which there is so far no solution yet. 
Norman then begins to structure knowledge into different categories: 
Knowledge OF = declarative knowledge
→ facts and rules, easy to write and teach
Knowledge HOW = procedural knowledge
→ things like muscle memory, only learned through practice, difficult to teach, largely subconscious
He then goes on to categorize memory, which is basically Knowledge in The Head, into Short-term and Long-Term memory.
Short Term Memory / Working memory
Short-term-memory, also known as working memory, is the memory of the just present. It is quite fragile and can easily get lost as soon as we get distracted. 
STM can only store about three to five “chunks” of information, whereas the complexity or length of these chunks seems to be less important than the amount. Therefore, it is easier to memorize long strings of random data by transforming them into meaningful chunks. Retention is affected primarily by time and number of repetitions, whereas meaningful information is easier to retain than random data.
What does this imply for design? We shouldn't count too much on people being able to remember things, especially when new information is always coming in. But, by using multiple sensory modalities: visual, haptic, auditory, we can avoid too much loss of concentration. Different modals of information are less likely to interfere with or distract from other elements stored in STM.
Long Term Memory 
Long-term memory is the memory for the past. It is still unclear exactly how it is transformed from short-term to long-term but it is clear that sleep plays an important role. 
Storage of information in the brain seems to be according to interpretation and occurs faster the more it is rehearsed and repeated. However, information stored in LTM is difficult to organize and not consciously available. As it is not directly retrieved but rather reconstructed each time it is recalled, it is, therefore, subject to bias and manipulation
Arbitrary Knowledge is knowledge for things with no underlying meaning or structure, such as commands, gestures, and procedures. It is learned by artificially providing structure, which is why a good conceptual model can help provide meaning and facilitate learning. 
Good use of constraints and natural mapping can also facilitate the use of new technology.
In summary, knowledge in the head and knowledge in the world are always working together and in a tradeoff with each other. It is the job of the designer to provide as many cues in the design (knowledge in the world) as possible to alleviate the load on a new user’s working memory (knowledge in the head).

You may also like

Back to Top