Cognitive Psychology and Human Factors

Course notes for my Human-Computer Interaction (HCI) course at the EPFL:

Cognitive science: study of how the mind works. Human behaviors.
Information process model (Card): This model consits of three interacting systems. The perceptual (seeing, touching, and hearing), motor (movement of body parts) and cognitive system (The main reasoning relying on facts and knowledge. Acts as processor).

Mental Models:
People have mental models of how things work. It allows them to make predictions on how things work.
A mental model missmatch appears when psychological variables differ from physical ones.
The designer model is the designer’s mental model on how the system should function
The user’s model is the user’s mental model of how their tasks should be accomplished.
The system image is what has been built (how the designer communicates with the user).
The gulf of execution is the distance between user’s goals and the means of achieving them through the system (unintended system image)
A design has to find a good balance between default values and personalization.

Learning:
It is hard for adults to purely learn. It leads them to minimize learning and learning new skill via skill transformation and skill replacement.

Minimize learning by metaphor
Using a metaphor to retrieve analogical information to get familiar with (e.g. desktop metaphor for office workers).
The key is to find the right metaphore that matches user’s mental models. (be carefull of overdoing, overly literal, overly cute, mismatched)

Minimize learning by consistency
Consistency of effects (same words, commands, actions always have the same effects -> predicatability)
Consistency of language and graphics (same information in same location)
Maintain a display inertia. Good interaction design changes as little as possible from one screen to the next

[We watched Apple's knowledge navigator video]

User-friendlyness:
Accelerate the process whereby novices begin to perform like experts. Improve the learning curve (advance the elevation stage of the learning curve)
Acquisition of skills (learning curve): developping skills in solving ever-larger chunks of a problem automatically

Types of users (that have to be met in one design)
Novice users have no syntactic knowledge (understanding of the structure, the form, the way it is presented) of the system and only a little sementic knowledge (understanding of the content). Attentive to low level details -> valuable cognitive resources are diverted from the central problem at hand
Intermittent users: maintain semantic knowledge of the system buy loose syntactic knowledge
Frequent users: they want fast interation, powerful tools, accelerator key. Experts solve large chunks of problem
Power tools vs loss of control: most successful applications offer the right amount of intelligence in their software (rule checker -> diagnose and repair)

Problem solving (-> task mental model)
Set up goals -> plan the action -> execute the action -> evaluate if actions help -> change strategies if necessary

Feedback:
There are two types of feedback. Articulatory (confirmation of the physical operation) and semantic (confirms of how the operation was performed (right-wrong))
Take into consideration the response time (how users perceive delays)
Mark exits and cancel