Models and Theories: Cognitive models

Notes

Study Notes

HCI Notes: Cognitive Models

HCI Models: Cognitive Models

Predicting user behavior and performance by modeling human information processing.

1. Purpose and Types of Cognitive Models

Cognitive models are tools used to analyze how users interact with a system by representing the user's cognitive processes (memory, perception, problem-solving). They help designers compare different interface options quantitatively before implementation.

Model Categories:

  • **Predictive Models:** Used to predict user performance (e.g., time to complete a task). They focus on efficiency. (Examples: GOMS, Fitts' Law).
  • **Descriptive Models:** Used to describe the user's knowledge and mental structure (e.g., Mental Models, Cognitive Walkthrough).

2. Predictive Model: GOMS (Goals, Operators, Methods, Selection Rules)

GOMS is a family of predictive modeling techniques used to calculate the time required for an experienced, error-free user to perform a routine task.

A. The Four Components

  • **Goals:** What the user intends to achieve (e.g., Delete File).
  • **Operators:** The cognitive, perceptual, or motor actions required to achieve the goal (e.g., Click, Read Text, Think, Type String).
  • **Methods:** Learned sequences of operators and subgoals used to accomplish a goal (e.g., Method for deleting a file: Locate icon -> Drag to trash -> Empty trash).
  • **Selection Rules:** Rules used to choose between competing methods for the same goal (e.g., If file is visible on desktop, use drag-and-drop; else use search function).

B. Key-Stroke Level Model (KLM)

A simplified GOMS model that focuses only on motor actions and a few cognitive operations to predict task execution time. Time is calculated by summing the duration of keystrokes, pointing (mouse movement), homing (hand moving to the keyboard/mouse), drawing, and mental preparation ($T_{execute} = \sum T_{K} + T_{P} + T_{H} + T_{D} + T_{M}$).

$$T_{execute} = \sum T_{K} + T_{P} + T_{H} + T_{D} + T_{M} + T_{R}$$

3. Predictive Model: Fitts' Law (Motor Performance)

Fitts' Law predicts the time required to rapidly move to a target area based on the distance to the target and the size of the target. It is fundamental to pointer-based GUI design.

A. The Law (Predicts Movement Time, MT)

$$MT = a + b \log_2 \left( \frac{D}{W} + 1 \right)$$ Where:

  • $MT$: Movement Time.
  • $D$: Distance from the starting point to the target center.
  • $W$: Width (size) of the target along the axis of movement.
  • $a, b$: Empirically derived constants.

B. Design Implications

  • **Target Size:** Larger targets are easier (faster) to acquire. (Make buttons big).
  • **Distance:** Closer targets are easier (faster) to acquire. (Place frequent controls nearby).
  • **Infinite Targets:** Targets placed at the screen edges (e.g., the Start button or browser scroll bar) are theoretically the easiest targets to hit, as their $W$ effectively approaches infinity (Corners and Edges are "hot zones").

4. Descriptive Model: Mental Models

A user's **Mental Model** (or Conceptual Model) is their internal representation of how a system works. It is built from prior experience, analogies, and learning, and is used to predict the system's behavior.

A. Models in HCI

  • **User's Mental Model:** The user's internal idea of how the system works. Often incomplete or incorrect.
  • **Designer's Conceptual Model:** The designer's articulation of the system structure (often documented in specifications).
  • **System Image:** The actual representation of the system (the interface, help documentation, manuals).

B. Importance for Design

Good design ensures that the **System Image** clearly conveys the **Designer's Conceptual Model**, which in turn helps the user quickly form an accurate **Mental Model**. Inconsistent interfaces lead to fragile and incorrect mental models.

Navigation

Back to Module
Module Complete