The Interaction Framework

Introduction

The interaction framework, also known as Abowd and Beale’s interaction framework, is an extension of Norman’s model of interaction. This model addresses the limitations of Norman’s model.

• Along with the user’s task language and system’s core language, input and output components also have their own language. The ways and actions allowed by the system to input data using the system interface are called the input component’s language. How the system outputs and presents information on the system interface is called the output component’s language.
• Both input and output form an interface.
• Interaction takes place through the translation of languages between components.

The four steps of the interaction framework

An interaction consists of four components, hence four translations are required, one from each component to another, to execute one cycle of an interaction. These translations are articulation, performance, presentation, and observation.

Articulation

Articulation translation is the translation of the user’s task language to the interface’s input language. The user first formulates the goal and the tasks to achieve the goal. Then the user performs the actions and converts them into stimuli for the input component. Articulation error occurs when a user formulates or performs wrong actions.
Example: A user types some text in MS Word and uses shortcuts to copy the selected text.

Good Articulation	Poor Articulation
User uses correct shortcut keys.	User mistakenly uses the delete shortcut key.

Performance

Performance translation is the translation of the input language to the system’s core language. The response of the input is converted into stimuli for system execution. The system perceives the actions, executes them, and updates the state. Performance error occurs either when the system does not allow the user to perform the intended action or when the system fails to understand the intended actions.
Example: A user types a person’s name in the MS Word processor.

Good Performance	Poor Performance
The system interprets that it’s a name and does not show a spelling error.	The system fails to interpret and shows an error along with suggesting possible English words.

Presentation

Presentation translation is the translation of the system’s core language to the output language. After the system has executed the actions, its state is updated. This updated state is converted into stimuli for the output component to be represented by the interface. The system presents the state on its interface. Presentation error occurs either when the system fails to update the interface according to its state or when the updated interface conveys incomplete or ambiguous information.
Example: A user selects a mobile application to install and starts the installation. After some time an error message appears and the installation stops.

Good Presentation	Poor Presentation
The error message says, “Unable to install the application. Phone memory is full. Please empty some space and try again later.”	The error message says, “Unable to install the application.”

Observation

Observation translation is the translation of the output language to the user’s task language. After the system’s updated state has been presented by the output component, the user converts this response into the stimuli for personal perception to evaluate the result.
Example: The user clicks to close the text editor without saving work and the system asks for confirmation before closing in the form of a dialog box with two options, “save and close,” and “close.”

Good Observation	Poor Observation
The user observes the dialog box and clicks on the save and close option.	The user fails to pay attention and clicks on the close option.