Kirsh points out that Norman's model makes an unrealistic assumption: The user is assumed to know the environment and its options and is merely picking an option. In practice, we do not always know what the options mean or even what options are available. Kirsh argued that users need to actively explore interfaces to become aware of the available functions and how they work.

Code-switching refers to a switch in language to match the capabilities of the communication partner. For example, you likely use different language when talking with friends and with family.

Human–machine interaction, according to Suchman, is similar to but different from human–human dialogue. It is similar in the sense that people pursue a shared understanding: They actively work to make themselves understood. It is different in the sense that the communication abilities of computers are limited, which requires humans to adapt.

Norman offered two central concepts to help us understand these cognitive efforts: the gulf of execution and the gulf of evaluation. These two concepts describe inferential breakpoints for users seeking to express their intentions and interpret feedback from the system, respectively.

The model subscribes to a theoretical assumption about dialogue: The defining cognitive challenge in dialogue is understanding the communication partner such that the appropriate next turn can be taken. In other words, the dialogue is intentional or goal-directed: Users aim to drive the computer to a particular desired state.

A significant early theory of dialogue interaction is the seven-stage model of Norman [600]. It considers interaction as goal-directed, turn-based dialogue.