The purpose of this paper is to show how tmta (Turing Machine Task Analysis), as a new model for an ecological approach to task analysis, can be applied to a complex task that involves the coordination of several separated components. Whereas cognitive engineering underlines the importance of modelling both ecological and cognitive constraints for designing a work system, only the first stage of the cognitive engineering framework (termed Work Domain Analysis), describes ecological constraints. tmta (Morineau et al., 2009) is a framework in which work domain affordances, patterns of information specifying affordance and agent operations were modelled with the help of Turing Machine formalism. Through Abstraction and Part-Whole Hierarchies (ah/pwh), the domain is described as a hierarchy of affordances. An agent who must perform a task picks up patterns of information from affordances. Information can take a value according to its relevance for the current task. On the basis of information, the agent can undertake three types of operation to change mental state through navigation in the Abstraction Hierarchy, make a transformation on the work domain objects, or move the object or itself. This model can simulate a task sequence and outline degraded situations in which information patterns cannot allow for progression in task achievement.
We present the application of tmta to the task of making a cup of tea. In neuropsychology, this task is used to assess action disorder syndromes with the help of a hierarchical task model that involves attainment of a sequence of goals and sub-goals. By comparing a simulation of the task with tmta and the usual hierarchical model, it can be seen that the tmta model is able to point out the complexity involved in such a routine task. In comparison with other methods of task analysis, tmta models constraints that arise from the sequencing of work domain affordances. In this way, this method provides the analyst with an operational definition of Gibson?s ecological concepts in order to understand a task.