Why dynamical systems approach to oral fluency research?

Oral fluency is an elusive construct. There are observable aspects of it which can be quantified. For example, time taken to speak, silent time during the turn, etc. There are also non-observable aspects that cannot be quantified or even clearly known about. For example, we do not know what are the underlying cognitive processes that lead to what we observe as speech. Among the observable aspects, there are elements which we do not know what to do with. For example, hesitations or repetitions: we don't know for sure if these are due to planning processes, lack of resources, or lower proficiency.

This is the where dynamical systems theory come in handy. Dynamical systems are complex systems or complex adaptive systems which have multiple interactive elements that change over time. The change itself dynamic in dynamical systems. How the systems changes over time changes over time.

The five features of dynamical systems as enumerated by Norman Segalowitz (2010) are as following:

1. Heterogeneity of elements and agents: there are a variety of interconnected elements within the system and they interact with each other. In case of fluency, they could be individual learner, cognitive and perceptual processes, beliefs, motivations, emotions, knowledge, aptitudes and attitudes of the learner.
2. System dynamics: connections and interconnections within the system, between elements change over time.
3. Non-linearity of the system: apart from change in interactions an connections, change itself changes within the system.
4. Openness of the system: dynamical systems are susceptible to external influences. They play a critical role in maintaining dynamic equilibrium (as opposed to steady state equilibrium).
5. Adaptation: a change in one part of the system rewsults in the system functioning in a changed or new way.

All components of dynamic systems thus are interconnected. This is challenging for the researcher since any predictions based on the whole system would be difficult to make since the system is dynamic and constantly changing. However, overestimating the system's flexibility may result in missing the regularities or patterns in the system. 

In the case of fluency research, dynamical systems approach can be useful because there are multiple interacting elements that constantly change/develop/evolve. Taking such interactions into consideration while designing research studies would do justice to the underlying processes of fluency, that are governed by these dynamic elements. Therefore, instead of looking at fluency as a static construct generated by a number of static elements, dynamical systems approach would be more meaningful for research.