Please enable javascript to view this page in its intended format.

Queen's University
 

Lab-Sub-Banner-Hollenstein-ADL-Small2.jpg
Dynamic Systems Approach to Development

Decades of research on developmental change and stability has yielded a great deal of knowledge. However, integrating this body of knowledge into a coherent picture of development has been difficult for several reasons:

  • Developmental change is most often non-linear, yet the methods to test for change are typically linear statistical models. Thus, we have learned what has changed (e.g., adolescent onset of depression) but often have little explanation as to how this change occurs over time.
  • Deterministic, unidirectional cause is the exception rather than the rule. That is, not only does X cause Y but Y also causes X. Recently, bidirectional or transactional models have become more prevalent (e.g., a child’s influence on parenting behaviour) and this is a necessary step in the right direction. However, these models still presume an additive effect of several linear, unidirectional causal processes, rather than truly reciprocal causation.
  • Complex interactions among many factors cannot be modeled parsimoniously. For example, the behaviour of young adolescents is known to be influenced by family relationships, peers, media, hormones, changes in brain functioning, nutrition, sleep patterns, genes, stressful events, school, and cultural norms. How can we account for the confluence of so many factors simultaneously?

Resolving these issues is beyond the scope of traditional, linear, closed-system methods. System theories of human behaviour and development have a long history (e.g., Lewin, von Bertalanffy, Bronfenbrenner) of trying to address these problems. However, it was not until physicists were able to document the properties of dynamic systems that both theoretical and empirical headway was possible. Complex patterns of many natural and social phenomena including population dynamics, embryogenesis, chemical reactions, economic trends, viral epidemics, brain activity, and motor development have been successfully measured and modeled as dynamic systems. Thus, viewing human development from a dynamic systems perspective makes it possible to examine non-linear, complex, and reciprocally causal processes more explicitly.

Defined most simply, a dynamic system is a system of elements that change over time. All dynamic systems share several properties in common. These include:

  • Self-organization. Novel forms emerge spontaneously from the complex interactions among lower-order system elements. Thus, the state of a system is not pre-determined, nor is it the product of external causes.
  • Hierarchical organization of nested structure. Lower-order elements self organize to form the structure at the next higher level and these structures are then the elements for the next higher level, and so on. Moreover, this nested structure exists in time as well: real-time (moment-to-moment) processes are nested within longer time scales (e.g. situations) and these are in turn nested within developmental time.
  • Reciprocal and circular cause. Within a level, interactions among system elements are reciprocally causal (X and Y cause each other). Across levels, causation is circular. The lower-order elements create the macro structure, but the macro structure constrains interactions among lower-order elements. This is a difficult concept but perhaps the best example from developmental psychology is the emergence of personality structures. Moment-to-moment emotional experiences in various contexts are the elements from which personality emerges. Personality coalesces across development and reduces the variety of emotional tendencies into a relatively small and predictable set.
  • Non-linear dynamics. The behaviour of a system is governed by feedback processes responsible for both stability and change. Negative feedback processes are self-stabilizing. Through negative feedback the elements continue to be linked in a similar fashion over time and the stability of the system is maintained. Positive feedback amplifies small variations in the lower-order interactions to create system instability. This instability is necessary to break down old patterns and for novel forms to emerge in their place. The dynamics of a system are the result of the interplay of both positive and negative feedback processes.
  • Perturbation reveals the nature of the system. A system can only be understood by the response pattern following a perturbation. A system may appear stable, for example, but become rapidly unstable following a relatively small perturbation. Conversely, a system that appears equally stable as the first may be relatively impervious to perturbation and thus confirm its stability.
  • System change occurs through the process of a phase transition. A phase transition is a period of instability and high variability observed when one stable pattern or structure breaks down and a new structure emerges in its place. A classic example is the boiling of a pot of water. The structure of cold water molecules is highly ordered, but as heat is applied the molecules begin to bounce into each other in a chaotic or variable pattern. At the boiling point, the molecules return to a new stable pattern when they form ordered columns moving up and down from bottom to surface and back.


The major premise of my own research is that a developing individual is part of a dynamic system that includes both lower-order elements (e.g., neuronal development and brain activity) and higher-order structure (e.g., interpersonal relationships and contexts). Thus, the individual cannot be isolated from his/her environment, nor can the constituent elements be neglected. For this reason, the majority of my research examines the dyad (usually parent and child) as the unit of analysis.

Kingston, Ontario, Canada. K7L 3N6. 613.533.2000