Usually, the founding-father status for Chaos and Complexity theory is given to Edward Lorenz in the 1960s. He was a weatherman interested in long-term forecasting using mathematical models. He found that a small error in the input to his computer (0.506 instead of 0.506127) led to extremely divergent patterns of weather.

From nearly the same starting point, the computer-generated patterns of weather grew further and further apart, until all similarity disappeared.

This phenomenon is known in Chaos theory as ‘sensitive dependence on initial conditions’, otherwise popularised as “The Butterfly Effect”. This characteristic of chaotic systems means that weather, and other naturally occurring systems are difficult to predict with any degree of accuracy far into the future.

The Butterfly Effect

The statistics that we use in psychology, (even the sophisticated ones), are associated with linear dynamics. We are used to ‘cause-and-effect’ relationships that can be calculated to give a predictable result. These linear systems work within clear definable limits; examine the assumptions for using parametric statistical tests for instance, or the idea of isolating all influences on the dependent variable apart from the one being manipulated by the experimenter. In this world of linear dynamics, the orderly and methodical are the norm; ‘things add-up’ to give predictable outcomes. It’s a world of clockwork; discover how the bits work and we can predict the time (as told by the clock!); generalising this research paradigm, we can discover how other ‘bits’ work and go on to discover the rules for the weather and even people. This will allow us to predict weather and people behaviour and thus ultimately control both.

(Un) fortunately, linear systems exist mainly in theory. Outside of the laboratory and mechanistic theorising, living beings and natural eco-systems are non-linear. A non-linear system cannot be described with traditional equations. The relationship between inputs and outputs in such systems is non-linear e.g.

Fig.2            Mathematical examples of linear and non-linear systems

This concept of chaos is extremely difficult for researchers who want to change one variable at a time and explore simple linear and causative relationships using traditional parametric measurements. Outside the laboratory, real life is both elaborate and seemingly erratic; relationships are unpredictable and capricious; twisted causal paths are interrupted and variables continuously affect each other in a network of rebounding feedback loops. And yet there are discernable patterns in this chaos e.g. traffic flow, weather changes, cardiac arrhythmias, crowd behaviour.