Negative and positive feedback
Apart from positive feedback, systems can also have negative feedback loops. Policy makers use negative feedback on a day to day basis as they take corrective action when events deviate from the norm or what is desired.
Negative feedback involves using feedback information to reduce the ‘error’ between desired and observed behaviour. It is called ‘negative’ because the response goes in the opposite direction to the ‘error’. You use negative feedback to drive a car, e.g. if the car is going too far to the right you steer a bit more to the left.
The feedback cycle of policy is shown above. For small differences between the desired and actual system state it is appropriate to make relatively small changes, whereas for large differences more radical policy action may be appropriate. Negative feedback corrects deviation – which is usually a good thing.
In contrast, positive feedback means that effects are increasing, not necessarily that they are desirable. For example, the positive feedback loop below is one of increasing frustration for patients and increasing pressure on the doctors.
In a situation like this, positive feedback can make the system spiral out of control. In the hospital the untreated patients might get increasingly angry and aggressive, requiring the police to be called in. More likely, one of the doctors would make a public announcement apologising for the long delay and explaining that interrupting the doctors only makes the delays worse.
Positive feedback can be a good thing for policy makers, for example the more people that give up smoking the easier it is for smokers to to give up, which increases the number of non-smokers. On the other hand positive feedback can bring the system crashing down. For example, during the financial crash of 2008 there were cascades of institutional failures that nearly destroyed the global financial system.1 As another example, as more people join a protest even more people may be inspired to join it 2.
What do you think?
Can you think of a positive feedback loop that has an undesired effect? What policy would you suggest to counter this? Share your thoughts in the comments.
1 Xuqing Huang, Irena Vodenska, Shlomo Havlin & H. Eugene Stanley, Cascading Failures in Bi-partite Graphs: Model for Systemic Risk Propagation, Nature Scientific Reports 3, Article number: 1219 (2013). doi:10.1038/srep01219
2 Sandra González-Bailón, Javier Borge-Holthoefer, Alejandro Rivero & Yamir Moreno, The Dynamics of Protest Recruitment through an Online Network, Nature Scientific Reports 1, Article number: 197 (2011). doi:10.1038/srep00197
© The UNESCO UniTwin Complex Systems Digital Campus