One reason people have difficulty understanding the implications of the current issues involved in climate change is that they often assess system dynamics using a mental model of “pattern matching” in which they seek correlations among data and use these to project future values. Humans can detect positive correlations in data well, indeed often perceiving patterns where none exist.
In the context of climate change, Sterman and Booth Sweeney found that many people used a pattern-matching approach to project future climate variables, concluding that system outputs (e.g. global mean temperature) are positively correlated with inputs (e.g. emissions). Pattern matching often works well in simple systems but fails in systems with stock and flow structures, like climate change.
According to Sterman, “The overwhelming majority of everyday experiences involves simple systems where cause and effect are closely related in time and space, time delays are short, and information cues are highly correlated. The water in the teakettle boils and the whistle sounds. Unfortunately, that doesn't apply to climate change and the consequences for this misunderstanding are high.”