Psychologists from The University of Manchester have shown how difficult it is for us to guess the true intention of each other's behaviour.
The study, published today in Attention, Perception, and Psychophysics, has important implications on public policies designed to impact on areas such as smoking, obesity, eating disorders, self-harm, alcohol use and gambling.
Clinical psychologist Dr Warren Mansell, who led the study, says policy makers need to accurately understand what a person is trying to control using their behaviour, rather than trying to change the behaviour itself.
He said: "We think we know what someone is doing just by observing them. For example if we see someone move a steering wheel of their car, we assume they are aiming to keep their car in the centre of the lane.
"But our study shows that it is incredibly easy to be mistaken – and that has important implications on anyone whose task is to change human behaviour.
"In psychological research, for example, this study suggests that some behaviour studied may be no more than a side effect of participants'true intentions.
"We should therefore avoid focusing on people's behaviour itself . That would lead to multiple and inevitably futile interventions for each and every problem."
He added: "In terms of public policy, we frequently we see money spent on another new initiative for 'behaviour change'.
"Yet if these behaviours are just side effects of people trying to exert control, then this multi-pronged approach to health is highly inefficient and fails to address the common root cause of people's difficulties.
"You need to ask people what they want in their life and how they solve their problems. Smoking, for example, is just one of many different ways in which a person might try to control something important to them – such as their social confidence, or emotional state."
Dr Mansell's team were able to fool over 350 people into thinking that a person they saw in a video controlling the location of a knot in a rubber band was doing something completely different.
The study participants wrongly thought it showed someone drawing a picture, exposing the phenomenon of 'control blindness'.
Most participants failed to correctly infer the person's intention, instead inferring complex but non-existent goals, such as 'tracing out two kangaroos boxing', based on the actions taken to keep the knot under control.
The effect persisted with many participants even when their awareness was successfully directed at the knot whose position was under control.
Much of the work was carried out by Andrew Willett, an honorary research assistant who spent three months at Manchester on secondment from Vassar College, Poughkeepsie.
The results were modelled on a computer by Dr Rick Marken, based at Antioch University.
Dr Mansell said: "It might surprise some, but this study backs up what we have found in studies of people's health and well-being. What's more, if control blindness is apparent with the simple behaviour in our study, then it is even more likely in everyday life where people's actions can be sophisticated or subtle.
"Observers with control blindness appear to be unable to guess what the true intention of someone's behaviour is, even if their attention is drawn towards it.
"Seeing the unintentionally produced pattern of pen movement as intentional is an example of what has been called'the illusion of control'
"Our study shows that the side effects of intentional behaviour that create the illusion of control can be so compelling that they blind people to the true intention, leading to control blindness."
Explore further: Dopaminergic drugs cause changes in deep brain areas of Parkinson's patient