Every day we are bombarded with messages that try to persuade us to change our behaviour for the good of ourselves and society. Drink less alcohol! Eat less food! Wear a mask! Get vaccinated!
Millions of dollars are spent each year on information campaigns designed to foster more cohesive and inclusive societies.
In developing economies, this includes efforts to foster democratic attitudes and more effective political systems.
Say no to gender-based violence! Reject corruption! Do not sell your vote!
Given how pervasive they are, the most surprising thing about these campaigns is that many just don't work.
Some even make the situation worse.
The reason why is important, because it tells us something about why it's so hard to change human behaviour.
Take anti-corruption campaigns.
Almost all of the programmes designed to reduce graft in high corruption countries include an awareness raising' component.
The idea is that through raising awareness about an issue, citizens will become more willing to reject requests for bribes and act against corruption.
The hope is they'll report it or vote against corrupt politicians.
But until the last five years, no one had systematically tested whether these messages were effective.
When researchers started to look into this, they got a nasty surprise: in most cases anti-corruption messages don't make citizens more likely to reject or condemn graft.
A test in Lagos, Nigeria, a nation with a history of high-level corruption, had some alarming results.
From 1200 randomly selected participants, it found people were actually more likely to pay a bribe.
The 1200 took part in a bribery game. They were told that paying a bribe would mean another participant would lose money much like in the real world.
Some participants received an anti-corruption message and some did not.
Those who received an anti-corruption message were on average more likely to pay a bribe. In other words, the effect of the messaging was the opposite of what was intended.
This means the millions of dollars being spent on anti-corruption messaging in countries like Nigeria may not just be wasted it may actually be making things worse.
Blame human nature.
Recent research in political and social psychology found that when we hear about topics that are pervasive, we tend to become cynical and focus on how big the problem is.
The way we behave is less shaped by what we are told to do than by what we believe our peers are already doing.
So, we are more likely to get vaccinated and follow social distancing protocols if we think all our friends have already done so than if the government tells us. The consequences of this for how governments need to communicate are profound.
If a campaign focuses on how corrupt the police are, or how few people are recycling, we point out how big the problem is and so run the risk of undermining people's confidence that a better future is possible while simultaneously suggesting that bad' behaviours are socially acceptable.
The effect of this double whammy is perfectly illustrated by anti-corruption messaging which confirms people's most cynical inner beliefs that the system is beyond repair and the only way to get things done is to take part in corruption by paying bribes.
Individuals who were already particularly pessimistic about the prevalence of corruption were significantly more likely to be negatively affected by reading an anti-corruption message.
The biggest worry is that this logic doesn't only apply to corruption.
One study tested messages designed to prevent the theft of wood from Arizona's Petrified Forest National Park.
It found that signs that described the problems caused by others' behaviour stealing wood were most likely to increase theft.
A more recent paper on how to encourage people to respect social distancing protocols during the pandemic in Australia concluded that telling citizens that others are not following the rules is dangerous because people match their behaviour to perceived social norms.
Similar conclusions have been reached with campaigns to stop gender-based violence and vote-buying.
It's important, then, to realise that raising awareness may actually make the problem worse.
Before embarking on awareness-raising campaigns policymakers could try and design less problematic messages.
One way to do this may be by focusing on those who behave appropriately', and the negative feelings people hold about those who do not, rather than highlighting the extent of the problem.
Emphasising the strength of feeling within society and the number of people who want to see change can inspire and motivate rather than depress and discourage.
But even then messages still have a risk of backfiring and so no message should be communicated to the public before it has been systematically tested.
Only then can we make sure that when we try and do good, we do no harm.
(360info.org: By Nic Cheeseman, University of Birmingham and Caryn Peiffer, University of Bristol)