It’s human nature: People generally behave better when they know someone is watching. That theory applies to police officers who are arresting people, and to the offenders who are being arrested — even if the "someone" is a tiny audio and video recording device attached to an officer’s collar or lapel.

Police departments that require on-duty officers to wear these body-mounted cameras have achieved a drastic reduction in use-of-force incidents by officers, and in citizens’ complaints against the cops. The enforcement tool carries both economic and privacy pitfalls, but neither are insurmountable — especially as costs of the recording equipment and data storage decline, and public awareness of the benefits of this community-policing strategy rise.

The Honolulu Police Department, which suffers from a lack of transparency and accountability that continues to fuel questions about the conduct of some of its officers, should join the growing number of law-enforcement agencies nationwide that are testing this technology.

The best model is Rialto, Calif., whose police chief conducted a carefully controlled study of the measurable impact of having uniformed officers wear devices that record their interactions with citizens. In the first year after the cameras were put into use in 2012, the use of force by officers fell 60 percent, and citizen complaints dropped 88 percent.

Criminologist Michael D. White, a professor at Arizona State University and author of a U.S. Department of Justice report on the use of these devices by police, writes of their "civilizing effects," noting that in Rialto, Phoenix and other cities he examined, the cameras’ ability to collect transparent and objective evidence that will be made public changes the dynamics of potentially explosive situations.

Rialto’s remarkable results have inspired a surge in interest, most recently by the U.S. Border Patrol, which announced this week that it would begin testing the technology in October, having been accused for years of misusing deadly force.

White has estimated that about 5,000 of the 18,000 police departments in the United States are experimenting with the cameras now, and he predicts widespread adoption over the next few years.

Departments that implement the technology successfully will be those that create a community consensus supporting their use — engaging the public to explain the benefits, overcoming resistance from police unions, and ensuring that the cameras are used consistently and fairly, taking special care to protect the privacy of sexual-assault victims, children and others.

There are obvious downsides to recording virtually all interactions with police — confidential informants, for example, would no longer talk — so some exceptions would be required. But rather than using the pitfalls as excuses for why the technology can’t be tried, treat them as problems to solve.

The urgent need for improvements in HPD is obvious. The circumstances surrounding recent fatal shootings by police, earlier allegations of police brutality and claims that police had sex with prostitutes all would have been more clear had officers been wearing cameras; perhaps some of the incidents and accusations could have been avoided altogether.

A pilot project of body-worn cameras should be implemented at HPD as part of an overhaul that also bolsters independent, transparent investigations whenever an officer kills someone in the line of duty, or breaks the law off the clock, and legislative corrections that lift the secrecy shielding policedisciplinary records from proper review. It will take ongoing, systematic reform to restore trust in a department that sometimes seems above the law.