The US Congress has told the National Science Foundation (NSF) to coordinate research on intelligence matters. But critics say that if the research agency acts as instructed, it will violate its tradition of openly disseminated science.

The Intelligence Authorization Act for 2004, signed into law on 13 December, calls for the NSF to hold two workshops to coordinate research in the behavioural, psychological and physiological sciences for the purpose of government security evaluations. NSF officials are also required to head a committee of experts in defence, law enforcement and intelligence advising the government on research into security screening procedures, such as lie-detector tests.

The act specifies that the workshops and panel are exempt from government “sunshine laws”, which would open the meetings and documents to public examination.

“This is at odds with an NSF-declared commitment to openness,” protests Steven Aftergood, who directs the Project on Government Secrecy at the Federation of American Scientists in Washington. Such exemptions are increasingly routine in new legislation related to security, he says.

The NSF and the White House Office of Science and Technology Policy would receive US$500,000 from the director of the Central Intelligence Agency to carry out the workshops and committee work. Because the final budget for intelligence is classified, officials were unable to say whether this expenditure is going ahead.

The NSF already works in areas related to homeland security. Its Approaches to Combat Terrorism initiative, for example, has spent roughly US$3.5 million on research for the intelligence community. But all this work is openly published.

The Senate intelligence committee is said to have inserted language mandating the workshops following a study by the National Academy of Sciences last year. The study had called for a research programme on security screening that would “operate under the normal rules of scientific freedom and openness to the extent possible while protecting national security”. A committee spokesman was not available for comment.

NSF officials say the programme will not violate the agency's commitment to openness. “No one at the NSF imagines that this is calling for some kind of closed or classified workshop,” says Curt Suplee, who directs the NSF office of legislative and public affairs.