#1198 1943 · Office of Strategic Services (OSS) · Intelligence and espionage
The OSS stopped interviewing spy candidates and staged a fake crisis to watch them break
the problem
Interviews and credentials kept passing agents who froze or cracked under real field pressure abroad
background
By 1943 the OSS had sent hundreds of agents through Europe and Asia trained on paper credentials, language skill, and a conventional interview board, and field reports described competent-seeming recruits who broke down, blew their cover story, or froze once real danger appeared. Nothing in a resume or an interview room predicted who would hold up alone in occupied territory.
Harvard psychologist Henry Murray was brought in to build something else: a three-day live-in program at a Maryland estate known only as Station S, where candidates operated under assumed names and false biographies for the whole stay, so assessors could watch how a person actually behaved under invented pressure.
what everyone would do
Screen harder on paper: tighten the language tests, add a psychological questionnaire, lengthen the interview board — more of the same instrument that had already let unfit agents through.
what they saw
Murray saw self-report fails exactly where it matters: a calm interview cannot show whether someone stays calm under real pressure. So the team built the pressure and scored the behavior, not the claim.
the move
Station S ran candidates through a simulated ordeal — 12 minutes to build a cover story before hostile interrogators shouted questions under a bare lightbulb, a locked-room escape under time pressure, and a construction task where two actor-assistants were secretly scripted to be lazy and insulting so assessors could watch how the candidate led an uncooperative team.
why it works
A live simulation removes the gap between claimed and actual behavior because the candidate cannot rehearse a performance for a scenario built to be unfamiliar and socially uncooperative — the scripted difficult assistants and hostile interrogation were designed so interview polish could not carry a candidate through; only the underlying trait could.
the payoff
Over 5,000 candidates ran through the stations; the staff's 1948 study found the simulations caught breakdowns no interview had surfaced.
where it breaks
It only works when the simulation is a faithful analogue of the real pressure, not a generic obstacle course — a fake crisis that misses what the job actually punishes teaches assessors nothing. It is also expensive: three days per candidate and trained actors, which is why most organizations still default to the interview it replaced.
what came after
Station S became the founding document of the modern assessment center, judging people by staged simulations of the job rather than by what they say about themselves; the CIA inherited it directly, and corporate assessment centers built through the 1950s and 1960s trace back to this program.
references
- [1]How OSS Operatives Were Assessed and SelectedSOFREP, 2021sofrep.com
- [2]Assessment of Men: Selection of Personnel for the Office of Strategic ServicesOSS Assessment Staff / Rinehart, 1948backcreekbooks.com