Years ago, Scientists proved years ago that toddlers who could wait longer than their peers to eat a promised reward of marshmallows were more likely to grow into more successful and better-adjusted adults. Now, the Marines may be looking to adapt "marshmallow test" science to help them screen for more resilient and determined officer candidates.

Late last month, Marine Corps Recruiting Command distributed a solicitation asking for research and studies to help them identify predictors for success at Officer Candidate School and The Basic School that follows. The document explains that officials have found consistent attrition patterns with current recruiting methods, which focus on a prospective candidate's grades, physical fitness scores, mental test scores and physical fitness performance, and grades, as well as accomplishments that demonstrated qualities such as integrity and leadership. Historically, anywhere from one-quarter to one-third of officer candidates do not make it through OCS. integrity and service from past accomplishments.

"To refine the recruiting process to produce higher quality officers while maximizing efficiency is a task that will require ingenuity and an adaptive force," the solicitation notes. "Obtaining a non-cognitive test that narrows the margin of error when selecting a Marine Corps officer candidate could potentially change the dynamic of the officer recruiting mission."

The new concept of non-cognitive testing for the Marine Corps is closely linked to an initiative in the Commandant's Planning Guidance, published by Gen. Joseph Dunford at the start of this year.

There was room for improvement in the Corps' recruit-screening processes, Dunford wrote in the document. He who pointed to psychological evaluations used by law enforcement, industry, and even special operations forces. He directed officials to assess available the tools currently available, screening the best for future use by Marine recruiters.

Recruiting Command spokesman Maj. Stuart Fugler said the search for non-cognitive testing had originated with the planning guidance. MCRC officials had pursued the commandant's directive with consultation from the Marine Corps Force Integration Office, he said.

While the current "whole person concept" of screening potential officer candidates wasn't causing high attrition rates, Fugler said, officials had observed limitations to the effectiveness of their conventional methods.

He said setting tougher standards for entry into the highly competitive OCS program did not cut attrition. In addition to slashing the OCS washout rate, Fugler said, Marine Corps leadership sought to use non-cognitive testing to better assign jobs and identify and retain top talent in the officer ranks.

"We observed high competition for attendance at OCS and increased quality indicators did not correspond to reduced attrition at OCS," he said. "We seek to not only reduce attrition at OCS but also better assign [military jobs] while identifying and retaining top talent within the officer corps."

How to develop the right tool to do that is another question entirely.

While the term "non-cognitive" is unpopular with psychologists and researchers, it refers to tests that measure qualities such as impulse control and grit, or resilience, character aspects of – all of which produce better-adjusted people. It's a relatively new field of study, and finding effective measurements of ways to measure these intangible qualities has been a challenge, said Dr. Daniel Willingham, a professor of psychology at the University of Virginia who has written on the topic.

A May study by field experts Angela Duckworth and David Scott Yeager described the benefits of self-reported and observer-reported questionnaires to identify determine these character qualities. But there are problems with these tools, they wrote, including misinterpretation of questions, lack of insight, and dishonest self-reporting.

Another option might be performance tasks that reveal character through the way a prospective candidate attempts a particular challenge, though these would have to be a lot more complex than a "marshmallow test" to work with Marines.

"One of the things that you intuitively know is that people get better at self regulation when you get older," Willingham said.

At this point, the Marines aren't offering any specifics about what sort of tools they're exploring. How this non-cognitive testing pilot program is rolled out to recruiting stations nationwide will depend on the options available and how accurate tests prove in forecasting success at OCS, Fugler said.

The timeline to adopt new tools, he said, is also unclear. It will depend on the success of the pilot program and the cost and usefulness of the tests.