Can instructional software based on brain imaging help struggling students read better? That's the hope of school administrators who invest in cutting-edge computer programs like Fast ForWord (FFW), a highly respected family of reading-intervention programs distributed by the Scientific Learning Corporation (SLC) and used by more than 120,000 students in the United States. To answer this question, Princeton University researchers Cecilia Elena Rouse and Alan B. Krueger recently conducted a randomized study of reading intervention technology, using FFW as the subject.

Fast ForWord incorporates findings from functional magnetic resonance imaging (fMRI), a tool for observing changes in brain activity. The programs are based on research indicating that struggling readers often need help with phonological processing-for instance, learning to distinguish between sounds like /ba/ and /da/. The software programs use computer games that slow down and magnify speech sounds to help children process them more accurately.

Several studies have shown that programs like FFW yield impressive results. One showed gains in auditory processing after 8-16 hours of training, and a subsequent study found that 90 percent of students who completed FFW made a remarkable 1.5 to 2 years of progress in reading over 6-8 weeks.

The Princeton researchers tested FFW in four schools in one district. The study included more than 500 children in third through sixth grade; about half were assigned to the experimental group using FFW and the remainder were in the control group. All of the students had scored in the bottom 20 percent on the state's standardized reading test.

The researchers found that the state standardized reading scores of students in both groups improved by an average of between four and five percentage points, regardless of whether they used the software program or not; scores on three other measures showed similar results. "Overall," the researchers write, "our estimates suggest that while the FFW programs may improve a few aspects of students' language skills, it does not appear that these gains translate into a broader measure of language acquisition or into actual reading skills."

Rouse points out that FFW is a challenging program to implement and that this difficulty may also account for some of the disparity between her group's findings and prior research. Students using FFW must train with the software for 90 to 100 minutes a day over 6 to 8 weeks, she notes. "During the study, we had a professional development day, fire alarms—the usual things that disrupt a school day—which meant that the students had a much harder time getting in that 90 to 100 minutes a day," Rouse recalls. Also, she and her colleagues factored into their results the test scores of students who were not able to complete the program—data that previous studies of FFW had excluded. "What other researchers might say is, 'Any student that didn't complete the training in a specific period of time, you don't include in your analysis,'" she says. "But if some people can't complete the program, for whatever reason, well, that's part of the program."

Other researchers warn against turning intervention programs for struggling readers into curricula. "This particular program is an example of a larger trend—to take programs originally designed for kids receiving special education services, for kids with particular (and often low-frequency) problems, and turn them into programs delivered to all or most or many kids," says Catherine Snow, a professor at the Harvard Graduate School of Education who specializes in children's reading and literacy development.

Rouse and Krueger add that their study challenges the extent to which brain imaging can be used to design and evaluate reading instruction programs. The authors caution: "There is little evidence . that responses detected in brain images translate into relevant skills and behaviors, such as reading ability."

Based on the new findings, Rouse advises teachers and administrators to "ask hard questions" when deciding whether to invest in educational software, brain-based or otherwise. "Ask not just whether or not it works in general, in theory, but whether it is actually going to work with all the warts, etc., of a school day," she says. "And show me the empirical evidence.. From the fMRI on down, does it ultimately mean kids will be able to read better?"

Note: Officials from the Scientific Learning Corporation were not available for comment.