Salem-Keizer spent about $400,000 of stimulus money for special education to buy the program.

SALEM — Valentin De Leon, 9, recently matched the word "sneak" with the picture of a snowman on his computer screen, because the two have the common "sn" sound between them.

The third-grader at Hallman Elementary School has spent a half hour each school day in recent weeks playing the computer game, an education intervention. Fast ForWord is being piloted at a dozen Salem-Keizer schools, most of which serve low-income students.

Salem-Keizer Superintendent Sandy Husk chose the program after reviewing research about brain development — that the same concept that allows stroke victims to rehabilitate and rewire their brains can help children who struggle with reading.

"I think we've just begun to scratch the surface in connecting what the medical field knows about brain development and applying it in the educational setting," Husk said.

Scientific Learning Corp. claims that students will gain one to two years of reading in eight to 12 weeks, but education research has showed mixed results.

The What Works Clearing House, a part of the U.S. Department of Education that reviews education research, says that it "was found to have positive effects on alphabetics and mixed effects on comprehension."

Salem-Keizer spent about $400,000 of stimulus money for special education to buy the program, which is not being limited to those students. More than 850 students in the district have used the program so far.

The cost includes software, servers and equipment such as headphones, as well as training, said Shawna Moran in the district's curriculum and instruction department.

Fast ForWord developers say the program draws on research about brain plasticity, the idea that the brain changes while learning new skills. It intends to improve cognitive skills such as memory, attention, processing rate and sequencing.

Yet other research has questioned how effective the learning games are.

A randomized and independent field test of the program showed that it generally did not improve second- and seventh-graders' language and reading comprehension test scores, which was reported in "Education Evaluation and Policy Analysis." Some problems were experienced in implementing the field setting, it said.

"There's really no evidence that this kind of intervention would have the kinds of impacts that the developer claims on students who are just more typically at risk at poor performance," said Geoffrey Borman, professor of education the University of Wisconsin-Madison.

"This intervention was originally designed for students who are experiencing very specific language difficulties," he said.

However, studies by Scientific Learning using various school district data show student growth. The company has more than 220 studies about the product to date, said Jessica Lindl, senior vice president.

In Clarke County School District in Georgia, for example, limited English, special-education and low-income students showed gains after using the program.

In 2006 about 40 percent of students who were not proficient became proficient the following year, compared with 27 percent in a comparison group, the study says.

"There aren't products out there like this," Lindl said. "Most curriculum products try to teach reading. We focus on building a better brain and building the capacity."

Salem-Keizer staffers are planning to conduct their own review of the program this summer, using student reading assessments and state tests results this summer, Moran said.

A focus has been proper implementation of the program, Husk said.

"If you try to do an implementation that has not got the rigor that you need to ensure the teachers and students understand the importance of consistency and amount of time spent, you're going to struggle," Husk said.

Built-in testing that comes with the software shows students are gaining, on average, one year and three months, Moran said.

"It's hard for me to report an impact yet," she said of the district's independent review. "So far, the students (that we have tested) are showing growth."

Houck Middle School Principal Sue Rieke-Smith has been so pleased with the program's results with special-education students that she's expanding its use to struggling English language learners and hopes to take it schoolwide in the fall.

The school is aligning the computer program's interventions with English language development standards.

"If we know a kid is missing a particular standard (we can) go to that particular product," she said. "It's trying to develop a standard of care for these kiddos."

Hallman teachers said the program's effect appears in hard-to-measure areas such as engagement.

"I do see growth. I really like that it individualizes their learning," said Bridget Crorey, a teacher at Hallman. "In the classroom, a lot of the time it's hard to make every minute count for every single student, especially in a dynamic room where you have kids at every level."

Yet some say the effect is not as clear.

"It's hard to see the direct impact to class ... it's brain based," said Julie Wojcicki, a teacher at Hallman. "I can't say specifically, 'They can read 25 more words.'"

But, she added, "They are making more connections in the classroom."