But three key ingredients need to be present - the right curriculum, teacher training and software, according to Jeremy Roschelle, director of the Center for Technology in Learning at SRI International, formerly part of California's Stanford University.
He told a numeracy conference in Glasgow, organised by Learning and Teaching Scotland, that, although American pupils' attainment in maths had improved steadily over the last 20 years, it had not done enough to make a difference.
"Children are still at the basic level of skill," Dr Roschelle said. "They are not yet proficient or advanced."
His study employed a software programme called SimCalc, for wireless hand-held computers, to teach 12-year-olds from a range of backgrounds in Texas how to tackle concepts such as proportionality in maths - a concept critical for the transition from middle school to high school mathematics and science. It cuts across the curriculum - from Newton's Law to chemistry.
The research involved 1,600 students and 95 specially-trained teachers. Schools were chosen randomly, either to use the computerised materials or their usual textbooks to teach rate and proportionality.
The SimCalc materials were based on interactive software which shows, for example, two football players running across a football field to a fixed point. The software helps students understand concepts of proportionality, linearity and rates of change; it gives them a better understanding of graphs and tables.
Pupils taught with conventional textbooks improved their Texas state test scores for maths by 19 per cent; those using SimCalc gained an average of 46 per cent. When the experiment was extended to eighth graders, or 13 year-olds, and focused on linear functions, a similar effect was found.
Teachers using SimCalc reported that they had more complex teaching goals, and their students reported that they learned more.
The results improved for pupils from a variety of backgrounds, including Hispanic children who have traditionally fared badly in academic attainment, Dr Roschelle said, so the approach "spoke to equity".
The lessons drawn from Jeremy Roschelle's research for Scotland's curriculum developments were:
- there needs to be a major focus on continuing professional development;
- there are lots of kinds of technology but they don't all work, and some even have a negative effect;
- the right software can play a key role by making learning more dynamic, and allowing students to visualise things that would otherwise be more difficult.