The Engineering and Physical Sciences Research Council's "shaping capability" agenda may have been the catalyst for the protest involving around 100 scientists delivering a coffin to 10 Downing Street last month. But the Science for the Future group's claim that the EPSRC is precipitating the "death of British science" was motivated just as much by its axeing of PhD project studentships to make way for Centres for Doctoral Training (CDTs).

The EPSRC established around 50 centres in 2009 to train small four-year cohorts of around a dozen students in "highly innovative, research-excellent environments where both depth and breadth are championed".

Centred around specific EPSRC priorities - the digital economy, nanoscience and energy - the centres "bring together diverse areas of expertise to train engineers and scientists with the skills, knowledge and confidence to tackle today's evolving issues and future challenges", the research council says.

They also aim to "create new working cultures, build relationships between teams in universities and forge lasting links with industry". To that end, CDT students are trained in technical and transferable skills, and are co-supervised by industrialists.

ÖйúAƬ

The cohort approach to doctoral training is certainly popular with funders, and similar approaches are being adopted by a number of research councils. But according to Adrian Sutton, chairman of the CDT in Theory and Simulation of Materials at Imperial College London, the approach is 60 per cent more expensive than traditional doctoral models. And in an era of declining real-terms budgets, that inevitably meant something had to give.

That "something" was project studentships, and a parliamentary question last year revealed that their loss would result in the EPSRC funding 1,000 fewer students in 2011-12 than in 2010-11.

ÖйúAƬ

Tony Barrett, head of synthesis at Imperial and the leading light behind the Science for the Future campaign, said he was dismayed by what he sees as a severe diminution of principal investigators' ability to recruit doctoral students for "exploratory research programmes", which are "where new ideas and innovation emerges".

He was also scathing about the restriction of CDTs to research council priorities. He believed such "excessive top-down control" was more "appropriate for Honecker's German Democratic Republic in the 1970s" - the research record of which compared poorly with West Germany.

Professor Barrett said that the centres would devalue UK doctorates by diluting the requirement for candidates to make an "original contribution to knowledge", while the time spent on developing "soft skills" was "an appalling waste of young scientists' talent and an utter waste of tax revenue".

He also had concerns about the prospects for research quality and doctoral recruitment at institutions without CDTs: "It is notable that a select group of universities have done very well out of the original call simply because they were quicker off the ground - and there have been no more funded since the original tranche."

But a recent mid-term review of the "progress and impact" made by CDTs is extremely positive.

The independent review, published at the end of March, says a majority are making "good progress or better". It adds that their approach is "an effective way of training a cohort of students", with the additional benefit of acting as a "catalyst for bringing people together" and a means to leverage "substantial industrial funding".

Nonetheless, the report also notes that there is a continuing need to monitor the centres' progress and to share best practice among them. The most successful appear to be driven by "strong involvement from a team of academics and staff".

"It does take someone passionate to drive a good CDT," Professor Sutton agreed. "The experience is not necessarily uniform across the different centres."

ÖйúAƬ

ÖйúAƬ

He applauded the CDTs' efforts to instil cultural, political and social acumen - the absence of which in standard PhD graduates employers often lamented. As a result, industrialists were already "lining up to employ our people".

Alwyn Seeds, head of the department of electronic and electrical engineering at University College London and director of the joint UCL-University of Cambridge CDT in Photonic Systems Development, said he believed his students were exceptional, noting that they were "publishing important papers in the first year of their PhD degrees", uncommon in other models.

But Alison Rodger, who directs the CDT in Molecular Organisation and Assembly in Cells at the University of Warwick, noted that the students' excellence may merely be because the centres attract the most able candidates to begin with. Her own efforts to collect evidence proving the case for CDTs had floundered because it was "just about impossible to work out what the control group is".

A Wellcome Trust programme in the 1990s piloting four-year degrees in neuroscience at UCL bore significant similarities to the CDTs. The trust compared the productivity of students who had taken up places with those who had been offered them but had preferred regular programmes. It found that nine years after starting their PhDs, the four-year cohort had published twice as many papers as those on three-year courses at UCL or elsewhere.

However, the EPSRC does not plan to record the CDT students' publication or employment-destination data. Nor are there likely to be a statistically significant number of students who turn down places at the centres, Professor Rodger noted.

Two-tier fears

Another criticism of CDTs, particularly from students, is the perceived creation of a two-tier system in doctoral training, with those outside the centres feeling like second-class citizens.

Professor Sutton sympathised, but said he believed the best practice created in CDTs was spilling over into other areas of doctoral training, benefiting all postgraduates.

"What we're doing is raising the bar for the whole of postgraduate education by being creative and innovative," he said.

Professor Rodger is less confident of the inevitability of such spillover, noting that it had taken "a very high level of commitment from me personally" to drive the implementation of CDT techniques across Warwick.

She also acknowledged that the multidisciplinary projects pursued by CDT students would "never have the depth" of those pursued by standard PhD students. But she remained "fanatically in favour" of them.

ÖйúAƬ

"Other models of training work very well, too, perhaps for different people and different areas of science," she said. "But I favour the CDT model for anything crossing disciplines since it facilitates the creation of communities. The training add-on more than justifies the extra cost and, personally, I always involve any single-discipline students I have in a CDT structure."

paul.jump@tsleducation.com.