"The use of a single generic question needs to be eliminated altogether, because it produces a situation where students are tempted to memorise responses," he said.
"Until this happens, New Zealand's supposedly 21st-century education system will continue to be operating on a 19th-century model."
Fountain's review found that generic questions have been used in NCEA history, social studies and media studies exams since the curriculum was relaxed in 2011 to let teachers choose any topic "based on their own interests with virtually no restriction".
Since 2011, the wording of the first question in every exam for Level 1 NCEA History unit 91006 has been, "Describe what happened in your chosen historical event."
In every year the second question asked for more detail about elements of the historical event which the student could choose, and the third question asked about the significance of the event.
Despite this generic wording, Fountain found that many teachers were actually still teaching only a narrow range of history topics listed on the official Te Kete Ipurangi website such as Māori leadership, New Zealand in World War II, and the changing role of women.
"The ongoing use of these tasks without adaptation has created a large body of completed work which could potentially be passed from one student to another, between years and even between schools," he found.
NZ History Teachers Association treasurer Greg Burnard said in June that memorising previous years' exemplars was "reasonably widespread across the country".
"Memorising an exemplar is not going to be punished, essentially," he said. "It's not seen as cheating, it's just seen as being well prepared."
Fountain recommends that NZQA consider publishing only "partial exemplars" for generic questions "which illustrate the skills required without providing a whole answer which could be rote-learned".
He says NZQA and the Ministry of Education should work with teachers to "develop teacher confidence in writing school-based local assessment tasks, reducing teacher reliance on the online sample assessments".
And he proposes that NZQA should "explore ways in which generic questions used to assess candidates in identified improvement standards can be varied from year to year".