When I came to work for the Dana Foundation in 2006, one of the innovative projects it was funding was called Arts Education, a series of pilot programs teaching artists how to be effective teachers in the classroom. Like most of our seed programs, the funding was short-term, to get the programs up and running and give enough time to prove their worth so other, bigger grant-givers (or government agencies) would fund them.
To me, the program didn’t seem to be on Dana’s regular track in 2006. Most of our grants targeted brain research, in labs and in hospitals. But then I learned that two years before, the foundation had called a meeting with neuroscientists from seven universities to try to design studies that might figure out why learning the arts is associated with higher academic performance. If science could prove some direct links, the reasoning went, perhaps the people who funded schools would stop cutting arts programs.
The researchers in the Dana Arts and Cognition Consortium, headed by Dana Alliance member Michael Gazzaniga, met and returned to their labs and did some solid basic behavioral research. Four years later, in 2008, we collected the results of their studies into Learning, Arts, and the Brain: The Dana Consortium Report on Arts and Cognition (web version; there’s also a PDF version). Though the report is now five years old, it is still one of our most popular pages on the site.
This was a project close to the heart of our chairman at the time, the late William Safire, a lover of both arts and politics. He wanted a strong case to argue for more arts funding, and pushed the researchers to keep going—and to do it right, sciencewise. When the results came in, they were suggestive but not definitive: Researchers could describe links but not prove a direct cause. This is true for nearly all research: Most science is a slow build toward knowledge.
For us in the communications department, it led to a debate: What could we actually say, in a press release and elsewhere, that would get people’s attention yet be true to the science? With Safire, words were important; precision, everything. We bounced back and forth between “Study hints at” and “Study suggests,” but then again, they really did find stuff, not just hint at it. So we settled on “Study finds strong links between arts education and cognitive development.”
At a neuro-education workshop later in 2008 in Baltimore, I listened as Safire and one of the consortium researchers, Elizabeth Spelke, parsed the words even more closely. Spelke’s lab had investigated the effects of music instruction on cognitive systems underlying math and science. The results “show that intensive music training is associated with improved performance in the core mathematical system for representing abstract geometry.” There were many caveats: sample size was small, they had not investigated less-intensive training, etc., etc. Safire pushed Spelke to say this result was promising; she stuck with “needs more research” (I’m paraphrasing).
Spelke and her colleagues continued that research, partly funded by Dana. This, month they released the report, “Two Randomized Trials Provide No Consistent Evidence for Nonmusical Cognitive Benefits of Brief Preschool Music Enrichment.” You can see how carefully they worded that title: “no consistent evidence,” “brief,” and “enrichment,” not instruction. Still, the press coverage (and, some might argue, the initial press release) used words with far less care. “Music lessons will not make your children smarter, says Harvard University,” “No, musical training won’t make your children more intelligent,” “Childhood Music Lessons And Brain Power Aren’t Linked, Study Suggests.” Maybe, maybe not.
The study’s lead author, Samuel Mehr, took to Reddit to counter the exaggerations (he explains the results and potential implications in a video). He goes into great detail on methods and results, answering the many solid questions Redditors asked:
Experimental studies, whose results are based on statistical hypothesis testing, cannot affirm a null hypothesis. So, the real conclusion, which we were careful to outline in our paper, is that we have no evidence in these two studies to support the idea that a short period of parent-child enrichment classes increases preschool children's cognitive skills in the domains of numerical discrimination, receptive vocabulary, visual form analysis, or symbolic navigation (in comparison to either visual arts or no training). Of course, this is not a headline that will sell many papers!
Research continues, at Harvard and elsewhere. We recently reported on studies of music instruction and auditory processing, “Past Music Training May Help Compensate for Some Age-Related Declines” (short-term lessons in long-ago past), “Music Offers a Boost to Education in Low-SES Environments” (2-year class instruction in high school), and we have one coming in January on visual arts and neuroplasticity. The evidence, pro and con, continues to build.
But I think it would be Point 2 in Mehr’s video (this argument doesn’t seem to be working) that would have resonated strongest with our late chairman. We’ve spent decades now building the science-based case for arts in the schools, and arts funding has continued to decline. Science arguments in general have not held much sway in political decisions lately. Perhaps it’s time to try a new tack in our arts-funding campaigns.