Sep 17 2009

Evidence-Based Education

In my field of medicine, for the past century or so, there has been a concerted effort to connect scientific research (both basic and applied) to actual practice. What’s the point of having a lot of scientific knowledge collecting dust in journals and books if it doesn’t affect what people do? In the last couple of decades, in fact, the medical profession has re-dedicated itself to evidence-based medicine and even formalized this process. Some of my colleagues and I even want to go a bit beyond EBM to science-based medicine, to make the connection between science and practice more effective (and to oppose the counter-movement to disconnect practice from science).

But strangely in many other fields the connection between published science and practice is not strong. Richard Wiseman, for example, just published a book called 59 seconds (out in the UK, and will be released in the US in January) in which he points out that there is a rich psychology literature full of practical advice that is simply not getting to the public. Instead we have a self-help industry peddling advice which is largely made-up and not evidence-based. (You can hear us discuss this issue on a recent SGU episode.)

It is also generally true that the common knowledge that most people absorb from the culture is largely false, or at least off in some significant way. Think about any topic that you happen to either be an expert in or at least highly interested and experienced in and compare your knowledge to what most people believe about that topic.

One area where we could use more of a connection between science and practice is education. This is especially true as our knowledge of how the brain works and learns rapidly increases. What knowledge we do have is not systematically applied to education. It further seems to me that we need more translational research – studies focusing on long-term outcomes of specific interventions.

As reported by New Scientist, a recent symposium on the Decade of the Mind focused on just these issues. The conference discussed recent research in neuroscience and what it tells us about education. This is a topic I have been following for a while, although this is not my area of expertise, I have tried to distill the current big picture from what is being published.

One of the themes of the conference is that one key to learning is executive function – the ability to focus attention and for self-control and planning. Improve executive function, and you improve overall learning. (This relates directly to the ADHD posts I recently wrote as ADHD is primarily a disorder of executive function.)

The question then becomes – how do we improve executive function in students. We can break this question down to two broad categories – how to optimize executive function in the short term during lessons, and how to improve (if it’s even possible) baseline executive function long term. The former involves transient effects, and the latter permanent changes to the brain through plasticity and learning.

Short term there are lots of things that can be done in the classroom to optimize attention. The evidence shows that on average people can pay attention to a lecture or lesson for about 15-20 minutes – that’s it. However, there are lots of ways to re-aquire attention. The key is that students cannot be expected to be passive and attentive for longer than about 20 minutes at a time – so activities need to be broken up. Various strategies can be employed, such as asking questions to engage the students, telling jokes, using exciting imagery (but not too much or it becomes distracting), and pretty much any physical activity.

There is evidence that chewing gum helps attention, as does doodling. That doesn’t mean teachers should allow these specific activities in class (although that may not be a bad idea) but rather that some physical activity or novel task should be interspersed throughout more demanding lessons to help engage and revitalize attention.

I have two young daughters in school, and so I know that many grade-school teachers already know this just from observation (probably because young children are more obvious about their inattentiveness while adults hide it better). They let the kids run around or do physical tasks throughout the day to break it up. But in later grades students are expected to sit through long classes without breaks.

Aerobic activity prior to demanding lessons also helps – so perhaps gym needs to take place at the beginning of the school day. Playing music also engages the mind and recharges attention. However, there does not appear to be anything special about certain types of music – the “Mozart effect” is bogus.

These all seem like simple methods to maximize learning during the school day. But also as we apply these methods to achieve this, we need to do follow up long-term studies to see which interventions work best and what their real long-term effects are.

Achieving improved long-term executive function is a trickier question, and researchers are still parsing this out. There definitely is plasticity in the brain – function does change with activity and learning. But at the same time there also seems to be fairly genetically determined potential that does not bend easily if at all to intervention.

Specifically, overall intelligence seems to be dominantly genetic and we all (barring extreme conditions) will achieve our genetic potential eventually. The biggest variable seems to be how long it takes to get there, and that may be a function of effective parenting and education. But what can be modified is how we use our intelligence – making the most of what we have. The same seems to be true of specific cognitive tasks, like reading and mathematics – these are largely genetically determined talents.

Having said that, there is evidence that certain factors can permanently improve some aspects of cognitive functi0n (but not others). Learning to play a musical instrument or learning a second language both improve performance in non-verbal spacial tasks, for example. This suggests that these activities should be encouraged at a young age or perhaps become part of the standard curriculum. But to be clear, they don’t improve overall intelligence.

It seems, therefore, that we already have a significant evidence-base of what works and what doesn’t in terms of optimizing education from a neuroscience point of view, mostly focused around improving attention and effective executive function. While this knowledge needs to be translated into the classroom, we also need further research to more deeply understand not only how the brain works but how it learns, and how to think about and measure the various components of cognitive function.

In short, we need more science-based education.

13 responses so far

13 thoughts on “Evidence-Based Education”

  1. eiskrystal says:

    There are many teaching techniques out there which work and should be used.

    Unfortunately learning by rote, random facts/figures being thrown at you and disconnection from real world applications were how I was taught in most subjects and I doubt it’s changed much.

  2. Excellent piece Steve — and totally agreed. Education research has a very long way to go; there are far too many wishy-washy and po-mo studies.

  3. Also relevant: a piece in Science on the foundations of a science of learning and The Economist on using computer games in education (instead of chalk and talk).

  4. See also: “Foundations for a New Science of Learning” in Science:;325/5938/284

  5. nowoo says:

    This sounds a lot like the work that Nobel-winning physicist Carl Wieman is doing at the University of British Columbia in the Carl Wieman Science Education Initiative:

    He might make a good interview subject for a future SGU episode.

  6. DevoutCatalyst says:

    The less patient amongst us can order the book 59 Seconds in about that much time right now from; got my copy in July. Use your current login.

    Has it been observed that kids in a place like Singapore, where English is the main language, benefit from not only being raised multilingual, but from having their little hooks set in a whale of a tongue called Mandarin, the writing system of which alone requires much hard work on their part to accomplish?

    As for learning a musical instrument, go to Venezuela and see all the hidden talent that can be discovered in any given slum. Many children who would have been lost instead possess skills that would make a skeptic’s jaw drop, children who have caught the passionate attention of institutions such as the Berlin Philharmonic.

    Evidence based education should strive to bring opportunity to all minds, as a society’s ultimate candle in the dark.

  7. nohayes says:

    This statement…

    “There definitely is plasticity in the brain – function does change with activity and learning.”

    Conflicts with this statement…

    “Specifically, overall intelligence seems to be dominantly genetic and we all (barring extreme conditions) will achieve our genetic potential eventually.”

    … unless you specifically define “overal intelligence” as the degree of plasticity in the brain.

    There is no way to salvage the fact that the initial statement conflicts with this statement…

    “… like reading and mathematics – these are largely genetically determined talents.”

    Certainly, those abilities are facilitated by our “hardware”, so you could easily say that reading and mathematics are genetically facilitated talents, but once you introduce determinism into the discussion you have to deal with it’s consequences.

    If a neurological trait is determined genetically, education is unnecessary. If education has a causal effect on literacy or mathematic reasoning, then those faculties aren’t genetically determined. “Making the most of what we have” doesn’t really apply to education unless you have an objective way of knowing when a person has hit their wall.

    With that out of the way, you’re right. Our functional education system (being the environment that actually takes place rather than the environment we educators advocate) is based on an inaccurate model of cognition, consciousness, and intelligence.

    It may be of some comfort to you that “Brain Based Learning” classes are cropping up as requirements in education programs. The course that I took specifically focused on much of what you talked about, facilitating learning by accounting for and planning around executive function (Breaking up lessons, introducing information through a variety of stimulus, etc).

    The largest problem that I see as an educator, (I teach chemistry to 10th, 11th, and 12th graders), is the prevalance and polite dismissal of unthinking belief. If we are to produce students with a passion for science, a strong sense of skeptacism, and analytical ability we must be allowed to tell them that the other intellectual and moral authorites in their lives are (almost always) wrong.

    I recently heard a speaker decribe public education as follows: “Teachers are afraid of principals, principals are afraid the superintendant, the superintendant is afraid of the school board, the school board is afraid of the parents, the parents are afraid of the kids, and the kids aren’t afraid of anybody.”

    That said, there are many schools where intellectual freedom and scientific truth are valued. I am fortunate enough to work in one.

  8. nohayes,

    To clarify my statements there is a difference between potential, which for some cognitive traits seems to be genetically determined, and skills, which are learned.

    Someone may have high native talent in math and great potential, but that potential will never be realized if they do not study math.

    Overall intelligence refers operationally to standardized IQ tests. People seem to approach their innate IQ asymptotically as we age, perhaps achieving it at different ages but all eventually pushing up against our genetic potential. But again – this is NOT the same thing as knowledge and skills.

    At the same time, if we look at brain function we do see that training does result in changes through plasticity. Sometimes scientific evidence seems to conflict depending upon how a question is approached.

    One way we can put this together is that there are ranges of potential that are genetically determined, but then use allows for growth or atrophy within those potential ranges.

    Further, we need to distinguish hardware and software. The raw power of the brain with respect to different kinds of processing and tasks is only manifest through the knowledge and mental habits of the person. For example, someone might be inherently brilliant but filled with misinformation and locked inside an internal belief system. Good hardware – bad software. Another individual may have average overall intelligence but very well educated with logic and critical thinking.

    Another way to look at it is that we all have our individual learning curves. We cannot change our curves, but education determines how fast and how far we move along our curves.

    Again – parsing all of this is challenging and we certainly have lots to learn and clarify (and I do individually as well – I am still trying to wrap my mind around this extensive literature).

    What this all means for the educational system is that the system should strive for efficiency, optimizing learning, and allowing for each individual to achieve their intellectual potential, including creating lifelong learners.

  9. nohayes says:

    Dr. Novella,

    Thanks for the response.

    My problem with the principle of “potential” or “native talent” is that I’ve never seen an objective way to measure either one. We have assessment items meant to take a snapshot of currrent intellectual ability and, taken in series, we can get data on growth rates or “how fast and how far we move along our curves”, but we have no way to measure a theoretical maximum.

    I’m not saying there isn’t a maximum. We can be sure that there is, and that it’s genetically influenced, if for no other reason than because cognition is a caloric expenditure and is thus subject to limted resources.

    Regarding IQ tests, I was under the impression that one’s score ultimately boiled down to one’s performance against a statistical average… meaning that IQ relies on the bell curve rather than setting the bell curve. Am I way off?

    Additionally, you said, “People seem to approach their innate IQ asymptotically as we age, perhaps achieving it at different ages but all eventually pushing up against our genetic potential.” Could that phenomenon actually be a measure of the decrease in brain plasticity that occurs with age? In short, how do we know that people aren’t improving by smaller and smaller increments because of the impact of age on the brains ability to learn?

  10. Woody says:

    nohayes –

    Excellent point about an age-associated decrease in brain plasticity conceivably causing the “ceiling” effect on serial IQ measures.

    The literature on aging and cognition increasingly suggests that both genetic and environmental factors can adversely impact cognition earlier in life than previously suspected. It may be that if such factors could be modified, such cognitive decline could be avoided. Still a pipe dream at this point.

  11. mindsparke says:

    Hi, Steve.

    This is a fascinating discussion. There’s been quite a bit of interest in computerized training that strengthens working memory. This kind of training has been extensively piloted in the UK (Scotland), for instance, with very promising results. Intensive working memory training has also been shown to increase fluid intelligence (PNAS 2008), and working memory is a better predictor of academic performance than intelligence.

    But it strikes me that a fundamental question needs to be answered: Does the educational system remain in its traditional domain of improving crystallized intelligence — delivering information and training critical analysis skills? Or does it break new ground and venture into the field of improving fluid intelligence?

    (Many educators, for good reason, have skipped the question and headed right down the new path.)

    Martin Walker

  12. lizditz says:

    As a current graduate student in education, I can say that there are some good (i.e., well-researched) “brain in education” education classes, and there’s also a load of codswallop being passed off as professional development.

    The The International Mind, Brain and Education Society is an example of the former. For an award-worthy example of the latter, I refer you to the discussion of Dan Hodkins at various sites, summarized in Pseudoneuroscience and Professional Development in Education.

  13. jaycm says:

    Part of the problem is political. The cognitive sciences literature gives conclusive evidence of effects that many educators and educational politicians would very much like to deny. First and foremost is genetic determinacy in intelligence, and variation in median intelligence among politically-relevant subpopulations.

    These objections aren’t trivial. A school optimized for effective teaching would probably separate students by measured IQ, for example. Of course, in practice this would enrage parents of dummy-track students. The lowest tracks would almost certainly have disproportionate numbers of African-American students, causing even more concern.

    Politically acceptable ideas include “no child left behind” and “teaching to students with different learning styles”. My admittedly minimal nature of the cognitive sciences literature indicates that the first dooms classes to progress at the rate of their dumbest student, and that students do not demonstrate meaningfully different learning styles. But that’s irrelevant, because the institutional pressures are to ignore genetic limits and promise equality.

    In a nutshell, we compromise our student’s educations for the sake of our political values. Whether or not this is wise is debatable.

Leave a Reply