Squaring the Circle: Correlation Does Not Equal Causation
How can we square the circle between competing ideologies in education and the science of learning? The convergence is no easy feat – ideologies prejudice and evolve over time and are guided by vested interest; scientific findings and the truth of reality does not (or should not) discriminate at its own convenience, nor does it have to conform to the reality in which we desire. If understanding the truth about the science of learning is our objective, then it stands to reason that ideologies must take a backwards step in the decisions made about education.
I recognise that some people are of the opinion that scientific research is also ideological. On this I make two points. Firstly, as noted above, the science of learning in epistemically concerned with truth. As such, it ought to be subject to the same rigorous and systematic testing of any scientific research; it must also and also be open to falsification. Secondly, and more importantly, scientific research differentiates on the basis of methodology. I have touched on this point in several of my past blogs. These two ideas are not distinct but rather inform one another. If this distinction is correct, and I believe it is, then it is imperative to understanding the paradigmatic shift currently underway in educational research - from intuition to evidence. Falsifiability requires a methodology and strict criteria that gives credibility to ideas – these ideas, whether born out of intuition or experiment, must provide evidence, explanatory power, predictive success, and survive counter-evidence that would otherwise undermine its credibility. Generally, educational research has not upheld this rigor. This is indicative of the fact that education has been a hallmark of social scientific study and not scientific study itself. In other blogs I have shown how this has been exploited in the education system.
The fact remains, we must begin with the science of learning and build theories of teaching to compliment. These theories of learning cannot be ideological, nor can they be credentialed on the basis of intuitive based justifications – they will claim truth, but they lay no bearing on providing the necessary evidence it takes to claim something is true. This is in the abstract – let’s look at something more concrete.
A major paradigm in educational research has been Constructivism. Fundamentally, constructivism provides a theory of learning which suggests that humans construct knowledge based on an ‘active creation’ between prior knowledge and new experiences in the world. Constructivism has provided a gamut of theories of teaching that translate the theory of learning into an applicable theory of teaching. Just think of repercussions that this theory of learning implies: it see’s teaching of knowledge transfer as futile insofar as information can only be subject to interpretation; it advocates for teacher as ‘facilitator’ and dismisses ‘chalk and talk’ or direct and explicit instruction; it asks for ‘minimal guidance, ‘problem-based’, ‘real world’, ‘experiential’, ‘Inquiry’ teaching; it helps market and provide evidence for a whole range of educational products. Now I am in no way saying constructivism has no place in education – I hope to establish the opposite in good time. But it cannot lay claim to providing an explanation of the truth behind the science of learning, and it cannot do this for one very good reason – it uses intuitive justifications as explanations of how learners learn. I will put it another way: constructivism uses our biologically driven theory of mind to attribute intentionality and mental states to other people – just look at the discourse of ‘active agent’ and ‘meaning maker’ and so forth. This discourse is all based in Folk Psychology, and as I have stated previously, folk psychological principles must be reducible to scientific principles in order to provide an accurate and objective account of the science of learning. So is constructivism consistent with scientific findings? Here is where it becomes rather complex.
Constructivism as a theory of learning is highly unscientific. All the folk psychological principles outlined above cannot be reduced to scientific principles: ‘meaning maker’ and ‘active agent’ are not elusive but, rather, illusory. According to Eliminativists, this means the concepts need to be abolished. But we cannot forget that we are in the business of interacting with human beings – human interaction is not epistemically concerned with truth but is concerned with all the wonderful experiences and connections we make with our students and the joy that brings to our lives. Our students, to us, are not (insofar as we interact with them and they experience the world) merely neurons and electro-physical interactions. We attribute them with mental states for three good reasons: firstly, it is biologically determined that we do so; secondly, because we ourselves experience in a manner which folk psychology describes with a high degree of accuracy and we attribute the same experiential phenomena to our students through empathic relationships; lastly, because students experience the world and have intentions, hopes and desires. Irrespective of the truth behind the science of learning, we can never square the circle between the reductionist truths that science discloses and the folk psychological ideologies that give meaning to our lives and define our experiences in the world because the two are necessarily incommensurable.
On the other hand, there is plenty of ‘evidence’ to suggest that constructivism is conducive to educational outcomes. Whether or not it is scientific in its explanatory power, it can still provide a means of facilitating learning and conceptual growth in students - It is not an inhibitor to learning and as a theory of teaching is widely used in order to achieve educational outcomes. How can this be so? How can it be a successful theory of teaching without there being any credible evidence for it? It’s simple: constructivism is a good theory of teaching but a false theory of learning. Let’s explore this in more detail.
Correlation does not equal causation. So often does this provide an answer to seemingly apparent paradoxes and contradictions that it seems almost trivial as a response. Alas the correlation between constructivist theories of teaching and the achievement of educational outcomes cannot, in itself, be the cause of those educational outcomes. In other words, the constructivist theory of teaching itself is not the cause of the educational outcomes but approximates something that can be reducible to a scientifically understood principle of learning. Surely we arrive at misnomer? But this must be true as in the case that all claims to truth must be consistent with the laws that govern everything in this universe. So what next?
Even though students are not ‘active agents’, their experience in the world tells them otherwise: their efficacy (science of learning) in achieving tasks relies on the feeling of agency and independent inquiry (feeling of meaning making); students self-reflect on their learning (feeling of ownership), even if they have far less deliberate, controlled and linear control than it appears they have (science of learning); students feel they are thinking about problems (active agents) even if they are recalling memory and storing reformed memories (science of learning) – these are the narratives that students tell themselves, and teachers tell of students, in order for learning to take place. Despite this, the concepts will always be elusive to science because they are illusory and products of physical interactions in the brain. This is why, when all things considered, we need to build theories of learning and teaching that consider both the phenomenological and the naturalistic perspectives: the phenomenological will never be able to disclose the truth of the science of learning, but by the same token, the science of learning (even through reduction) will never be able to disregard the phenomenology of conscious learning experience, even if it can account for it. Both are needed in order to design better educational systems.
If everything I have stated so far is correct (as a scientific hypothesis it’s open to falsification) then it has a very interesting implication, which I will get to briefly. But my first point is this: I am sceptical of the ease by which translating theories of learning into theories of teaching is currently being done. Not only is it a great way of marketing products as ‘evidence based’, but it seems an overly superficial process (but nonetheless an intuitive one) by which a theory of teaching is designed to superficially embody on the theory of learning – pretty logical and simple. In Homo Economicus I stated:
“For Theories of Learning, we can use Behavioural Economics as an exemplar in the utilisation of naturalism as beneficial for enhancing an understanding of the underlying principles involved in decision making – in our case, educational not economic. For Theories of Teaching, there is an exemplary application of these understandings to enhance economic outcomes, or in our case, educational outcomes. These outcomes, however, come at a contentious and foreseeable cost: they are deceptive insofar as they utilise these understandings to enhance economic outcomes. The great moral question for education, however, would be whether we should use the same deception to enhance learning outcomes. Theories of Teaching, built on these principles, would look vastly different from the way in which we design Theories of Teaching now as we currently aim to provide a transparent and harmonious association between Theories of Learning and Theories of Teaching. Behavioural Economics does not disseminate its theories in manifestly transparent ways in which allow the individual to transcend their impulses – it exploits the consumer on the basis of deception in order to enhance capital gain. I am not sure what is to be taken away from this… perhaps I will return to this at some time in the future.”
If we are to take the science of learning seriously, just as behavioural economics took the science of economics seriously, then we necessarily cannot produce theories of teaching that ‘superficially embody’ the theory of learning – that would disregard the narratives and phenomenological perspectives of the learner which would have catastrophic implications for learning. Take for example Willingham’s thesis that students don’t like school because the brain is not built for thinking. Now this theory, consistent with scientific research, is a theory of learning: students rarely think when faced with problems but rather revert to memory recall as a ‘quick and dirty’ solution that conserves energy best spent on more pressing matters like survival or reproduction. If we built a theory of teaching that ‘superficially embodied’ this theory, not only would it be highly unethical but it would by highly demoralising and counterproductive for producing educational outcomes. It also, might I add, runs contrary to our folk psychological intuitions that we attribute to the learner too – despite knowing that the reduction of what we could perceive as “lazy thinking” is actually optimal and conducive to the survival of the individual, we still attribute the folk psychological principle of ‘disengaged’ or ‘lazy’ because we know that is a sign of ‘lazy thinking’ and ‘disengaged learning’. When communicating to the student it would be of no benefit to say that they are behaving in an optimal way as our measurement of success is learning and educational outcomes. We can’t let the truth get in the way of a good story – if we did it would become an impediment to learning.
So what then? How do we build a theory of teaching based on what we know about the science of learning if it can’t mirror the science of learning? Remember in Homo Educationist I stated economists deceive customers into thinking they have more agency in consumer choice than they really have? And in The Story so Far I referred to psychotherapy and the ‘talking cure’ as a means of ailing psychological conditions despite the correlation between improvement and lack of causation? The same must be the true of learning. In the same way talking deceives patients into thinking they have overcome their ailments; in the same manner behavioral economists deceive their customers into thinking they are making choices about products; teachers should be deceiving students in a manner that nudges them out of their default modes and engages them more deeply in their learning – and I say deeply in the purely folk psychological sense, whatever it means.
The science of learning has disclosed how learning takes place - memory recall is a great example of the same type of deception as noted in behavioural economics. You are not telling the students that their brains are mediating memory readjustment in the hippocampus – you are providing experiences for this to occur and utilising the narrative of phenomenology (or how this physical process appears in consciousness) to allow for this process to happen. I have said it before, and I will say it again: to build successful theories of teaching, we must begin with the science of learning. Only from here can we build theories of teaching that will truly and accurately utilise the science of learning to produce quality educational outcomes. We must overcome our disposition to build theories of teaching based on theories of learning that superficially model and are born out of logical and intuitive inferences. Translation ought not be a logical process but a deceptive one that converges two incommensurable frameworks (the phenomenological and the scientific) in a manner that enhances and compliments the principles that the science of learning has disclosed.