The Imagined School

“Of course it is happening inside your head, Harry, but why on earth should that mean that it is not real?”
J. K. Rowling, Harry Potter and the Deathly Hallows

A paracosm is a detailed imaginary world, often formed in childhood, which blends the real world with imaginary places, characters, and events. Children may develop a deep relationship with such imagined worlds, offering them a source of escapism and a means by which they come to understand the peculiarities of concrete reality.

Arguably, as adults we continue to operate in such imagined worlds, though they tend to become less fantastical and more grounded. We each carry around a small-scale model (or models) of reality within our minds. These models are representations of reality, not replicants. They are what we imagine reality to be.

As we develop, we might hope that our mental models enable us to function effectively in the world. Our models are tested continuously as we observe how our actions affect our environment. We make predictions which may or may not pan out, and so develop a theory for how we may act upon the world to achieve our goals. Our competency, in whichever realm we operate in, is determined by the utility of our mental models – the extent to which they allow us to predict, therefore to assert some influence over the world.

I listened to a fascinating podcast recently about the research into colour constancy by the neuroscientist Pascal Wallish. Wallish was inspired by the phenomenon of ‘the dress’: the image that swept social media a few years ago of a dress which, depending on how one’s brain interpreted the incoming visual data, could be ‘seen’ as blue and green, or as white and gold. What Wallish sought to understand was how this image could lead to such vehement disagreement: how we could all be so certain that our perception of the dress was the ‘right’ perception.

If you want to know the full answer to the question of ‘the dress’, you will have to listen to the podcast (although I’ll give you a little insight here). What I would like to draw out here is Wallish’s description of the mental models we hold, by which we make sense of our experiences.

Wallish suggests there are four layers in the hierarchy of our mental models by which we enquire about the world:

Description – We ask ‘what is it?’ We have a taxonomy available against which we compare what we observe (“that looks like a bird”, “this is a conflict”).
Explanation – We ask ‘why is it like that?’ We have causal models available which help us form explanatory hypothesis.
Prediction – We ask ‘what is about to happen?’ We project forward by drawing upon how we have seen events play out in the past.
Control – We ask ‘how can I influence events?’ We act upon the world with intent to create or control in such a way as our goals are met.

Mental models constitute your assumptions about the world. Wallish asserts that when you are confronted with substantial uncertainty – with data which cannot be made sense of by your existing assumptions – this causes cognitive dissonance.

How does the brain react to this uncomfortable confusion?

Wallish suggests that our mind disambiguates an ambiguous situation without our conscious awareness. In other words, it provides us with more certainty than we have a right to feel. It does this by inferring what ought to be there, not what is there. For example, we see the world in three dimensions, yet we have no three-dimensional data entering our brain. This facet of reality is constructed by our brains so that we may operate in a three dimensional environment. The mind infers this extra dimension: it imagines it to be there so that we can function – so we don’t keep bumping into things!

This reflexive action serves a very useful purpose, but it can also deceive us. It helps explain why we feel so certain that the dress is blue/green or white/gold, when the data entering our brain is ambiguous. We build this certainty on ‘priors’: our past experiences which determine the way we perceive the present. (For those interested, this is how we explain ‘colour constancy’ – the tendency to see objects as the colour we ‘know’ them to be even when colour data is not available. For example, in near-dark we still ‘see’ a banana as yellow, even though we are not receiving this light frequency.)

Substantial uncertainty. How often are we exposed to such ambiguity and how often do our minds do their work by inferring what ought to be there?

This idea was also explored by Frank Keil, termed the Illusion Of Explanatory Depth (IOED). Keil argues that ‘humans are driven to seek useful explanations of their environment’ but that ‘partial explanatory understandings may induce the compelling sense of knowing more than they really do.’ Keil is referring to the problem of weak mental models and the false certainty this creates. This is most likely to occur when confronted with a complex, multi-faceted reality, which cannot be accurately captured by simple heuristics, and when there are aspects of this reality which are more readily observable and easier to understand. In other words, we notice the noticeable and the understandable. From this we build a theory, filling the gaps with inferences from our past experience as some of the data we need is hidden from us, or beyond our comprehension.

There is a risk, when confronted with substantial uncertainty, that we construct our own paracosm: an imagined world which only partially reflects the complex reality, but leaves us more certain than we have the right to be.

And all this is what I wanted to say when I took part this week in a discussion about leadership development on Clubhouse (the trendy new social media app). Of course, this is far too niche and specific for a debate on how to develop school leaders! Besides which, the focus of the discussion was about the notion of expertise and it was a great opportunity to debate this with Dr. Amanda Goodall and Tom Rees, both highly knowledgeable in this field and more articulate on the subject than me.

But what has all this got to do with leadership development or expertise? A great deal, I think, for what is developing leadership expertise if not the growth of ‘better’ mental models? By ‘better’, I mean models which are sufficiently representative of reality to allow accurate predictions and foster wise action.

It seems to me that the gap between how the school is and how the school leader imagines it to be is where poor leadership thrives. Leaders’ decisions are made with reference to the imagined school but play out in reality. It is in our interests to assist leaders in closing these gaps.

I believe that this is a fundamental problem for school leaders, and therefore for leadership development. There are a number of forces acting upon school leaders which compel them towards the creation of an inaccurate representation of the school and a false certainty about their insight. Firstly, leaders are at a distance from classroom complexity and therefore reliant on proxies and samples for estimating the success of the school’s core business. Whenever leaders sit and examine progress data, aggregate observation judgements, or paw over pupils’ work books, they risk the illusion of explanatory depth – of noticing the noticeable and the understandable and filling the gaps with assumptions. Secondly, leaders, under pressure of time and pressure to act, can assume their perspective is universally shared or somehow privileged. Whenever leaders state ‘this is how I see it’ without stopping to wonder whether there might be equally valid ways to see ‘it’ (the dress) differently, they risk over-confidence. Thirdly, the feedback leaders receive on their actions may falsely reinforce their notion of the imagined school rather than cause them to question their assumptions, bringing them closer to an accurate depiction of reality. Whenever leaders assume that their predictions come to fruition, seek out only confirmatory evidence of an initiative’s success, or turn their back on dissenting voices, they reinforce their illusions and maintain the fantasy inherent in their paracosm.

If we were to observe school leaders in their attempts to improve their school, how often would we see them acting on the imagined school – the abstraction rather than the reality? This is a more comfortable place to operate. Leaders may leave a meeting knowing that they have pushed for better numbers on a spreadsheet, finish giving lesson observation feedback knowing they have suggested doing more of whatever technique is backed by research, or be satisfied that the new curriculum plan is better sequenced than the previous version. These imprints may or may not lead to actual school improvement, but the leader can take comfort that the imagined school is moving forward nicely.

Given this line of reasoning, the scope for improving leadership development is threefold:

Supporting school leaders in developing better descriptive, explanatory, and predictive mental models, and;
Encouraging, and equipping, leaders to better understand the nature of the challenges the school faces, and the various ways these challenges can be perceived, and;
Helping leaders appreciate and seek out useful and accurate feedback which provides leaders’ with insights which make future action more informed and effective.

We want school leaders to effectively act upon the school, not an illusory version of it. The imagined school – the model leaders hold in mind – should as far as possible reflect the nuanced, ambiguous, uncertain, reality of the school.

For such development to occur, participants must feel psychologically safe. Trauma, excessive psychological pressure, or risks to our self-esteem, push us towards comfortable fantasies rather than an acceptance of the flaws in our imagined worlds. Any attempt at leadership development must pay due regard to the need to create a safe space to explore our deficiencies and illusions.

Which leaves the question of how we will know if our attempts to develop leadership expertise have been successful. Pascal Wallish suggests an answer to this. He argues that mental models are invisible to us, therefore to attempt to judge their efficacy we must look to whether our visible successes suggest that we truly comprehend the complex reality upon which this success is built. He provides the example of humans’ ability to harness the power of the atom and the evidence this provides for the sufficiency of our knowledge. Rather than attempt to interrogate the ‘truth’ of our understanding of nuclear physics, or the sophistication of our mental models, instead we can point to the evidence of our ability to control and create in this domain. Our ability to achieve our intent is the evidence we need that our understanding of a complex endeavour has advanced. Similarly, we may look to the success of school leaders in steering their schools towards meaningful organisational goals as evidence of their expertise. But we must be careful that what we observe is not the illusion of improvement: a shift in the imagined school rather than an improvement in the lives or capacities of children.

We all live within our own imagined world, but we operate within the real world. However, we can only improve the real school if we work on our imagined one. This is what I wanted to say about leadership development. This is how I imagine we can improve it.