How can we ever be sure? Thoughts on climate, Parkinson’s disease and Let’s Think

Written by Stuart Twiss

Climate and the weather

 “The weather forecasters are always getting it wrong, it promised rain all day but look at this sunshine!”

“This is global warming, is it?”

The everyday descriptions of weather, forecasting and the causes of our experienced weather are riddled with misconceptions.  Weather forecasting is expected to be precise rather than probabilistic. We attribute the causes of a weather event to the weather itself and we’re often confused by climatic patterns that have a longer timescale and greater geographic reach than the moment by moment shifts in temperature and dewpoint that cause the rain that spoils our enjoyment of a sunny day.

As weather amateurs we can rely on the views of the ‘so-called experts’, or trust our own past experience and the sometimes accurate lore of a world before science. ‘Red sky at night’ thinking.  So we carry an umbrella, a coat and some hope!

As climate amateurs we might ignore the evidence of incremental change in global temperatures because of our immediate experience of the weather.  We might therefore hope that climate change will go away or even deny that there is any causal link.

The difference is that although an umbrella is adequate for weather it is a poor defence against climate!

Parkinson’s disease and the RCT

My mother had Parkinson’s disease for over 15 years before she passed away last October.  She had a slow progressing form of the disease unlike Philip Adey, one of the originators of Let’s Think who sadly lost his health so quickly.   My mother would have been very interested in the recent BBC documentary on GDNF, a nerve growth factor first identified from mouse tumours.  In patients such as Darren Calder it had profound effects, life changing and ‘miraculous’ but it also involved an expensive and risky process of administering the drug directly into the brain.

The current view is that such promising interventions be subject to larger scale Randomised Control Trials, RCTs.  This is a model of verification based on medical trials that has been broadly successful in identifying investment worthy drugs and procedures.  In an RCT as large a number of potentially willing recipients is gathered. The members of the group are randomly assigned to the proposed intervention or a non-intervention (sometimes a mock intervention such as a placebo).  After it is believed that sufficient time has elapsed for the treatment to have had the effect the results of both cohorts, treatment and non-treatment are statistically compared using the measures of independent evaluators to see if there is a significant difference.  This would seem to provide the necessary proof, uncontroversial evidence that something works, or does not. As such it has been employed within medicine for decades and within education in the more recent past.

There are however some significant issues with the approach, as good as it is.

  • For an RCT to be successful the treatment has to be uniformly and consistently applied in line with the hypothesis it is based upon.
  • The RCT relies on agreement about the intended effect and a shared view on appropriate measures.
  • An RCT ignores the positive and negative effects on individuals that may be based on their unique circumstances in favour of the overall impact on the group. Ignoring these circumstances is felt necessary in an RCT because a view has to be taken on whether the treatment can be replicated outside the trial.
  • Some interventions can only be conducted with small groups either due to recruitment or the cost or risk or disruption of the intervention.
  • Randomisation of smaller groups ignores the profound effects of differences between the two groups that could be balanced by stratification and more adequate blinding..
  • The endpoints in an RCT can be poorly selected, often with a preference for easily assessed endpoints.
  • There is little use of selection criteria for the individuals in an RCT trial and particularly responsive and other unsuitable individuals affect the results markedly, and especially so in smaller cohorts.
  • An RCT is inappropriate if there is insufficient sample size or power to make claims that would impact many future recipients.

Although the GDNF trial had a profound effect on some recipients like Darren it had an effect size overall that was inconclusive using the measures. Thus it was considered to have failed the RCT test.  But further research into the brain scans taken in tandem with the assessments show promise, even though these were not part of the original assessment regime.

Let’s Think

How are weather reporting, climate predictions and trials of Parkinson’s interventions informative for Let’s Think?

Let’s Think is an intervention that has a long term outcome, delayed over years, but is made up of numerous short term events and actions. Like our attempts to halt and reverse climate change it aims at a change that is incremental in the short term but which has profound benefits in the long term. Its proposal for teaching is based on a hypothesis of long term development and brain activity. It’s benefits are measured by cognitive gains.

Let’s Think addresses the challenging evidence that, despite apparent increases in academic success, there is an underlying decline in cognitive ability as measured by Piagetian tests. This is the ‘climate change’ that is happening in our classrooms, the unnoticed change in performance hidden by the annually reported ‘weather’.

Within Let’s Think there is a broad prescription for teacher activity to address the ‘climate change’ but the teacher, like a meteorologist is expected to use the many assessments and data points that come from her class to forecast the next most effective move in the ‘weather’ of the classroom.  Although the ‘weather’ of her classroom is the most important experience in the short term; long term it’s a change in the ‘climate’ that has the most profound effect on the child. The climate of challenge, co-operation and construction, reflected on together in a mixed class has to be created by each Let’s Think teacher out of the weather of everyday classroom life and it is this that addresses the underlying climate change, the decline in cognitive ability.

The Let’s Think Forum has gathered ample evidence of the underlying decline in cognition and also has evidence from many small scale and international trials of the positive impact on cognitive gains for children as young as 5 and as old as 14.  In particular from Let’s Think interventions in mathematics and science.

Our goal is to regularly replicate the positive effects of halting cognitive decline that have been seen in smaller and quasi experimental trials in larger scale trials.  We also aim to communicate and clarify the Let’s Think intervention so that it can be used with precision by teachers.

Previous Randomised Control Trials, RCTs, such as that for the Education Endowment Funded Let’s Think Secondary Science showed, like the Parkinson’s trial, no conclusive evidence of effect with the measures that were used.

Fortunately, unlike the Parkinson’s trial, there is very little risk in implementing Let’s Think in secondary science because in the trial taking away the ‘normal’ science lessons caused no decline in the gains of scientific knowledge in test schools compared to control schools. In addition the trial revealed that there were significant benefits experienced by teachers to their general ability to question and to promote engaging lessons.  Importantly it also revealed the difficulties experienced by teachers to develop a co-operative climate where students worked on challenges and reflected together on their efforts. Something we would want to work on especially in secondary settings with future interventions.