Saturday, 3 August 2013

Learning to play table tennis on a windy day, lessons for conservation

I have played table tennis from a young age and have received an amount of coaching.  The role of the coach is to ensure that the player does not improve their stroke in a manner which is beneficial in the short term but limits the potential of the player, the player is responsible for fine tuning the prescribed technique, gaining ‘feel’ in the process.  In this blog I will show that controlled, non random, conditions aid an individual’s attempt to gain feel, or its intellectual equivalent, expert intuition.  In a random environment the quality of feedback is reduced and the process of gaining intuition is disrupted.  Conservation is practiced in a random environment meaning that the intuition of individuals, resulting from their own experiences should not be trusted.  I will consider the consequences of this conclusion and whether a role analogous to that of the coach exists in conservation.

When purposefully practicing (training) with the aim of improving their stroke, a player will try to isolate one aspect of their stroke technique e.g. the extent to which one uses their wrist to generate power and spin.  Small adjustments will then be made to that aspect of the stroke and the player will take note of the relationship between the adjusted technique and the outcome (the flight of the ball).  Feel will be gained as the player gains positive feedback from the technique which results in good shots. 

Now suppose that the player is playing outside on a windy day.  The wind changes in direction (blowing either towards the player, their opponent or not at all) randomly from shot to shot.  A player trials a technique and plays a shot.  The ball flies into the net.  There are two possible explanations: 1) the technique was a bad one or 2) the technique was good but the wind caused the shot be a bad one.  Similarly, if the ball goes onto the table then there are two possible explanations: 1) the technique was a good one or 2) the technique was a good one or 2) the technique was bad but the wind caused the shot to be a good one.  The player would be much less able to gain feel and would have to trial each technique many more times to gain the same level of feel. 

There are many disciplines in which feel can be trusted, disciplines in which shot and outcome are closely linked in a non random environment (see chapter 22 of the fantastic ‘Thinking, fast and slow’).  If conservation was practised in such an environment it would not be a problem that, as Sutherland observed in his 2004 paper ‘much of conservation practice is based upon anecdote’.  However, every conservation intervention is carried out on a windy day.  Natural populations fluctuate according to the weather as well as other factors we are unable to predict.  A system we cannot predict is random.  For instance, I recently read that butterflies have benefited from the weather over the past few months, an effect which will continue to be important over the next year (link here).  The effect of the weather is so important that human actions could have degraded butterfly habitats over the past 5 year with butterfly numbers peaking this year, the year when habitat is at its lowest quality.  Any individual landowner who followed this trend (with their land lowering in value to butterflies over the past 5 years), who was also unaware of the effect of the good weather and who measured butterfly population sizes would conclude that the habitat was of the highest quality in the final year.  This is analogous to playing using a bad technique but obtaining a good outcome due to the effect of the wind when playing table tennis outdoors.

One partial solution is to simply play more shots and gain feel more slowly.  The effect of the wind will even out over many shots and good technique will result in a good outcome more often than bad technique.  Similarly, one cannot learn from one conservation intervention not carried out under controlled conditions.  To identify good and bad conservation interventions the outcome of many interventions must be compared.  It is here that conservation scientists have a role to play by comparing the results of many trials of a similar conservation technique across many different conservation strategies.  Every conservation practitioner, that is every individual who purposefully manipulates the land for a conservation charity, business, government or personal enjoyment has the potential to turn their ‘shot’ into a small piece of feedback by measuring the effect of this shot, by checking if this shot lands on the table.  Currently conservation practitioners are playing shots, carrying out interventions they hope will benefit nature, and turning around not bothering to check if the ball lands on the table, if the intervention is a good one.  This is the strategy adopted by Sutherland, founder of www.conservationevidence.com

Alternatively, or alongside this approach, scientists can exclude the random by conducting well designed experiments.  The feedback gained will only apply to a very precise set of conditions which may not be relevant in reality.  However, general principles can be tested and lessons learnt.  A framework of understanding can be built.  For instance, if you asked a coach why an overarm tennis serve is better than an underarm one then they may point out to you that by striking the ball above the height of the net with a fully extended arm the player is able to generate more power whilst ensuring that ball lands in the correct section of the court.  The coach can work from theory (a grasp of mechanics) to technique.  Likewise, an understanding of the factors determining algal growth can be used to predict the effect of fertiliser running into a lake, for example.  Using this understanding an intervention can be recommended, the effectiveness of which can then be measured as outlined above.

Conservation science throws up counterintuitive findings, findings that feel wrong.  This is not surprising and we should be wary of our sense of feel where that sense is based upon personal experience in a random environment.  Even in a non random environment it is a misconception that the correct theory or technique should ‘feel’ right when described or tried for the first time.  It doesn’t always.  I remember someone who coached me telling me how to adjust the position of my feet when playing a forehand shot, ‘it will feel odd and alien, like wearing shorts on a windy day’ he said.  He was right.  At first the new technique just felt wrong.  Similarly, when I learnt to juggle I had to trust the guy who had taken the time to make an instructional youtube tutorial.  When I tried to replicate what he showed me it felt wrong.  Now I can juggle and it feels right but if I try to learn a new trick, a new pattern of movement it feels wrong again

Conservation is practiced in a random and complex environment.  In an environment which does not lend itself to the acquisition of expert intuition.  We should be wary of trusting our own experiences too much.  As a profession conservation must realise the value of feedback and the limits of our understanding.  At best, by pooling the experiences of many conservation practitioners, conservation might succeed in being broadly right.