The neuroscience of conditioning
Sometimes the best confirmations of the dhamma come from sources that have nothing to do with Buddhism. On Intelligence by Jeff Hawkins is just such a source. Hawkins is an electrical engineer and entrepreneur whose interest in Artificial Intelligence has convinced him that the key to developing AI lies in understanding the brain. If that sounds a little obvious, it’s necessary to say that much of AI research – even on neural networks – has ignored the biology of the brain. As the name of the book suggests, this is not about consciousness or experience at an abstract level. It’s about human intelligence and how that distinctly human (well, mammalian) part of the brain – the neocortex – makes us as smart as we are. What I propose to do here is pick out some of the main points that Hawkins makes and show how they relate to the kind of things that we might notice in ourselves as we peer through the microscope of meditation. In particular, this book offers biological explanations for our ingrained habits – the conditioned responses that arise in us despite our best intellectual intentions and endeavours to behave otherwise. The closer we understand the nature of the mind, the better we can work with it.
The neocortex is structured in a uniform way in its entirety, regardless of function or location. That structure consists of 6 hierarchical layers, each as thick as a business card. Those 6 layers are further interconnected across sections of the neocortex to form hierarchies of hierarchies. Signals come in from the sensory neurons – like the nerves coming from your eyes or ears – in a rapidly changing fashion (both spatially and temporally) but as these chaotic signals filter up through the hierarchies, they stabilize and solidify. For example, input from the optic nerve (one million sensory neurons) is a firehose of light, colour and line shape changes due to the ever changing nature of the photons hitting the retina, and also the constant eye movements (saccades) we involuntarily make to scan our field of vision. By the time it filters through several layers of neocortex, this cacophony of electrical impulse has become something stable like, for example, a face.
It might even be your Aunt Susan’s face. If it is Aunt Susan, then effectively the memory of Aunt Susan’s face is encoded high up in the hierarchies of the cortex, and can be triggered by Aunt Susan no matter what the lighting conditions or angle of view are. The important thing to notice here is that the brain has an invariant representation for a vastly changeable (to all practical purposes infinitely changeable) set of input signals. But here’s where it starts to get really interesting. The flow of signal is not only upstream from the optic nerves to the memory of your aunt’s face. It’s also (perhaps even prevalently) back downstream. If higher layers are starting to see things that correspond to Aunt Susan, they feed this back down the line, and hone the incoming signal to check for Aunt Susan-ness. This is very efficient, as you can imagine, as it involves a narrowing of the search as early as possible. It’s a little like what happens when you type in a search term on Google and you are offered increasingly specific choice to select from. (For more details see the book’s wikipedia entry, or read the book!) This way of processing signals is elegant and much faster than a computational approach, but it comes at a price: At a very biological level, we decide what we perceive based on what we have already experienced. If that’s not an recipe for habitual reactive behaviour, I don’t know what is.
So if this is the way our brain works, how does it effect our everyday life, and how can we use this understanding to work better with our minds? On Intelligence has nothing to say about this, and what follows are my own – hopefully rational – extrapolations from Hawkins’ conclusions. His model of perception explains why changing habits through a conscious, intellectual application of will can be so frustratingly difficult. The processing described above takes place long before the filtered sensory signals reach our consciousness. The key interpretations of what we are experiencing are made long before they reach the ‘selfing’ processes of the brain, and so the the ego can really only dress things up as best it can – to claim ownership of that interpretation. But by that time, those interpretations have already begun to send signals to other centres of the cortex, including our motor neurons. In this way, habitually-wired reactive thoughts and actions are triggered. Our conscious ego, always behind the curve, tends to either justify the resulting behaviour in some way, or in general to tell some story about it. Brute force application of will is our favourite way to try to change those habitual thoughts and actions, but it can never reach into the depth of where those habits come from.
If this description invokes a certain hopelessness, and calls into question the notion of free will, then I think there’s no harm in that. I think it is probably hopeless and pointless to believe that we can impose our conscious will on activities that are upstream of the consciousness process. We can certainly modulate and moderate some of the grosser behaviours that we perceive to be in need of change. But we cannot by sheer intellectual will simply decide to be, say, more compassionate individuals from one day to the next. So what can be done?
Surrender. First and foremost to the nature of your own mind. You can’t work well with a system if you don’t have some understanding of how it operates, and the science is telling us in an ever-clearer way: we are not who we think we are. Our minds are not a centralized command-and-control system. Control is distributed across thousands of drivers (to borrow an image from Bodhipaksa), each struggling to grab the steering wheel. We are bags of competing habits, so let’s give up all hope and pretense of being in charge and let’s look instead to work with what we really have. Instead of trying to pull imagined levers and throw non-existent switches, we can plant seeds, in the form of new habits.
Incidentally, Buddhism has been saying the same thing for a very long time. The cognitive function of recognizing Aunt Susan’s face is called sañña in Pali, sometimes translated as perception. It is one of the Five Aggregates (khandhas) and is described by Bhikku Bodhi as follows:
The characteristic of perception is the perceiving of the qualities of the object. Its function is to make a sign as a condition for perceiving again that “this is the same,” or its function is recognizing what has been previously perceived. It becomes manifest as the interpreting of the object…by way of the features that had been apprehended.
The sign referred to by Bhikku Bodhi, corresponds to Hawkins’ invariant representation.
Once we surrender to this unintuitive and perhaps emotionally difficult way of understanding the brain, we can work with it instead of against it by gently initiating new regular habits – without any grand expectations – and see where it leads. This is where the blunt instrument of conscious intellectual will comes into play. The will is just another process struggling for control. It arises and falls away like everything else and we cannot expect it to change our habitual thinking and behaviour. But our will can help us to initiate new habits and support them in their infancy. In those moments when it is available, we can use its direction and energy to take the micro-actions that, in the long run, can rewire habits. We can decide to ‘feed the wolf of love‘ day in, day out as Rick Hanson suggests, and then let that work its own way through deeper and mostly unseen habits. And we can introduce meditation as a life habit. One of the most important habits that I’ve built up recently is the habit of daily meditation. Thanks to the 100 day challenge in the Wildmind Google+ community, I can say that the habit of sitting has become strong and has very little trouble grabbing the steering wheel at least once a day. I have found over the years that meditation itself has the effect of rooting out old useless or harmful habits. A few of them have just fallen away. Others put up a fight and fade in and out over time.
So what might be learned here? I have come for now to this conclusion: A better understanding of how our minds work tells us that there is no self there in control, but there are features to human intelligence that can be harnessed in order to favour certain drivers over others. The way to bring about positive change is to plant seeds, water them regularly and be patient. If that sounds a lot like love, I don’t think it’s a coincidence.
On Intelligence can be bought on Amazon.com
(I’m very grateful to Bodhipaksa for reviewing this article and bringing my attention to Bhikku Bodhi’s definition of sañña.)