Constructing reality: some thoughts on the nature of perception

Constructing reality
Exploring the nature of perception, and how this might impact on wine tasting

Cobb: 'They say we are only using a fraction of our brain's true potential. That's when we are awake. When we are asleep, our mind can do almost anything'

Ariadne: 'Such as?'

Cobb: 'Imagine you are designing a building. You consciously create each aspect. But sometimes it feels like it is creating itself.'

Ariadne: 'Like discovering...'

Cobb: '...Genuine inspiration, right. Now, In a dream, our mind continuously does this: we create and perceive our world simultaneously. Our mind does this so well we don't even know it is happening. This allows us to get right in the middle of that process by taking over the creative part.'

Lines from a conversation between the characters Cobb (played by Leonardo di Caprio) and Ariadne (played by Ellen Page) in the film Inception (2010).

Consider the following common expressions:

'We're living in different worlds.'
'What planet are you on?'
'He's in his own world.'

These all reflect the fact that it's not uncommon for two people people to share their version of an experience in very different ways. Is this because of their poor recall of events? Or is it that they actually experienced something very different?

Is seeing believing?
Is what we see what is really there? Skilled illusionists are able to trick us remarkably consistently – they have an innate understanding of how the human visual system works, and take advantage of this. We're amazed by what they do, because we trust our eyes.

Visual illusions – and there are many of them – show us clearly that we don't always see what is actually there. If our visual systems are constructing a view of reality complete with flaws, then this poses a crucial question: how much of our experience of reality is actually real?

I suspect that my experience of the world is very different from the way my dog experiences the world. When we take a walk together, if she were able to speak, her account of the reality that we shared on the walk would be very different from mine. My experience is very visually dominated. Hers seems to be dominated by smell. As she stops and sniffs, she detects signals that are hidden from me. She leaves her own signals. When she meets another dog, a large part of the interaction takes place through mutual smelling. It's a world largely closed to humans.

The assumption here, of course, is there is a reality out there, that we have in common: an external environment in which we live and move and breath. How we experience this reality depends on the particular sensory apparatus we have and – crucially – how the we use our brain processing to interpret this sensory input and assemble it into a conscious experience.

Rodents have an important sensory system involving a specialized form of touch: their whiskers are part of a vibrissal system that allows them to map the world spatially even in the dark. Bats use ultrasound to navigate. All these sensory systems are creating a slightly different account of the same reality - the external world. Yet we are so used to our own particular account of the world that our unspoken assumption is that how we perceive it is the way that it is.

Of course, it makes sense in practical terms to make this assumption, because our sensory systems and the processing of these signals allows us to model reality in a way that extracts out the information useful to us for our decision making. But our perception is, by necessity, partial. More importantly, I will argue that what we perceive is not necessarily what is there. 'Reality', as we perceive it, is something unique to each of us. To a degree.

Sensation is a unity
Sensation is a unity. We are used to thinking of senses as discrete entities: sight, smell, taste, touch, sound. But our experience of the world is a single, unified perception. At the moment I'm receiving lots of information from the outside world, and also from my own internal state. But unless I specifically attend to various elements of this information, I'm not really aware of most of it. As long as I don't feel hot or cold, I'm not thinking about the air temperature. As long as I don't have a sore back, or have been sitting in the same position for too long, for example, I'm not aware of touch.

I concentrate on my laptop, unaware of all the information in my peripheral vision, save for motion, which seems to pop out. If someone moves particularly fast, or comes very close to me, I can easily ignore it. There's a hum of background noise: I will only be aware of this when something alerts me, and then my stereo vision will guide my eyes to look in the direction of where the loud or unusual sound has come from.

By the time we are aware of the environment around us, a lot of processing has already taken place to the sensory information we have picked up. It is as if our perception is akin to the chief executive of a large company who doesn't want to know all the little details, but just what is important in helping her make critical decisions. She receives an edited version of what takes place in the company with the uneccessary material taken out. This is the only way she can work effectively.

This higher-order processing by the brain fuses together sensory input from a variety of sources, adds into the mix our own internal state, and then delivers to our consciousness a model of the world out there that is relevant to us, with the largely irrelevant details relegated to the background. With vision, this processing is clear. We have specialized modules for face processing, for example. In our field of view, faces are very important. So part of our visual processing is involved in recognizing faces from a scene and affording them special significance. Likewise, our peripheral vision is very sensitive to motion, because moving objects just out of our main field of view are likely to be important.

What is ‘red’?
Colour is interesting. Let's take just one colour: red. The colour 'red' does not exist in nature. It is entirely subjective. This can be shown by context effects, where a 'red' shape can be placed in the context of surrounding colours that cause it to be perceived differently. Looking at it the other way round, 'red' is normally experienced when our photoreceptors are hit by light of a certain wavelength. But context experiments show that the brain then processes this information such that in some instances red isn't red at all, demonstrating that redness is something that we construct.

Another interesting observation is the response of captive-bred infant monkeys to a plastic snake. As you'd expect, they have no fear of snakes. But they develop fear of snakes if they are shown a film of a monkey showing a fear response to a toy snake. If the film is changed so that the fear response is to another object, say a flower, then the captive-reared monkeys don't develop a fear of flowers. So it is as if there is something innate in monkeys that prepares them to learn to fear snakes. Even though these infant monkeys have never seen a snake, they have an innate emotional response to the visual stimulus of seeing a lifelike model snake once they have seen another monkey react fearfully to that snake. This means that hard-wired in their brains is some processing that attributes significance to a snake-like object. Pigeons have a similar response to the shadow of a hawk – one that seems hard wired.

Constructing reality
Where am I going with all this? I am trying to make a case for the subjective nature of our perception of the world around us. Our perception, which is a unity, is based on a common shared reality of what is out there. But it also includes important components that come from us. We model reality. By the time we are aware of what is out there, we are experiencing something that is, to a degree, unique to us. We are, in effect, constructing reality.

This line of argument seems a little dangerous. It sounds strongly relativist. But it is not to dismiss all objectivity, and it does not mean that anything goes. For all intents and purposes, our notion of the world around us is largely an accurate one, but it is not as accurate as we would like to believe.

In order for us to respond rapidly to a changing and sometimes surprising environment, it seems that we have had to do a fair amount of interpreting and modelling of what is out there before we are consciously aware of it. We take sensory information about the environment around us, but we don't have time or processing power to take all of it and process it accurately, so instead we focus on those elements that are most likely to be of use to us, and use these to model reality.

The fact that this extracted information is then subjected to interpretation is of great interest to me. This could be the reason that different people have such different experiences of the same events. Their experience of reality is subtly but significantly different, because of the way they interpret it even before they are aware of it.

This all suggests that the brain has some way of generating 'reality' from relatively limited external input. And could it be that the mechanism for generating 'reality' could even function in the absence of external input?

Dreams and hallucinations
This is where dreams come in. I've always been intrigued by my dreams. The experiences I have as I dreams sometimes seem incredibly real. Yet they are not simply based on experiences I have had. They involve situations that are entirely novel. Yes, sometimes I can see a link between recent experience and the content of my dreams. But often there is no such connection, and the ideas and experiences I have in my dreaming state are often quite remarkable.

Dreams could be an artefact of the brain's reality-generating module. I would go further, and say that they represent evidence of what the brain is doing in our perception of reality. The cues received from our senses kick into action this reality-generating module which then creates our perception. Thus perception itself could be internally generated, albeit based on an external reality. This seems quite a scary thought, but it fits well with the reality of perception.

Further evidence for this reality generation come from the effects of psychotropic substances. Hallucinating hippies, high on drugs, experience a starkly different 'reality' to the rest of us. Could it be that the ability of some chemicals to induce hallucinations is because they tap into the brain's reality-generating module, with surprising results?

Why would the brain have the ability to create reality if this were not a useful or important feature? It is unlikely to be a side-effect of some other functional process. And it is hard to imagine how a psychotropic drug might create reality unless it is tapping into some pre-existing module or mechanism designed for this purpose.

Implications for wine?
So what are the implications of this for wine? There is no reason not to think that our perception of flavour is something we generate, based on the physical properties of the food or drink we are consuming. These physical properties include chemistry, but also – importantly – the appearance of the food/drink. We bring a good deal to the encounter, too. Our experience, knowledge and expectations, as well as our internal state, will shape our perception.

Perception of flavour isn’t just about taste and smell. Our perception is a unity, and flavour is a part of that unity. It’s more than just multimodal (that is, it involves several senses) – part of flavour perception is internally generated. In this sense, flavour is not a property of the wine we taste.

This is something we need to bear in mind in the way that we discuss wine, the way we carry out tastings, and also the way that we communicate about wine. It doesn't mean that wine is totally subjective – after all, there is a surprising amount of agreement when we taste wine together. But it calls for an enriched, more grown-up understanding of the nature of wine tasting than the current model that is taught to students of wine.