“Everyone knows what attention is. It is the taking possession by the mind in clear and vivid form, of one out of what seem several simultaneously possible objects or trains of thought…It implies withdrawl from some things in order to deal effectively with others, and is a condition which has a real opposite in the confused, dazed, scatterbrained state.”

says American psychologist William James.
The fact that most psychology textbook introductions to the concept of “Attention” start with this sentence, “Everyone knows what attention is” strikes me as unsatisfactory. Do we really know what Attention is? And if so, meaning that we all have an clear idea of how it feels to pay attention to something or someone, would that suffice as an account to the phenomena of attention?
William James uses a common trick of defining something by describing what it is not: He contrasts “Attention” with what he thinks is its opposite, namely the “confused, dazed, scatterbrained state [of mind].” But an answer like this misses the point.

Consider this definition as yielded by a Google “define:” search:

A psychological construct describing detection, selection, discrimination of stimuli, as well as allocating of limited processing resources to competing attentional demands. Functional attentional processing is essential for higher order cognitive processes such as learning and memory.

Attention, a psychological construct, that just serves higher order processes?
Does this mean that there was no such thing like “attention” before some psychologist decided to construe the concept?
Definitions like this one clearly cause more confusion than they shed light on what the original question was asking for.

Seth Goldstein, CEO of Root Markets, came up with a very unorthodox way of describing what Attention is:

Attention is the substance of focus.

This one abandons any kind of psychological vocabularly and focuses on the phenomena of Attention itself, which is described as being a substance standing in a certain relationship to focus. (where focus IMO is nothing but intentionality).
Firstly, why call it a substance? Consider the ways the word “attention” is used:
We pay attention to something. In psychology there is talk about the delegation of attentional resources. We have limited attention. Someone receives attention.
This is all substance talk.
In the German language the Attention substance is given away for free: “Man schenkt jemandem oder etwas Aufmerksamkeit”, just like a gift - but that’s not true (given that people never expect any returned value for their gifts). Giving away Attention always involves the implicit wish for a reception of value.
The exchange of Attention substance for a desired value is fundamental to life: The food hunting animal focusses on its prey - everything else is rendered background to the hunt and all attentional resources are being delegated to the target. Attention is the hunt’s fuel. The aim is to catch the prey before you run out of it, tired, exhausted and still hungry. Satisfying your hunger is the promised value that is being proposed and eventually exchanged for this purest form of Attention.
Viewing Attention as a substance comes with considerable metaphorical advantages when talking about which role Attention plays in our world. At its heart, attention is about our interaction with the world - it’s the currency of intentionality.
What makes it unique among any kind of currency is that it even can be exchanged for “itself”: I pay attention to you, so please give me some of yours back for it. This indicates how my attention is different from yours. It’s not like money where an exchange of my 10 bucks for your 10 bucks would be an odd thing to do. There are as many instances of Attention as there are conscious beings.

Social Saliency Maps

A saliency map is a neuroscientific method that is used to investigate brain processes that govern bottom-up overt attention. The paradigm of “overt” attention means that it is apparent where or to what a subject is paying attention to by following its eye-movements (preferably with an eye-tracker) - naturally assuming that the places we look at equal the places we are paying attention to. The idea of “bottom-up” processes (as contrasted with “top-down”) is that we do not consciously decide where to look first when viewing images, scenes, stimuli, but that this is rather predetermined by the properties of our nervous system and the properties of the stimululs. Very simply put, if I’d show you an image which is completly black, except for a red dot in its periphery, I can predict that this is where you are going to move your eyeballs at, and hence, draw your attention to. Now it is possible to take an image/stimulus that you are going to display to a subject and compute a more sophisticated saliency map by extracting several feature maps(including maps for contrast, color, luminance etc.), from the original:

The method of saliency maps, invented by Laurent Itti & Christof Koch does very well in predicting the first few locations of an image where subjects are going to pay attention to. In such experiments subjects are usually merely told to “investigate the images carefully”, trying to avoid any task-related biases of attention.
These bottom-up, unbiased attention processes have been investigated very carefully in the last decade, though there are still many labs working on it trying to come up with more precise accounts of this form of determined attention, and hence more sophisticated saliency maps. There are also efforts to incorporate top-down effects into neuroscientific models and theories of attention - i.e. attention processes that are governed by concious decisions.
Yet it is rarely the case in real life that you are sitting in front of a screen, a weird infrared camera-armed helmet mounted on your head, ordered to view a bunch of boring images taken from more or less natural scenes.
In real life we are constantly exposed to other people, engaging in communication and interaction. Consider the case when you are with another person and some more or less salient objects in a room. Suddenly the other shifts her gaze towards a spot - you can’t help but to look there too to see what caught her attention. This would surely override the possibly predetermined hierarchy of object saliency in the room - and it is still “bottom-up” - since you can’t help it. However, this behaviour is not merely due to the structure of our nervous system but needs to be explained by the presence of the other person and her influence on your behaviour (even though one could argue with the yet relatively shaky idea of mirror neurons).
What we need here is a model I’d like to call the Social Saliency Map - a saliency map that incorporates the mechanics of social interaction and - if you want so - the functionality of mirror neurons.
Such a model would also be able to account for several attention-related phenomena that only become apparent when expanding one’s view from the individual to the social domain:
As just described in the example, the mere shift of attentional resources of one person can be enough to elicit a similar shift for everyone who was paying attention to this person before, whereas the “amplitude” of this shift possibly correlates with the amount of Attention that was invested in the person in the first place - think of great leaders, think of stars.
Another related remarkable thing about attention in societies is that people tend to invest their attention into what seems to attract the attention of many others - probably the strongest mechanism of societies in general?

If I find some time in the next days, I’ll write about what this all could possibly mean

PS: Yip, it’s been quite a while since I posted last here, and yip, knowlogo.org still’s not up. I guess I’ll have to work on my personal attention spending policy and finally get to read GTD.

technorati tags: attention

Thanks for visiting my blog! If you like it, subscribe to my RSS feed or share this post via one of the social media services below.