In his Principles of Psychology, William James devotes a chapter to "Attention." There, he describes attention as a "reactive spontaneity" that is both passively dependent on what is given and actively selective in terms of its focus.
Without this selective emphasis, the world would show up as a structureless chaos. Attention is a sorting machine that slots the oncoming line of objects into foreground or background, center or periphery, light or shadow. We can only foreground one thing at a time and the criterion used for sorting things is "interest." We foreground those objects that provoke our interest.
Here is James' own explicit definition of attention:
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 is worth noting that, on this definition, attention implies the necessity of its own withdrawal. We attend to something only by not attending something else. Or, to put it more bluntly: paying attention is the business of selectively ignoring.
It would seem that, of necessity, those who are good at paying attention to stuff will also be those who are good at ignoring at stuff. The cultivation of attention must split this difference: we must learn how to focus by ignoring without also succumbing ignorance.
As a general rule, we can only pay attention to one thing at a time. This "one" thing may well be an aggregate of multiple objects (what isn't an aggregate after all?), but to attend to them all simultaneously we must address them as a "single" thing.
This leads to a secondary definition of attention in terms of "gathering": attention is what gathers out and then gathers together an aggregate.
Or, more comprehensively: attention is what gathers objects, gathers me, and then gathers me in relation to the gathered object. Attention not only gathers objects in relation to me, it gathers me in relation to the anchor point of an object.
Here, attention depends on our ability to generate (unifying) networks of associations. We're able to pay attention to things only to the degree that we have some "place" to put them, some other stuff with which we can associate them, some compatriots with which we can number them.
The opposite of attention is distraction. If attention is a gathering, distraction is a scattering. Distraction is both a dispersion of objects and a dispersion of "me." Distraction is a focusless fog.
Distraction, though, is different from a lack of sustained attention because attention will and must move on. Attention is unavoidably serial in character. We must pay attention to this . . . and then this . . . and then this . . . and then this . . . and then this . . .
Attention is geared to flit. As soon as there is nothing new to notice, it will be on to the next thing. The conveyor belt of attention never stops moving. As James puts it:
No one can possibly attend continuously to an object that does not change.
This serial character may predispose me to hazy distraction, but it is not itself distraction. James claims that the only way to keep our attention from moving to a new object is to constantly work to keep seeing new things in the object we're already attending to. But only novelty can sustain attention: if I don't find novelty in the object at hand, my attention will move on.
In this sense, anicca is the key to samadhi:
If we wish to keep [attention] on the same object, we must seek continually to find out something new about the latter.
It follows, then, that my observation of the object, in order to be sustained, must be intensively interrogative. Every object needs to taste like a koan. Every object needs to be "rolled" over and over as we continually consider different aspects of it.
James also usefully distinguishes a number of different kinds of attention along multiple axises.
We might, for instance, distinguish kinds of attention in terms of (1) whether they are immediate or derived, or (2) in terms of whether the are passive or voluntary.
Objects of immediate attention are those that are interesting to us in their own right. Objects of derived attention are of interest only because we associate them with objects that are of immediate interest.
Objects of passive attention do not require active effort. Objects of voluntary attention do.
The point of connection between these opposed pairs is especially interesting. James claims that voluntary attention is always derived because attention only requires effort when an object is not of immediate interest.
The result, James says, is that when we talk about "sustained voluntary attention" we really mean either one of two things: (1) we're naming a repeated (and ongoing) series of successive (but failed) efforts to pay attention, or (2) we're naming a kind of attention that starts off as derivative and voluntary but then becomes passive and potentially immediate. (In either words, there is no such thing as sustained "voluntary" attention :)
James additionally points out that the better someone is at forming novel associations with an object, the better they will be at paying attention to that object. And, similarly, those whose minds are already equipped with rich sets of associative networks will have an easier time of gathering novel objects into those networks.
Two final notes.
First, there is a connection between attention and memory. Objects that are paid attention to will be remembered (cf. "sati").
And second, the nature of the attentive process (1) involves a physiological disposition of sensory/muscular readiness (one must not discount the role of the body, of posture, etc. in attention), and (2) involves a mental disposition of anticipation, preparation, excitation.
Because this dimension of anticipation is so important, James proposes that we describe all attention as a kind of pre-perception.
The only things that we commonly see are those that we pre-perceive or anticipate as being of interest to us. Attention itself amounts to a kind of "looking for" or proto-desire that screens and selectively emphasizes on the basis of interest.
Though we may initially notice many objects as being of derived importance because they function as "signs" indicating something about some other objects that are of immediate importance to us, the habitual pathways worn by these signs to their referents will soon lead us to skip over the sign altogether and pass directly to the object of primary interest. As a result, pre-perception makes attention possible because it primes us to ignore the habitual.
One thing can come to the fore only if many others recede into the background.
Pre-perception both enables focused attention AND impedes my ability to attend to what is habitual or unexpected.
Here, as elsewhere, the business of cultivating attention will have to grapple (and, almost certainly, continue grappling) with the inconvenient double-bind of attention itself.