Attention and consciousness are often thought to be inseparable or even two words for the same thing, but under unusual circumstances they can be separated. It is not easy to separate them and so what consciousness does to assist attention would be an important function of consciousness to the extent that attention was important to survival. Well that’s a no brainer – all you have to do is not attend to what is happening when cross a busy road to understand the risk to survival of not attending to the important things. Surely this function alone would pay the biological cost of consciousness.
Consciousness, in effect, is a canvas on which the spotlight of attention can fall so that all areas of the brain can know what is currently important and so that part of the canvas can be rendered in more detail. What steers the spotlight? We have mechanisms to keep the brain focused on a task or goal, thought of as top-down control, and mechanisms to shift focus to unusual or alarming sensory input, thought of as bottom-up control. This keeps what is significant in awareness and available to the whole brain. All the resources of the brain can be brought to the important problem of the moment. Because what is currently significant is in consciousness, it is also in memory. We start memories with the most important aspect of each consecutive moment.
Victor Lamme and his group have studied attention and consciousness with, to my mind, a very reasonable attitude – forget introspection etc. and let the neurological evidence rule. Looking at the neural correlates of consciousness – the fast forward wave and the feedback wave with its synchrony across large areas of the cortex – and controlling the visual stimulus, they find the following.
Because depth of processing (attention) and the fast forward sweep (FFS)/ recurrent processing (RP) distinction are orthogonal, a visual stimulus can reach any of four stages of processing:
• Stage 1: Superficial processing during the FFS; This would happen when a stimulus is not attended and is masked. Unattended and masked words, for example, do not activate word-form selective areas, only visual areas, so do not even penetrate deeply into the ventral stream hierarchy.
• Stage 2: Deep processing during the FFS; for example, a stimulus that is attended, yet masked (and hence invisible). This stimulus does travel through the whole hierarchy of sensory to motor and prefrontal areas, and may influence behavior, as in unconscious priming.
• Stage 3: Superficial processing of a recurrent/ re-entrant nature (RP); for example, a visual stimulus that is given sufficient time to evoke RP (i.e., is not masked within ∼50 ms) yet is not attended or is neglected, as in neglect, inattentional blindness , change blindness, or the attentional blink.
• Stage 4: Deep (or a better word may be “wide-spread”) RP. This is the case when RP spans the whole hierarchy from low level sensory to high level executive areas. This occurs when a stimulus is given sufficient time to engage in RP and is attended. Others have equated this to the situation that a stimulus has entered global workspace .
Or this description:
Initially, all objects are processed by low level areas in a feedforward fashion, so that basic features are extracted at about 100ms (Stage 1). Some objects are processed more deeply at about 200ms (Stage 2), depending on top down and bottom up attentional selection. Meanwhile, recurrent processing in early visual areas emerges also at about 200ms (Stage 3) for all or most of the objects. Later still, at about 300 msec, recurrent processing grows more widespread (Stage 4) for those objects that are selected by attention (potentially slightly different ones than those that were favored initially, as attentional selection is influenced by previous processing). After stimulus removal, Stage 3 processing turns into iconic memory, while Stage 4 processing turns into working memory.
It is difficult to envisage how what is significant could be decided and broadcast without the use of some structure like consciousness awareness in which it can be tagged. And it is hard to see how the brain could work without global priorities.
Besides the significance, there is an aspect of attention that is connected with language. E.B. Bolles (here) has been writing about this for a few years – and in his theory, the words in language steer attention so that the speaker and listener can mutually focus on the same topic. So if I start by saying ‘the car’ then my listener will focus attention on some object that fits that description and/or the concept of ‘car’ in the listener’s mind. This steering of attention may be a further reason (beside the need to use working memory) why language is almost always conscious. I will be returning to language again later.
There is more to come. Previous posts in this series: