Verified by 
(This is Part 2 of a two-part blog series.)
Take a look below at the famous Müller-Lyer illusion. Every student of psychology knows that the two horizontal lines are actually the same length, even though the lines certainly do not appear that way. Knowledge of the true length of the lines cannot turn off or change the nature of the illusion. Hence, the illusion is said to be “encapsulated,” insulated from the influence of higher-order knowledge (that the two lines are actually of the same length). In short, because of encapsulation, the Müller-Lyer illusion cannot be modified or turned off as a result of one’s beliefs, desires, or knowledge. Many perceptual processes are encapsulated in this way. Similarly, conscious states, such as pain, nausea, and guilt, too, can be encapsulated. When one receives an injection from the doctor, for example, the pain associated with the injection cannot be turned off by the knowledge that the injection will benefit one’s health. Similarly, while holding one’s breath while underwater, one cannot turn off the desire to inhale, nor the negative feelings associated holding one’s breath (which is the adaptive thing to do in that context).
In cases such as these, encapsulation seems disadvantageous. However, over the course of development, encapsulation is adaptive: It would not be adaptive to turn off states such as pain, guilt, or nausea at will, for the role that these states play in guiding behavior would be gone. Just imagine the negative consequences that would arise if a young child could turn off pain signals at will. It is likely that the behavior of such a child would become riskier. In summary, many percepts, urges, and bodily feelings are encapsulated, and they should be. Regarding urges and the behavioral inclinations associated with them, one can suppress overt behavior (e.g., to grab a cupcake that belongs to someone else), but not the urges associated with those behaviors (e.g., the desire to eat the cupcake). One can suppress the action, but not the urge.
Because of encapsulation, each “conscious content” in the conscious field does not in a sense “know” of the nature of the other conscious contents composing the field. (A “conscious content” is anything that one is aware of; the conscious field contains everything [all the conscious contents] that one is aware of at one moment in time.) Because of encapsulation, each content in the conscious field also does not know whether it is relevant to current goals and actions.
This leads to a question: How come behavior does not always reflect encapsulation, and how can one, say, hold one’s breath while underwater, not grab someone else’s cupcake, or report that the two lines in the Müller-Lyer illusion are actually of the same length? This is because normal behavior is not dictated by any single conscious content, but rather by the whole conscious field. The conscious field thus permits the “collective influence” of all conscious contents activated at a given time. When consciousness fails, behavior is not “integrated” in this way. For example, in anarchic hand syndrome, the anarchic hand just might grab a cupcake that belongs to someone else or might out of the blue unbutton a button in the sleeve. It is important to note that these actions are not unsophisticated behaviors (neither a robot nor a 3-year-old can unbutton clothing). Rather, these actions are “unintegrated” actions that reflect encapsulation and not a fully operational conscious field.
Hence, from one theoretical standpoint, the conscious field could be construed as a “frame” that affords adaptive action selection, specifically for the the skeletal muscle system, which is the effector system for what in everyday life is called “voluntary behavior.” The conscious field itself is passive, like a car window, but essential for encapsulated conscious contents to influence action collectively.
In order for encapsulation to yield adaptive behavior, several conditions must be met. These conditions could be construed as the five burdens of encapsulation. First, because no content knows whether it is action-relevant or not, and also does not know of the nature of the other contents composing the conscious field, the conscious field must be very thorough and represent as many (potentially actionable) contents as possible, just in case. This explains why, even though the conscious field is for adaptive behavior, one is often aware of things to which one does not need to respond. Thus, encapsulation explains why the field is so capacious and inclusive.
Second, in order to benefit action selection, each content (e.g., the color blue versus the smell of lavender) must differentiate itself from all the other contents in the field, for a contrast not apparent in the field cannot be reflected in voluntary action. Each content must differentiate itself not only from contents within the same modality (vision), but also from contents from other modalities (smell). These contrasts must arise even though all content must somehow exist in the same decision space, and therefore share the same underlying format.
Third, because of encapsulation, the spatial layout of the stimulus scene must represent spatial coordinates as thoroughly as possible. This occurs for many sensory modalities (but not olfaction). This is because it is often the case that the “discriminative stimulus” that determines which action should be performed is not a single stimulus, but rather the spatial distance between two stimuli, as in the case of driving. Thus, our conscious field must have a rich and thorough representation of the spatial dimensions of the external world. The field does not know which such spatial relation might be essential for adaptive action selection.
Fourth, for action to be adaptive, such a spatial model of the world must include the emergence of first-person perspective, due to the demands of action selection, for example, when deciding between reaching for a large (but faraway) banana on one’s right or a smaller (but nearby) banana on one’s left. The first-person perspective is essential for this kind of action selection. (This first-person perspective also emerges in the dream world.) Fifth, because of encapsulation, and in order for action to be adaptive, the contents that compose the conscious field must all be comparable at some level, for they must exist as comparable tokens in a common decision space. These contents include information about the immediate environment, the representations of anticipated actions (e.g., mental imagery of to-be-produced actions), the effects of actual action (e.g., proprioceptive feedback), and even high-level cognitions. All of these contents, which tend to have a perceptual-like format, are sampled not by other conscious contents (which would violate encapsulation), but rather by the action systems in the Skeletal Muscle Output System. These systems are unconscious. In short, encapsulation explains why the conscious field, though in the service of adaptive action, contains contents that are not action-relevant, and why it has a first-person perspective and is so thorough (both in terms of its contents and the representation of spatial coordinates).
References
Morsella, E., Godwin, C. A., Jantz, T. K., Krieger, S. C., & Gazzaley, A. (2016). Homing in on consciousness in the nervous system: An action-based synthesis. Behavioral and Brain Sciences [Target Article], 39, 1-17.
