Verified by 
Trying to characterize the “architecture of the mind” is one of the central projects in cognitive science. From understanding what neurons actually do, to designating certain brain regions as responsible for specific activities, this interdisciplinary endeavor has defined the work of many neuroscientists, experimental psychologists, and philosophers.
One characterization of the mind, which generates plenty of debate, is that it is organized in specific “modules” responsible for specific tasks. Theoretically, these modules operate independently of each other in order to perform their designated role in cognition, and many cannot be influenced by other modules or be processes of which we are consciously aware. For example, there are claims that language acquisition is reliant on an innate module that specifically evolved for linguistic functions (see Noam Chomsky’s proposal on “universal grammar”). Without this language-specific module, our ability to communicate in the sophisticated way that we do would be impossible. It seems there are several areas that support language (e.g., Broca’s area, Wernicke’s area), and damage to one of these particular areas makes the individual lose an aspect of linguistic processing, such as speech production or comprehension.
The modularity of mind hypothesis, as defended by the philosopher Jerry Fodor (1983), has many implications for understanding how the mind works and how neuroscientists might approach the brain in clinical settings. For example, if one region suffers damage, the potential for successful rehabilitation depends on that region’s plasticity or else the individual might be unable to recover lost abilities. This modularity proposal also can help us understand how cognition evolved in organisms—the more ancient the “module”, the earlier the abilities associated with it.
If the architecture of the mind and the anatomy of the brain are such that a specific ability is completely encapsulated within a module, then it seems that the brain has a very rigid architecture that cannot recover from damage or adapt to the environment. We know from clinical examples that this is not the case. Another problem with a too rigidly modular brain is that there would be less crosstalk between regions and would challenge the integration of information from different processing areas in the brain (e.g., those responsible for processing sound and visual information)—contrary to the conscious experience that we have. But then if the brain is too disorganized, it would be difficult to maintain systematic cognitive operations (e.g., think about how our internal organs are organized). There should be some “belief-independent” systems, for example, responsible for processing sound and some for processing vision, since these inputs have different physical properties. Clearly the brain cannot be completely modular or completely lacking organization. How can we identify the balance?
One idea to consider that can help us understand the architecture of the brain is that of cognitive penetrability (see Pylyshyn, 1999; Raftopoulos, 2001). What this idea addresses is just how much one module (or function) of the brain can influence another. Sources of influence include things like concepts, desires, and beliefs. If there is pervasive cognitive penetrability, then beliefs can affect perception at a very low level. For example, your belief that the sun will be purple today would change the way visual information is processed so that it is processed as something which is purple, even though it has the same physical properties it had yesterday when it appeared to be yellowish. On the other hand, if all modules are cognitively impenetrable, then it would not be possible for the different modules to communicate or interact with each other to achieve the overarching goals of an organism, particularly a complex one like a human being. There must be some modules of the mind that are more hard-wired and independent (and impenetrable), while others can receive inputs that modulate the mechanics of the module (but these are uninteresting cases of penetration, as they simply change the input; see Firestone & Scholl, 2016). So at what level of processing can belief penetrate and alter perception?
Keeping in theme with our posts on the dissociation between consciousness and attention, or CAD (Montemayor & Haladjian, 2015), it is possible to understand how impenetrable a module may be based on understanding the relationship between attention and consciousness. Since consciousness may be generally dissociated from attention, we can argue that some fundamental aspects of brain processing (e.g., attention to features) that happen in the background outside of our awareness are cognitively impenetrable. Indeed, such older neural systems related to basic attentional processes are less likely to be penetrable and only operate outside of awareness, like the ability to bind different features together into an object-based representation. Some other aspects of the mind, such as reasoning and conceptual attention, are more fluid and influenced by belief as well as perceptual input.
In the end, we can take a middle ground and claim that some modules cannot be penetrated by beliefs, such as those that fundamentally process perceptual information (although this is a hotly debated issue). This modularity is necessary in order for us to interact with the environment in a more consistent manner (think of how dangerous it would be if our beliefs and desires about size influenced how we perceive the size of road obstacles while riding a bicycle). This sort of natural standardization of perceptual inputs also helps with producing shared experiences, which supports social interactions. Beliefs can play a role later on in the processing of information, but not at low-level processes. The information that enters our subjective conscious experience is a combination of outputs from fully encapsulated modules and influences from beliefs that shape these outputs.
- Harry Haladjian & Carlos Montemayor
Note: Since this continues to be a debated topic in various disciplines, there is an upcoming special issue in Frontiers dedicated to it: http://journal.frontiersin.org/researchtopic/4600/pre-cueing-effects-on-perception-and-cognitive-penetrability
References
Firestone, C., & Scholl, B. J. (2016). Cognition does not affect perception: Evaluating the evidence for ‘top-down’ effects. Behavioral and Brain Sciences, FirstView, 1-72.
Fodor, J. A. (1983). The Modularity of Mind: An Essay on Faculty Psychology. Cambridge, MA: MIT Press.
Haladjian, H. H., & Montemayor, C. (2015). On the evolution of conscious attention. Psychonomic Bulletin & Review, 22(3), 595-613.
Montemayor, C., & Haladjian, H. H. (2015). Consciousness, Attention, and Conscious Attention. Cambridge, MA: MIT Press.
Pylyshyn, Z. W. (1999). Is vision continuous with cognition? The case for cognitive impenetrability of visual perception. Behavioral and Brain Sciences, 22(3), 341-365; discussion 366-423.
Raftopoulos, A. (2001). Is perception informationally encapsulated?: The issue of the theory-ladenness of perception. Cognitive Science, 25(3), 423-451.
