Archive for December, 2012

h1

First-Mate Z

December 21, 2012
Shaun: "If this thing doesn't have intentionality, how does it have a brain eating goal?"Ed: "Who cares! I just wish he'd jog on."

Shaun: “If this thing doesn’t have intentionality, how does it have a brain-eating goal?”
Ed: “Who cares! Let’s just tell him to jog on.”

First-Mate Z

or

How Your Inner Zombie Rules Your World

by Mike Stayton

In this essay I argue that mental modularity can explain the existence and function of our inner zombie mind. I will first explicate a modular hypothesis that attempts to account for complex unconscious actions.This will be followed by experimental and clinical research providing support for the hypothesis. To close I provide further explanation why modularity of mind can make sense of the zombie and conscious minds.

Most have seen a zombie on film. But what characteristics do these living dead creatures share and on what basis can I claim that we are all part zombie? Other than an insatiable taste for brains, classic zombies uniformly tend to exhibit slow clumsy movement, blank stares through glazed eyes, and general absence of sensations, feelings and a general lack of mental awareness in general.

Few realize that within each of us lies an active zombie. Sufferers of somnambulism, commonly known as sleepwalkers, regularly exhibit these same zombie-like characteristics. Sleepwalkers rise from bed and engage in unconscious activities ranging from preparing a meal to driving a car. The expression of these and similar unconscious actions are the manifestations I speak of as the actions of the zombie within. Somnambulism illustrates the existence of low-level mechanisms capable of controlling actions behind the back of conscious awareness.

Christof Koch claims that “zombie agents” are the mental mechanisms responsible for the unconscious processing of background information and activation of automatic motor operations. These zombie agents control the sleepwalker’s automatic “zombie behaviors.”[1] Surprisingly, the same zombie agents run quietly in the background of our own waking minds. To provide evidence for this, we must observe subjects whose conscious sensations have been restricted, allowing the zombie behaviors to shine through. Before we do this we will examine Goodale and Milner’s (G&D) modular hypothesis.

Goodale and Milner’s Modular Hypothesis

Perceptual research is providing a bulk of the current information regarding unconscious mechanisms. In Sight Unseen, G&D argue that there are two distinct visual systems – the ventral and dorsal streams. The ventral stream specializes in gathering information about the world around us and is responsible for providing conscious perception. The function of the dorsal stream is vision for action. These streams corroborate to give us the dual abilities of gathering information and interacting with our environment.

The ventral stream is scene-based, providing us with a cognitive photograph of the world. Its function is to construct a useful knowledge-based representation of our environment. The ventral stream forwards information to higher-level cognitive areas for identification and categorization. This information is then stored in longer-term memory. The snap-shot-like ability allows us to remember and think of the past, present and future, in turn giving us the ability to plan. Ventral perception provides relative, object-based, metrics comparing objects in terms of each other, e.g. bigger/smaller than, closer/further than. Its higher-level connections provide us with information necessary for correctly identifying people and objects. Accurate identification requires a high level of processing causing this stream to run relatively slowly. The ventral stream provides intentionality – the content for our thoughts.

The dorsal stream relates to what I am calling zombie agents. It provides the information necessary for action. To do this, it processes the immediate present, allowing us to interact with our environment in real-time. In addition, it relies on more precise egocentric metrics based on the relation between agent and object. While the ventral stream informs us that our spoon is further than our cereal bowl, this pictorial knowledge is not sufficient to accurately guide our hand to our spoon. We need to know the exact location of objects to act upon them. Such precise metrics are accomplished by a derivation comparing an object in relation to the known location of the hand and assisted by visual stereopsis. Because the relation between an agent and his environment is in constant flux, especially during agent action, the dorsal stream is required to process information at near-instantaneous speeds. This immediacy negates any need for long-term memory. It instead utilizes only enough unconscious episodic memory for accomplishing the task at hand. Object-specific details are also unnecessary.

Evidence for the Everyday Zombie

It is these zombie agents that enable the unconscious sleepwalker to avoid stubbing his toe. Night walking experiments[2] show it is possible to navigate your environment without conscious attention to the surroundings. The subject attempts to walk in the dark, outside, while focusing his attention on a phosphorescent sphere suspended from the bill of his baseball cap. This purposeful fixation directs the subject’s attention away from his environment. Eventually, the subject is able to move confidently through the dark while successfully avoiding ground-level obstacles. Not only can your inner zombie process visual feedback without conscious vision, but it can also signal the appropriate feed-forward responses to coordinate the plethora of complex mechanisms required for navigating your environment while simultaneously preventing you from tripping over the ottoman.

Koch explains nightwalking[3] by positing that the dorsal stream is the mechanism monitoring the extreme lower visual peripheries and controlling the coordination and placement of the feet. Zombie agents are all that is needed for successful navigation. Koch states that higher-level categorization, controlled by scene-based consciousness, is unnecessary at the periphery of vision.

Visual agnosia patients can see but have a strange inability to visually recognize certain aspects of objects. One patient, called DF, can see colors and textures, navigate her environment and grasp and catch objects. She cannot, however, differentiate the shapes and orientation of objects. Despite the inability to discern whether an object is vertical or horizontal, she can correctly orient her hand to insert a card into a randomly angled slot. Furthermore, she can’t tell a smaller from a larger object but her grip correctly adjusts to the size of an object in the process of grasping it. The ability to grasp indicates that DF’s quick processing zombie brain is functioning to correctly adjust her grip in real time as her hand moves closer to the object. Additional support can be derived from the fact that she can correctly orient her hand to perform an action on an object, all the while lacking the ability to conceptualize and discuss the same objects orientation. This evidence indicates damage to the ventral stream.

On the opposing end of the spectrum, a patient called IG suffers from dorsal stream damage. IG lacks the zombie agents necessary to perform actions in real-time. She has difficulty scaling the size of her grasp when reaching for items, but accurately scales it when delayed and forced to pantomime the action from memory. These two opposing afflictions show a direct correlation between ventral and dorsal actions. DF can perform immediate tasks better because her zombie agents are functional. IG can complete delayed tasks better because she has a relatively intact ventral stream.

What’s the meaning of this?

            The separability of our zombie and conscious agents in terms of function, as exposed by experimental and clinical research, provides positive evidence for a modularity of mind thesis. One version of this theory[4] states that the mind is composed of a central system, responsible for high-level processes, combined with additional modules providing low-level processing. The conscious mind would be what we might relate to the central system while the automatic zombie agents would be the auxiliary modules. We have seen that the central system may not have access to all information from zombie modules. This mental sequestration is called informational encapsulation. Informational encapsulation explains why visual illusions cannot be rationalized even though we know it to be an illusion. Physical and conceptual modularity would explain why damage to certain areas of the brain causes consequences such as those illustrated above.

Each module plays its own crucial role in directing our actions and thought processes. The zombie directs precise movements and expeditiously scans for both potential interests and dangers. Simultaneously the conscious brain is free to tackle more complex intellectual and memory-heavy challenges.  If one of these modules is damaged an individual may continue to function, although certain functions may be impaired. Damage to the conscious central system may impair our ability to think and plan, while damage to a zombie module inhibits our ability to interact in real-time. It is only through cooperation that operations of the highest and most complex cognitive capacities can occur.

All of this knowledge can be deepened with biological evidence showing that cognitively lesser animals, such as reptiles, possess physical components related to visual dorsal streams but lack evidence for the more complex ventral streams.[5]  This may indicate that the ventral stream, including consciousness, is a more recent evolutionary advancement that is unnecessary for living beings. It is a feature that most likely survived due to the benefits gained from its ability to process complex information, without detrimental loss of cognitive or neuromotor resources. By splitting the total computations between them, the parallel processing of modular unconscious and conscious agents allow us to process a huge amount of information. The role of the zombie is to free up resources allowing consciousness to do what it does best. Considering the conscious mind as the captain of a ship, he is allowed to deliberate, analyze, judge and plan. The only reason he has the resources for this is thanks to First Mate Zombie automatically handling much of the day-to-day operations. So, if you ever find you literally cannot walk and chew gum at the same time, it might be time to see your friendly neighborhood neuroscientist to see if you have a zombie problem.


[1] Koch, Christof. 2004. The quest for consciousness: a neurobiological approach. Denver, Colo: Roberts and Co, 206.

[2] Koch, Christof. 2004. The quest for consciousness: a neurobiological approach. Denver, Colo: Roberts and Co, 210.

[3] Koch, Christof. 2004. The quest for consciousness: a neurobiological approach. Denver, Colo: Roberts and Co, 211.

[4] Crane, Tim. 2003. The mechanical mind: a philosophical introduction to minds, machines, and mental representation. London: Routledge, 150-152.

[5] Ramachandran, V. S., Christopher Rawlence, Emma Crichton-Miller, and Rena Baskin. 2007. Secrets of the mind. Boston, Mass: WGBH Boston Video.

h1

THEORY LADEN OBSERVATIONS (Super Fun…?)

December 21, 2012

microscope

THEORY LADEN OBSERVATIONS

By: Mike Stayton

Objective observation seems to be at the heart of scientific endeavor. Many philosophers of science argue that observations are theory-laden, meaning they are influenced by, or even depend upon, preconceived theories held by those conducting the research. I will argue that observations are inherently dependent on theories and cause problems for the objectivity of science. First, I will describe a common view of science, introducing the problems posed by the idea of theory-laden observations. Second, I will expound on the views of Thomas Kuhn, counter objections by Jerry Fodor and lean on Immanuel Kant to show that observation is necessarily objective. Finally, I will pose solutions to offer solace to the empiricist.

Because of its methodological approach of deriving knowledge from carefully collected and observable facts, science is almost universally viewed as an especially certain form of knowledge. Most individuals, however, hold a naïve view of science based on the following three assumptions[1]:

i.         Facts are directly given to careful unprejudiced observers via the senses.

ii.         Facts are prior to and independent of theory.

iii.         Facts constitute a firm and reliable foundation of knowledge.

These assumptions attempt to explain the certainty of science. If brought under scrutiny these assumptions are found to be unsound.

Though our visual mechanics are similar, each of us can remember an instance where we disagreed with someone on an object we saw before us. A simple example would be looking at clouds as a kid. I see a spaceship while my friend, Robert, sees a fish. Though the light is forming the exact same image on both our retinas, we interpret that image differently. We view the clouds differently because we interpret the world based on individual mental frameworks. This example shines a light on human subjectivity, undermining the idea that facts simply present themselves.

This cloud example illustrates that perception is not simple and that our minds play a role in interpreting even basic experiences. Let us revise its complexity to make a further point. I see the spaceship because I love astronomy. If Robert is trained in meteorology he may see a cumuliform cloud. This cloud informs him that it’s probably going to rain. Though the cloud is the same our celestial observations are filtered through our minds and combined with relevant interests and facts. Robert’s observation is a good example of an observation that is theory-laden. He is trained to understand the information that clouds can convey. If well trained, he may even do so intuitively. These personal filters, both conceptual and theoretical, interpret what we see on an individual basis. This adjusted example shows that facts may be prior to our perceptual experience.

If the first two assumptions are questioned, a direct challenge is posed to scientific reliability. Thus the problem with accepting the above arguments is that science would be found on an unreliable source, undermining its objectivity. If objective human observation is impossible then science cannot be considered a firm and reliable foundation of knowledge.

It could be argued that professional observers, such as scientists, should be able to separate theories and expectations from their observations. Thomas Kuhn[2] utilizes psychological research to claim that even trained, careful, observers can be influenced by preconceptions and expectations. In the experiment, scientists are shown simple playing cards some of which are anomalous (e.g. a red ace of clubs). They are asked to report their observations after each card. The results show that the anomalous cards were regularly described as normal. For example, a red ace of clubs may be normalized as a black ace of clubs or red ace of hearts. Even these trained observers are synthesizing their perceptions with expectations. Their expectations categorize the anomalous cards into previously experienced categories. Though many eventually noticed the irregularities, some never detect them. The explanation given is that some are unable to acquire the anomalous concept (red club). This infers that not only is perception influenced by basic concepts but that expectations help form our perceptual experience. Just as in the cloud example, scientists with differing backgrounds observing the same phenomenon can also have different observations depending on their preconceptions. It is more difficult for one to perceive a red club if one’s accepted beliefs do not accept their existence.

Jerry Fodor claims that theory-laden observations that rely on theoretical assumptions interpreting perceptual sensations, such as Kuhn’s, only occur in low-level perceptual processing[3]. He uses visual illusions to bolster his claim that most held theories, even those explaining illusions, have no affect on our perception of a given illusion. He extends this claim to state that theories that do influence our perception are not high-level scientific theories. Fodor’s argument rests on the modularity of mind theory[4] that posits individual modules, like car parts, compose the inner workings of the mind. Fodor limits his version of modularity to low-level modules including perceptual observation. Each low-level module functions as an inherent and unconscious mental processor with no access to higher-level concepts, such as theories. The low-level modules send processing results to the higher central cognitive processors. It is only there that more complex concepts may be applied to observations.

While modularity of mind seems perfectly reasonable, I would argue that Fodor’s view is overly simplistic and underestimates the complexity necessary for even the simplest cognitive operations. Returning to the sky, it could be said that on the lowest level we all see “Cloud.” If I experience a rocket and Robert experiences a fish, it could be inferred that low-level processes are less affected by concepts than higher-level cognitive processes, such as those that give rise to experience, are most susceptible to preconceptions. Pre-existing mental structures contribute more so to what is consciously experienced than what is directly seen.

Immanuel Kant became one of the first to attempt analyzing the function of reason in organizing our experiences[5]. Kant’s view supports the idea of theory-laden observations by arguing that the human mind plays an active role in constructing experience. Kant noted that physical sensations are given to us via sensory organs. These sensations are filtered through a pre-existing mental framework conceptually categorizing the incoming sensations into what we experience. This leads to the counter-intuitive claim that senses cannot provide direct knowledge of objects in the world. Instead, “objects must conform to our knowledge[6]” as all sense data are filtered through our existing mental interpretation of the world. Thus we cannot know things objectively as they are outside the mind. We can only know things subjectively as they appear to be. Information about an object can only be acquired after the perceptual image is processed by the mind.

With all this doubt thrown upon the objective observation it seems science is in dire straits. Alan Chalmers claims subjective observation does not undermine scientific knowledge[7]. The inevitability of subjectivity does mean that observers must practice caution. Countering methods have been developed. Researchers carefully attend to established observation methods ensuring observations are repeatable and testable by others. Although the theory-ladeness of observation has shown that empirical observation is inherently flawed, it has not undermined science as humankind’s most reliable method for making sense of the world. Instead it taught us that if we must strive to ensure we guard against human imperfection.


[1] Chalmers, A.F. What is This Thing Called Science. Third Edition. Indianapolis: Hackett Publishing Company, 1999.

[2] Kuhn, Thomas S. The Structure of Scientific Revolutions. Fourth. Chicago: The University of Chicago Press, 2012.

[3] Godfrey-Smith, Peter. Theory and Reality. Chicago: The University of Chicago Press, 2003.

[4] Robbins, Philip. Modularity of Mind. Summer 2010. Robbins, Philip, “Modularity of Mind”, The Stanford Encyclopedia of Philosophy (Summer 2010 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/sum2010/entries/modularity-mind/&gt;. (accessed November 10, 2012)

[5] Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp Smith. Basingstoke: Palgrave Macmillan, 2007.

[6] Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp Smith. Basingstoke: Palgrave Macmillan, 2007.

[7] Chalmers, A.F. What is This Thing Called Science. Third Edition. Indianapolis: Hackett Publishing Company, 1999.

%d bloggers like this: