Archive for December, 2012

Semester 6 Summary: Will or No

Sarah Williams provided this great perspective on the scope of semester 6, in which we discussed Cognitive Neuroscience and Free Will.  She highlights elements of each discussion.

6.1: Sam Harris’ view: There is no agency. This is the case for two reasons:

1)    Agency can only be conscious and never unconscious. There is no conscious agency because all conscious choice is determined completely by unconscious causes. Some empirical support of this claim is found in the Libet (1983), Wegner (2002) and Ebert & Wegner (2010) experiments.

2)    Contemporary physical theory does not support the existence of agents.

Some problems with Harris’ view could include the fact that we seem to have conscious control over conditioning our impulse control and over how our brains respond to stimuli. If there was support for unconscious agency it would also be problematic for the argument. The causal chain of our behavior consists of several stages. Harris’ claim requires that the first cause be at the level of our neurobiology. This seems arbitrary, and Harris does not offer support for why this must be the case.

6.2 Daniel Kahneman’s view:

Human cognition operates using two systems, System 1 (for immediate, automatic responses) and System 2 (for less immediate, more complex calculations). Both systems use heuristics, but they are automatic for System 1 and conscious for System 2. A brief summary of the two systems:

System 1: automatic, short time scale, no voluntary control or conscious effort needed, metabolically less costly, trainable by conditioning, handles familiar tasks.

System 2: conscious, deliberate and requiring mental effort, metabolically more costly, handles complex and novel tasks, can take over some system 1 tasks.

From an evolutionary standpoint, having a fast, metabolically cheap system that can be trained to handle repetitive tasks and situations, and a slower, more deliberative system that can train the fast system seems like a useful adaptation given our limited ability to process stimuli (invisible gorilla experiment) and our limited metabolic resources (ego depletion).

Kahneman’s paper added a new dimension to the discussion on free will. System 1 handles seemingly unconscious activities and System 1 can be trained by our conscious System 2, does this mean that there ultimately is a form of unconscious agency?

6.3 Shaun Nichols’ view:

Most people think that they have agency, but seem to mistake a difference in the antecedent conditions with actual agency. These intuitions track indeterminist agency/free will. We probably have these intuitions because they most closely describe the way we experience our actions. When we are at a decision point, we experience that all of the antecedent conditions are fixed and yet we still believe that we have the option to choose differently.

However, these intuitions do not adequately justify indeterminism for a number of reasons:

  • Our experiences could be informed by determined causes
  • The indeterministic perspective could be learned/conditioned
  • Raw experience as a counterfactual doesn’t really make sense. Something like a toothache is primary, whereas counterfactuals are a second order analysis of primary experiences.
  • Spinoza: We believe in indeterminist free will because we are ignorant of the actual causes of our actions
  • We don’t have introspective access to all the factors that influence our choices (in response to Glimcher’s argument)

6.4 Roy Baumeister’s view:

Choice is not an epiphenomenon, lacking causal power over our actions. Conscious thought does influence actions (although, so does unconscious thought). Some examples: mental rehearsing and simulation of actions, planning, reflection, counterfactual reasoning, and overriding automatic behaviors. Conscious thought and unconscious thought have a two-way influence over each other via priming (unconscious influences conscious) and conditioning/learning/synaptic plasticity (conscious influences unconscious).

For dualists, this influence could also be non-physical.

Baumeister is attempting to counter the view that conscious thought is a “steam whistle,” having no influence on our behavior and existing as an epiphenomenon of our unconscious processes.

6.5 History of Will:

The views of the Greeks:

There were two major camps that divided over whether choice was found in the head/brain area/organ of our bodies or the heart area/organ. Plato was in the “head” camp. He located virtue/masculinity/energy in the chest. This contrasts with Aristotle, who located all three aspects (Rational/Motive/Nutritive) in the heart. For Aristotle, the brain was merely a cooling device for the blood.

Some historical views:

Thomas Aquinas separated the processes into worldly/imperfect and extra-wordly/perfect. Our senses perceived the data from the world and we processed this information with our common sense (physical), while our intellect perceived the perfect realm of ideas (spiritual – akin to Plato’s Forms). Only humans have intellect, while animals and humans have common sense. Descartes combined common sense, intellect and will into the concept of the mind and located it outside the physical realm. Contemporary neuroscientists suggest that there is no one seat of consciousness, will or identity.

6.6 How to define agency:

  • If agency is defined as intent, this implies a relationship with another entity, living or otherwise. Atoms could not have agency, however, because they must follow a set of rules.
  • If agency is the ability to make rules, then it can be easily related to morality.
  • If we have an attitude of approval toward some options or the consequences of those options more than others, morality could be a way to express those attitudes. (I phrased this differently than the summary on the blog because I thought it was heavily biased toward a consequentialist interpretation of morality. I use the word attitude technically here.)
  • Whatever the choice available, value and reason ought to be used to evaluate which choice should be picked. If we lose the concept of agency, we seem to lose the concept of rationality. The sense of accountability and responsibility are both grounded in our sense of agency.

What if there is no such thing as agency?

Perhaps Spinoza is right: we invented agency as a way to make sense of cognitive dissonance or to explain our experience of the world. To relieve ourselves of this misconception would not necessarily have any impact on the institutions of punishment or morality.

Perhaps the question of agency and free will hinges on whether it is possible to have a bifurcating universe, as agency requires this. Sub-atomic physical stochasticity complicates the question of a bifurcating universe, as ultimate causes are obscured and the limits of observation may cover true determinism, contingency, or will.

With the advances in neuroscience, social behavior theory and complex dynamical systems theory, how can moral agency, one of the classic examples of human “uniqueness”, be reconstructed?

6.7 Emergence:

Moral agency may be an emergent property of consciousness.

Weak emergence: Complex systems produce patterns that, by convention, are modeled or described in terms of higher-order properties: i.e. human thought emerges from interactions between large numbers of neurons, social insect colonies emerge from the interactions of many individual insects (I think this is an entailment relation from the lower to the higher processes). Weak emergence is widely accepted and uncontroversial.

Strong emergence: the higher-order phenomena that emerge from lower level processes are not causally determined by them (this is a supervenient relation, see: http://plato.stanford.edu/entries/supervenience/ – 3.2). This has the characteristics of dualism. This is the more controversial of the two views.

Consensus on the topic of agency was not an emergent property of this forum!


Read Full Post »

The reading this week was chapter 8 of “Neuroscience, Psychology & Religion: Illusions, Delusions, and Realities about Human Nature,” by Jeeves and Brown. This excerpt deals with religious (specifically, Christian) perspectives on human uniqueness, and how these perspectives might be informed and updated in light of modern scientific developments. We began by discussing the notion of human uniqueness, and how some religious perspectives on human uniqueness compare with more secular perspectives.  As mentioned in the reading, a classical religious perspective on human uniqueness is the idea that humans are made “in the image of god,” and are therefore unique from other species. It was noted that from a secular, biological perspective, humans are certainly unique in a variety of ways – but so is every other species, in one way or another, and therefore humans are not thought to be “special” .  Moral agency is often listed as an example (or sometimes, the example) of a uniquely human attribute, and in the reading, the authors discuss various ways how one might construct a notion of moral agency in light of recent developments in neuroscience, social behavior, and complex dynamical systems.

This led to a discussion of emergence, and in particular the distinction between “weak” emergence and “strong” emergence. Weak emergence is the idea that complex systems can possess higher order emergent properties which are so difficult to describe in terms of the complicated interactions of their constituent parts that in practice it is much easier to model them in terms of a higher level of description. Examples of emergent phenomena given in the reading include human thought (which emerges from the synaptic interactions between large numbers of individual neurons), and the collective behavior of social insect colonies (which can be thought of as emerging from the interactions between many individual insects).

Strong emergence makes a much stronger claim about emergent phenomena, in that it assumes that higher-order phenomena can emerge from complex lower-level processes but are not causally determined by them, and may not be bound by the same fundamental physical laws. Unlike weak emergence, in which the higher order emergent properties of a system are assumed to in principle explainable by the complex low level processes which give rise to them (even if this is not feasible in practice), strong emergence takes the view that the emergent properties of a complex system are fundamentally irreducible and cannot (even in principle) be explained in terms of processes at a lower level of description. It was noted that this type of emergence has a somewhat dualist character (in the sense that emergent properties are treated as somehow fundamentally distinct from the components of the system that give rise to them), and in fact, the authors of the reading discuss the notion of “emergent dualism.” Strong emergence is sometimes invoked to explain phenomena like consciousness, mind, “qualia”, or agency, however these explanations (as well as the very notion of strong emergence) remain controversial. Weak emergence however, is widely accepted and uncontroversial. In the reading, the authors appear to be advocating for a sort of strongly emergent view of human thought and consciousness, and relate this to religious notions of human uniqueness and moral agency.

Finally, toward the end of the class period, in an attempt to sum up the semester, we went around the room discussing our individual beliefs about agency, in light of all that had been discussed throughout the semester. As in the first meeting, and unsurprisingly, there was no consensus regarding whether or not humans possess agency.


Read Full Post »