2008 April 22
at 5pm
Reducing Cognitive Control to Chunks of Cortex
Catherine Stinson
Attention is typically divided into two types that roughly correspond
to 'bottom-up' and 'top-down' processing. Although attention is
a heavily researched topic within psychology and neuroscience,
progress has only recently begun to be made on the top-down variety,
called 'executive attention' or 'cognitive control'. This variety
of attention is thought to be the capacity that manages the goal-directed
planning of complex action sequences, and that directs the allotment
of processing resources during introspection. Much of the research
in this field points towards these capacities being implemented
by particular regions of frontal cortex, including anterior cingulate
cortex. Although most of the conclusions being drawn in this area
of research are modest--that this area is "involved in"
executive attention, for example--some of the labs working in
this area are professing to have located the "source"
of attentional control. By this they seem to mean that they have
reduced attentional control to the activities of a particular
chunk of brain (or at least that they have taken an important
first step in localizing the chunk of brain, even if it remains
to work out the mechanisms functioning there). In this paper I
examine these claims and the evidence being put forward for them.
It is not immediately clear what notion of reduction is being
invoked in these claims, nor what sort of explanation is being
offered. To answer these questions, I look to various papers on
reduction and mechanisms by Schaffner, Machamer, Darden, and Craver.
I argue that the attempt at reduction exemplified in recent research
on executive attention relies on several problematic assumptions
and confusions. I will also discuss broader questions about the
status of explanations in psychology, and what sorts of links
can and should be made between these and neurophysiological theories.
2008 April 18
at 5pm
Maxwell demons and the epistemology of thought experiments
Rawad Skaff
How can we gain knowledge about nature just by thinking about
an imaginary, hypothetical or counterfactual, situation or scenario?
This is the main puzzle concerning the epistemic power of thought
experiments (TE).
I will start by analyzing the different TEs known as "Maxwell
demon" as well as their purposes. My first aim is to test
a criterion required by most philosophers in the literature on
TEs: that the scenario of the TE is "possible". Generally,
the possibility criterion is required in order to justify the
use of TEs and to sort successful TEs from unsuccessful ones.
In other words, it is claimed that an impossible scenario leads
to an unsuccessful TE since it is hard to see how we can obtain
knowledge from them. I argue that the possibility of the scenario
fails as a criterion for their success. My conclusion is based
on the claim that Maxwell original demon is a good TE. This will
bring me to my second aim which is to test several approaches
to TEs in science; in particular those who claim that TEs are
logical arguments or limiting case of 'real' experiments. If my
claim concerning Maxwell original demon is correct than TE can
not be logical arguments neither limiting cases of experiments.
2008 April 4
at 4pm
The Causal Markov Condition: Should You Choose to Accept It?
Karen Zwier
The Causal Markov Condition (CMC) is an axiom specifying a relationship
between a causal graph and the probability distribution over its
vertices. In recent years, there has been much debate (particularly
in BJPS) over this axiom. In this talk, I classify the criticisms
against the CMC into two groups. The first type of criticism is
metaphysical: how do we know that causal relationships, in reality,
always exhibit the precise statistical relationships specified
by the CMC? Could there not be situations, for example, in which
two distinct effects of the same cause are correlated, even when
all common causes are given? The second type of criticism involves
the application of the CMC: Why should we continue to use the
axiom when there are many known "counterexamples"?
In regard to the first type of criticism, I will not attempt
to argue on a metaphysical basis that causal processes must be
independent. I will argue, however, that use of the CMC is reasonable
given the aims of science. I will also give several arguments
to motivate its use. In regard to the second type of criticism,
I will confirm the accusation that CMC is not applicable in every
case. I will, however, show that many cases that have been cited
as "counterexamples" of the CMC are actually instances
of misapplication or misunderstanding. Furthermore, I will argue
that the strength of the CMC lies precisely in an ability to discern
appropriate situations for its use.
2008 March 21
at 5pm
Types of Domains of Knowledge
Selja Seppälä
The research project I am presenting is part of my doctorate
dissertation on the conceptual characteristics and structure of
terminological definitions. Terminology is both an applied activity
and a theoretical discipline. As an applied activity, it consists
in writing mono- or multilingual dictionaries for specialized
domains (sciences, activities, practices, etc.). Thus, definition
writing plays an important role in conveying the knowledge (the
concepts) of the described fields and their overall organization
(the underlying conceptual system). The writing of these definitions
is done through the synthesis of a number of defining or knowledge
rich contexts extracted from domain specific texts (MEYER 2001).
Though subject to a small number of formal restrictions (such
as a single sentence), their writing does not follow precise rules,
at least at the content level, leaving this activity to the intuition
and experience of the terminologist. Therefore, open questions
for work in the field of terminology, as a theoretical discipline,
are the following: what makes a context more relevant than another
to define a certain concept? Moreover, since the definition is
rather limited in space, what are the characteristics of the concept
to be included in the definition?
One of my working hypotheses is that the conceptual structure
of a definition is at least partly dependent on the type of domain
to which it is attached. To be tested, it needs however further
clarification: what are types of domains of knowledge? My main
objective is, therefore, to try to find what characteristic features
could be used to define different types of domains. I will first
see what is meant by domain in terminology. I will then make a
brief overview of some attempts to define types of domains and
examine other possible criteria that could serve that purpose.
I will finally delineate possible bottom-up methods to empirically
define types of domains.
2008 February
8 at 4pm
Status Bias and the Matthew Effect in Philosophy
Jason Byron
The "Matthew Effect" was introduced by Robert Merton
to describe the disproportionate credit high-status scientists
get for their work, at the expense of low-status scientists for
work of the same or similar quality. The effect can be generated
by biased attention (if the scientific community is not be aware
of the work of low-status scientists and so accords it little
credit) or by biased assessments of work already known to the
community in favor of high status scientists. I present data from
top philosophy journals that show a very strong negative correlation
between blind review policies and the percentage of high-status
authors published (r = -0.90, P = 0.001). This result suggests
that philosophers assess the quality of research papers at least
in part according to author status, even when carefully attending
to the content of the papers, as for example during peer review.
I argue that this form of status bias, even if appropriate in
science, is generally inappropriate in philosophy, given general
differences in the epistemic communities of scientists and philosophers.
