Kate
Arrington
Department of
Psychology
Lehigh University
Research Interests
For much of the history of human
inquiry, questions of will and choice have been the purview of
philosophers and
theologians. Applying scientific methods to the study of volitional
behavior
presents a challenge: The control inherent in scientific
investigation limits participants’
free will. Faced with this dilemma, psychologists have danced around
issues of
voluntary control of behavior. However, a hallmark of human behavior
is the
flexibility to choose how to respond or even not to respond to
stimuli in our
environments. This choice of action requires executive control of a
cognitive
system designed to receive signals from the external world,
integrate these
signals with internal information and goals, and generate behavior.
The
mechanisms of cognitive control that allow for this flexible
behavior within a
complex world are only beginning to be understood (Brass &
Haggard, 2008;
Miller & Cohen, 2001). The questions that drive my research are
at the core
of the scientific study of cognitive control. What are the interactions between stimulus-driven and
goal-directed
forces that determine cognitive control? Why does the ability to
implement
cognitive control vary across individuals and across time and
situations within
the same individuals? What are the mental representations and
processes that
guide task selection and performance? While my research is
primarily aimed
at understanding cognitive functioning through analyses of
behavioral data and mathematical
modeling, I have strong interests in studying the neural basis of
cognition and
am pursuing these interests through collaborations involving
functional
neuroimaging research. I believe that the converging methodologies
approach
that defines the field of cognitive neuroscience provides an
investigative
environment where knowledge of neural function can inform cognitive
theory.
Multitask environments are ideal
situations in which
to study executive control. When multiple behavioral paths are
available, cognitive
control is necessary to coordinate input of perceptual
information, retrieve information
from memory, and direct behavior to competing tasks. Task
switching
methodologies are widely used to study control processes; however,
they are
limited in their application to the study of voluntary behavior,
because the
experiment typically dictates the task to be performed on each
trial. In
response to this limitation, I developed the voluntary task
switching (VTS) paradigm
during my postdoctoral fellowship with Gordon Logan (Arrington
& Logan,
2004-a; 2005 see accompanying materials). In VTS, participants are
provided
with multivalent stimuli and allowed to choose what task to
perform on each
trial. Typically, the participants are instructed to perform each
task equally
often and in a random order. These instructions are intended to
encourage participants
to actively choose the task to perform on every trial and to
discourage them
from selecting a strategy of performing one task for a large
number of trials or
alternating strictly between tasks. Within this paradigm, task
performance measures
of reaction time (RT) and accuracy show costs in performance
associated with
switching tasks. As in standard task switching studies, these
switch costs
likely reflect a variety of processes including time needed to
establish a new
mental set, or task set (Rogers & Monsell, 1995), and
competing activation
from recently active task sets (Allport, et al., 1994), but may
also include
choice costs (Arrington & Logan, 2005). VTS also provides
measures of task
choice. Choice behavior is reflected both in the particular task
participants
choose on a given trial type and in the transitions they make
between tasks, or
the switch probability. The ability of this procedure to measure
both task
choice and task performance provides a comprehensive approach to
the study of
cognitive control processes engaged during volitional, multitask
behavior. The
idea that control processes may serve distinct roles of
“endogenous control of
task set, in the sense of controlling which
task is performed” and “‘task readiness’ …measured by speed
of performance”
(Allport & Wylie, 2000) has been largely ignored in the
standard task
switching literature over the past decade. In my own work, I
consider a
distinction between task selection and task preparation. Task
selection refers
to the intent to perform a particular task and the associated
activation of a
task-level representation. Task preparation refers to the
activation of the
task set in terms of the attention, decision, and response
parameters necessary
to perform the task. Ongoing research in my lab is aimed at
understanding these
processes and specifying a model of behavior in multitask
environments.
Current
Projects
Volitional control over planned sequences of actions using a
voluntary task span procedure
The balance of goal-directed and stimulus-driven influences on task
selection in tasks with unequal task difficulty