| Distributed
Thinking: A Symposium. University of Hertfordshire, 19 June 2008
Distributed thinking is concerned
with how we do what we do in the world. Making action and interaction
into the domain of analysis, we ask what humans do with what they
know (knowledge) and how their intelligence plays out. Rather than
focus exclusively on what goes on inside the head, we also attend
to the resources used in acting and thinking. We came to distributed
thinking from different perspectives. Starting from investigation
of how children learn to talk, the nature of language impairment
and the nature of concepts , we gradually moved away from traditional
cognitive views. We now see language as a resource available for
problem solving, which, much of the time, may also stand in the
way. Emphasis is thus given to how children learn to talk and, by
solving problems, come to think. Experience is central to action
and language functions as a major developmental resource. Language
impairment is not separable from thinking (as often assumed) but,
rather, intrinsic to distributed thinking.
Whereas distributed cognition emphasises
processing and models of working memory, we regard cognition as
spreading beyond memory and perception/ attention. These processes
take place in neural timescales as evidenced by reaction time studies,
fmri studies and so on. On the distributed view, however, emphasis
falls on how activities are integrated across timescales. Distributed
thinking thus incorporates rapid neural reactions such as eye movements
or gaze with more complex cycles of action. Experience is central
to how, in lived time, we think and language together while integrating
external resources with concerted activity. To pursue investigation
of distributed thinking, therefore, a primary challenge is to develop
new methodologies.
In cognitive psychology, we look
at human cognition in tasks during where brain, body and world often
interact. Our challenge, then, is to tease out the cognitive roles
of brain, body and world. To this end, we invoke both (inner) capacities
and (what can be called) co-acting assemblages. We thus need new
ways of examining –and quantifying –observations. We
wish to investigate: ( 1) how the same assemblages adapt to different
environments; and 2) how activity varies as environmental conditions/
human components change. Further, we need a model of assemblages
that may integrate (a) executive control systems); (b) neural slave
systems; and (c) bodily processes (micro-action). As a result body-world
relations arise as brain-side functions mutually regulate the bodily
processes that allow shifting forms of control over the system’s
external components.
Speakers:
(Use name link to go to speakers
abstract)
Anthony,
Sue (Hertfordshire)
Baber,
Chris (Birmingham) - download presentation
/slides 
Bonnefon,
Jean-François (CNRS)
Clowes,
Rob (Sussex)
Cowley, Stephen (Hertfordshire)
- download presentation
/slides
Fioratou,
Evie (Aberdeen)
Healey, Patrick (Queen
Mary) - download presentation
/slides 
Ryder,
Nuala (Hertfordshire)
Perry,
Mark (Brunel) - download presentation
/slides
Spiekermann,
Kai (University of Warwick) - download presentation
/slides
Vallee-Tourangeau,
Fred (Kingston) - download presentation
/slides
Abstracts:
Sue Anthony
and Nuala Ryder: Hertfordshire
"Thinking ain't (all) in the head"
Understanding performance on a number of cognitive tasks such as
problem solving, memory and comprehension has traditionally focused
on internal representations and the processes that act upon them.
Results of a number of studies suggest that this approach provides
only a partial explanation of how people perform on these tasks.
What seems to be missing is a systematic account of the role of
their experiences with past and present external resources. These
resources may include people, the physical environment and specific
artefacts. The presentation will discuss this apparent oversight
in cognitive psychology in the light of methodological challenges
that are posed.
return to speakers
Chris Baber:
Birmingham
"Crime Scene Examination as Distributed Cognition: theory and
technology"–
There are three main strands to this talk. First, I will argue that
the process of Crime Scene Examination can be usefully considered
as a form of Distributed Cognition in that the CSE needs to translate
artifacts in the world into evidence and this requires an interaction
between objects-in-the-world and the CSE's interpretation of what
constitutes evidence. Second, the notion of 'evidence' requires
a broad consideration of cognition as being performed by several
parties within the criminal investigation process.
This requires not only consensus on what constitutes evidence but
also managing the exchange of information and artifacts. Third,
we have developed wearable computers that can support the recovery
and logging of artifacts in a manner that we believe can support
distributed cognition.
return to speakers
Bonnefon,
Jean-François: CNRS
"An interdisciplinary approach to decision makers' trust in
vague and partial expert recommendations mediated by computer platforms."
Sylvie Leblois, Jonathan Ben-Naim, Jean-François Bonnefon,
Andreas Herzig, Emiliano Lorini
Important decisions are rarely made in isolation, in particular
because individual decision-makers may not have all the knowledge
and expertise they would need. In such a situation, it is not uncommon
for the decision-maker to collect the recommendation of one or several
experts before making the decision. Nowadays, this consultation
can take place through dedicated web-based computer platforms. It
has been claimed that the most important issue raised by this medium
of
distributed cognition is that of Trust, as it is essential to understand
the determinants and the consequences of the trust that the decision-maker
develops in the expert recommendations he or she is receiving through
the platform. These questions are even more salient in situations
where there are several potential experts to consult, in
situations where experts commonly make vague recommendations, and
in situations where experts cannot be considered totally impartial,
because they are likely to have a stake in the consequences of the
decision they are informing. We adopt an interdisciplinary approach
to the issue of trust in vague and partial expert recommendations
mediated by computer platforms. More precisely, we describe how
the problem can be tackled through a combination of the experimental
and formal methods of cognitive psychology and artificial intelligence,
and we review the methodological innovations required for that task,
both within and accross our disciplinary borders. The art marketprovides
us with a paradigmatical case where naive sellers can seek the recommendations
of experts through computer platforms, where the experts cannot
usually give point-estimations of the value of a piece, and where
the experts may have a personal interest in the value that is attributed
to a piece.
return to speakers
Rob Clowes:
Sussex.
"Distributed Thinking and Internalisation"
It seems there is some growing consensus among the research community
that there is such a thing as distributed thinking. This is presumably
supposed to be something more than many individuals thinking about
the same thing separately. But what is distributed thinking? Core
cases of what people refer to as thinking seem to be at least attributed
by many people as some sort of 'inward' process albeit with external
correlates (Hurlburt, Koch, & Heavey, 2002). Speaking of distributed
thinking could of course be a redefinition of terms, building on
past talk of distributed cognition (Hutchins, 1995), or the extended
or external mind (Clark & Chalmers, 1998). By analogy an analysis
of distributed thinking might proceed from some of the principles
which are currently being deployed here (i.e. the use of the parity
principle). And yet, inwardness itself stands in some need of explanation,
and further might help us clarify what distributed thinking might
actually be. I will look at this question with the help of some
minimal simulation models that might help explain some of the dynamics
of medium time-scale human thought, and how even "inward"
thinking can be seen to depend in some ways upon social action.
Clark, A., & Chalmers, D. (1998).
The Extended Mind. Analysis, 58, 10-23.
Hurlburt, R., Koch, M., & Heavey, C. (2002). Descriptive Experience
Sampling Demonstrates the Connection of Thinking to Externally Observable
Behavior. Cognitive Therapy and Research, 26(1), 117-134.
Hutchins, E. (1995). Cognition in the Wild. Cambridge MA: MIT Press.
return to speakers
Evie Fioratou,
University of Aberdeen
"Essential DEEDS for Improving Anaesthetists' Cognition."
In this presentation, I will explore the role of the DEEDS (Dynamical,
Embodied, Extended, Distributed, & Situated) approach in improving
anaesthetists' cognition in the operating theatre. I will argue
for the importance of studying the dynamic interaction between the
anaesthetist, the patient, the surgical team and all the external
resources in the OR environment (e.g., monitors, charts) in order
to understand the implications of such dynamic an interaction on
cognition. Furthermore, I will consider the development of cognitive
skills in anaesthesia training from a DEEDS perspective, in which
"scaffolding", embodied, and emergent experiences play
crucial roles. The practical relevance for applications to simulation
training will thus be explored and future research ideas will be
suggested.
return to speakers
Pat Healey:
Queen Mary
“Interactive Misalignment: The Role of Repair in the Development
of Group Sub-languages”
The argument developed in this paper is that interaction provides
a key source of constraints on language change. In particular, that
the local processes involved in the detection and resolution of
interpretive misalignments directly constrain the kinds of sub-language
that people can develop. The main evidence for these claims is drawn
from two sets of experimental studies involving collaborative tasks
in which communication is either graphically or verbally mediated.
I present evidence from a set of group sub-language experiments
that particular histories of direct interaction -- independently
of task exposure or experience -- play a critical role in group
sub-language convergence. I then present evidence from a second
set of experiments that these effects depend on specific interactional
mechanisms; people's ability to juxtapose and contrast elements
of one another's turns. Finally I argue that the primary function
of these mechanisms is to detect and resolve misalignments.
return to speakers
Mark Perry:
Brunel
"Making use of things to think with: the mobilisation of social
and physical resources in enabling distributed cognition"
The presentation will explore the role of distributed cognition
in human activity, and how we as researchers can come to understand
more about the role of embodiment and situatedness in human behaviour.
As a research or theoretical tool, DCog explores the interplay between
mind and body, and the physical, social and cultural contexts within
which activity occurs, and which form an intelligent system as an
emergent property of this interaction (or, at least, can be usefully
considered as forming a meta-cognitive system). Here, the role of
internal mental cognitive behaviour is seen in managing the co-ordination
of representations and processes. It is in this area that the presentation
will focus - how the resources in settings are co-ordinated by human
actors and the physical manipulations that are applied to enable
the symbolic transformations necessary for distributed problem solving.
Yet whilst it is itself seen as a correction to the problems of
traditional cognitive science, DCog itself has some practical limitations
to its use and a number of academic questions regarding its theoretical
foundations, and these will also be addressed.
return to speakers
Kai
Spiekermann: University of Warwick
"Judgement Aggregation, Social Choice, and Distributed Thinking"
Social choice theory aims to explore the possibility of aggregating
individual preferences into collective preferences. Aggregation
procedures should meet certain conditions, in particular regarding
rationality, fairness, responsiveness, and non-manipulability. Social
choice theory shows that aggregation is often impossible, given
some seemingly weak conditions for the aggregation procedure. Arrow’s
Theorem, for instance, shows that there is no non-dictatorial aggregation
of preference rankings if the aggregation function is required to
produce reflexive, connected and transitive social orderings and
meets three conditions: universal domain, independence of irrelevant
alternatives, and the weak Pareto principle. In recent years, judgement
aggregation has emerged as an important area of social choice theory.
Judgement aggregation is concerned with aggregating sets of individual
judgements over logically connected propositions into a set of collective
judgements. Again, it can be shown that even seemingly weak conditions
on the aggregation function lead to impossibility results. This
implies that the step from individual judgements to collective judgements
faces dilemmas and trade-offs between different desiderata, such
as universal domain, rationality, epistemological quality, fairness,
et cetera. These dilemmas challenge us to decide which conditions
we should relax. The typical application for judgement aggregation
is the problem of group decision making. Juries and expert committees
are the stock examples. However, the relevance of judgement aggregation
goes beyond these cases. It conveys easily to other settings where
distributed sets of logically connected information must be aggregated
into collective information, that is to cases of distributed thinking
and collective intelligence.
return to speakers
Fred Vallee-Tourangeau:
Kingston
“Verbal Fluency and Interactive Skills in a Word Generation
Task”
In attempting to solve a wide variety of tasks, people naturally
seek to modify their external environment such that the physical
space in which they work is more amenable or ‘congenial’
to achieving a desired outcome. Scientists build and tinker with
physical models of their object of investigation, an interactive
process that provides important perceptual feedback that drives
the generation and evaluation of novel hypotheses. On a more quotidian
level, people use simple artifacts or alter the physical environment
to enhance their prospective memory or facilitate the execution
of everyday tasks. In doing so, a reasoner delegates some of the
information storage or computational costs onto her immediate surroundings,
and the problem solving activity is distributed among resources
internal and external to the reasoner.
Attempts to determine the effectiveness
of certain artifacts or spatial reorganizations in aiding reasoners
solve problems must be relativised to the difficulty of the task
and the cognitive abilities of the reasoners. These factors were
examined using a simple word production task with letter tiles.
Two sets of tiles that differed in terms of word-production difficulty
were selected. Participants were asked to produce as many words
as they could within a finite time period for each letter set. In
one group, participants were encouraged to rearrange or touch the
tiles when producing words, and in the other group, participants
could not interact with or point to the tiles. Participants were
further split in a low and high verbal fluency group as a function
of their score on the Thurstone word fluency test taken at the end
of the experiment. In the high fluency group, letter rearrangement
did not improve the participants’ ability to generate words.
In contrast, in the low verbal fluency group, letter rearrangement
significantly enhanced the ability to produce new words from both
the hard and easy letter set. For these participants, the task was
more taxing and the opportunity to restructure the letter set substantially
elevated their performance.
return to speakers
|
