Transcript Conflicting Role of Technology in Memory Research

Traditional v Distributed
Cognition
in Memory Research
Heather Brown
&
Joanne Bower
Contents







Traditional approaches to memory research
Their problems – exemplar study
An Alternative - Distributed Cognition
Methodology utilised in traditional and
distributed cognition approaches
Technologies used in these approaches
Research of the future
Conclusions
Traditional Memory Research

Environmental Input
Atkinson and Shiffrin’s
modal model (1968)

Sensory Registers
Visual, Auditory and Haptic
Short-Term Memory Store
Temporary working Memory
Long-Term Memory Store
Memory as an
information processing
system
Response output
The Modal Model: Short term
memory store
Problems
 Neuropsychological
Evidence: Shallice and
Warrington (1970)

Solution
 Baddeley and Hitch (1974)
The Working Memory Model

An attentional controlling
system coordinates a
number of subsystems
Grossly impaired auditory
memory span does not
effect Long term learning
capacity
The Working Memory Model
Visuo-Spatial Sketch
Pad
Central Executive
Phonological Loop
The Modal Model: Long-Term
Learning
Problem
 Laboratory studies
suggest that
transference into the
long term memory store
is not a direct function
of time in Short term
memory store

Tulving (1966)
Solution
 Craig and Lockhart
(1972) Levels of
Processing



Shallow processing
Deep processing
Rehearsal
 Maintenance
 Elaborative
Traditional Memory Research



Evidence from laboratory studies involves
testing participants on simplistic and
unrealistic tasks.
Neuropsychological evidence shows extreme
cases often with hard to localise problems.
Research has typically focused towards the
individual using evidence removed from
context.
The Misinformation Effect



Memory can be affected by social and
contextual factors that occur after the event.
These factors can provide correct or incorrect
information, which then becomes
incorporated into memory for the actual
event.
This conflicts with traditional memory
research which ignores the effect of such
variables.
Neural activity during encoding predicts
false memories created by
misinformation (Okado & Stark, 2005):


Used fMRI scanning to try to locate which
areas of the brain were active during
encoding and retrieval of event information
and misinformation.
Concluded that whether the event or
misinformation was recalled depended on the
level of activity during encoding. Greater
activity during encoding lead to higher
chance of recall.
Technology and Methodology
used in Okado & Stark (2005)




Attempt to investigate social variables using fMRI
data.
fMRI is an example of technological advancement,
however it reinforces the traditional individualistic
approach to memory research.
The advances in technology outside of the
laboratory have increased the opportunity for all
forms of distributed cognition.
Therefore, fMRI scanning can only inform us
regarding part of the constructive memory process.
Okado & Stark (cont’d)
They Assume:
 Although the events leading
to encoding occurred in a
social framework, encoding
and retrieval occur within
individual cognition.
 Memory is a constructive
processes.
 The loss of social context
that occurs within the fMRI
scanner does not change
the underlying cognitive
processes of encoding and
recall.



However…
Cognition is not limited to the
individual, but can be
distributed across an
environment.
They discuss this process
within an individual framework.
The processes engaged in the
fMRI scanner may differ from
those experienced ‘In the Wild’
(Hutchins, 1995).
Hutchins (2000); Rogers & Scaife
(1997)
Bridging the Gap:



The assumptions made by Okado & Stark
(2005) are typical of those made throughout
traditional memory research.
However, frameworks for memory research
should encompass both the cognitive and
social aspects.
Distributed cognition is one such approach.
Distributed Cognition - Similarities
to traditional memory research


Both seek to understand the organisation of
cognitive systems including memory.
Both acknowledge a role of individuals within
this framework
HOWEVER…….
Distributed Cognition –
Differences to traditional research


Distributed cognition does not limit cognition
to the individual.
Cognition can be distributed in three ways



Socially (e.g. across co-witnesses to a crime)
Over time (e.g. eyewitness memory can be
influenced between the event and subsequent
recall e.g. at a trial).
Internally-externally (e.g. memory is influenced by
external features such as news reports).
Hutchins (2000)
Methodologies used in
memory research:
Good
control of
variables
Traditional:
Lab-based
experiments
Distributed
cognition:
Ethnography
☺
Easy
to
Use
Fast
Good
depth of
data
Good
use of
context
Accounts for
interrelationship
between
variables
☺ ☺
☺
☺
☺
Influence of technology on
cognitive research:

Brain Scanning:




Video and audio taping:



Allows good insight into individual brain
However, cannot be used to study social distribution of
cognition.
Therefore it is good at studying ONE aspect of cognition.
Equipment has become smaller, easier to use and less
intrusive.
Enables exact recording of events for later analysis.
Internet:


Can reinforce traditional studies (e.g. www.coghealth.com).
Also necessitates a distributed approach (e.g. it is an
additional medium for encountering misinformation).
Influence of technology on
everyday cognition:



Advancements communications technology
e.g. Internet, MSN Messenger, text
messaging and wireless communication.
Advancements in memory and processing
aids.
Wider incorporation of technology into
everyday life.
Norman (1993)
Research of the future:




Traditional approaches provide a useful starting
point for memory research.
However, their results may be best considered as
one part of a larger body of investigation.
Distributed cognition has evolved from traditional
memory approaches.
However, some of the current methodologies
associated with this framework are not suitable for
all aspects of memory research.
Two possible Solutions:

Virtual Environments:



Allows utilisation of traditional experimental
approaches.
Enables ethnographic research to be conducted
alongside traditional approaches (i.e. better control of
variables but observation of natural processes).
Connectionist modelling:


Used to represent information at an individual level
and beyond.
It has controllable predictive power, if accepted as a
model rather than a mirror for reality.
Rogers & Scaife (1997)
Conclusions:




Traditional memory research fails to account for
important contextual variables.
Distributed cognition provides an alternative, wider
conceptual framework for memory research.
Within this framework, traditional methodologies can be
used in conjunction with developing techniques to
provide a wider understanding of everyday cognition.
Together, connectionist models and virtual environments
can help to bridge the traditional-distributed cognition
gap.
References:



Atkinson RC and Shiffrin RM (1968) Human memory: a proposed
system and it’s control processes. In Gardiner JM (1976) Readings
in Human Memory Meuthen & Co Ltd
Baddeley (2001) Human Memory; Theory and practice revised
edition psychology press
Bick (1999) Coghealth www.coghealth.com Accessed April 2005.

Craik and Lockhart (1972) Levels of Processing: A framework for
Memory Research Journal of Verbal Learning and Verbal Behaviour
11 671-684

Hutchins (2000) Distributed Cognition
http://eclectic.ss.uci.edu/~drwhite/Anthro179a/DistributedCognition.p
df Accessed April 2005.

Hutchins (1995) Cognition in the Wild MIT Press: Cambridge.
References 2:





Norman (1993) Things that make us Smart Perseus Books: USA.
Okado & Stark (2005) Neural activity during encoding predicts false
memories created by misinformation. In Learning and Memory 12(1)
pp3-11.
Rogers & Scaife (1997) http://wwwsv.cict.fr/cotcos/pjs/TheoreticalApproaches/DistributedCog/DistCogn
itionpaperRogers.htm Accessed April 2005.
Shallice and Warrington (1970) Independent functioning of verbal
memory stores: A neuropsychological study Quarterly Journal of
Experimental Psychology, 22, 261-273
Tulving (1966) Subjective organisation and effects of repetition in
mulit-trial free-recall learning. In Gardiner JM (1976) Readings in
Human Memory Meuthen & Co Ltd.