Improvement of Interactions in Interactive Environments and

Knowledge Management from the Distributed Cognition Approach

Carmen Rearte

, Hernan Castillo

and Mabel Sosa

Facultad de Ciencias Exactas y Tecnologías, Universidad Nacional de Santiago del Estero, Av. Belgrano (s) 1912,

Santiago del Estero, Argentina

Instituto de Investigaciones en Informática y Sistemas de Información, Universidad Nacional de Santiago del Estero,

Av. Belgrano (s) 1912, Santiago del Estero, Argentina

Keywords: Distributed Cognition Approach, Interactions, Knowledge Management.

Abstract: In the field of Computer Supported Collaborative Work and Knowledge Management supported by

computer, the cognitive and sociological dimensions can not be neglected in the accessing analysis of their

designs. The material and social environment models the cognitive processes since the vast majority of them

are mediated by interaction with other agents and other artefacts. Following these premises, in this work, a

methodological framework for the analysis of the interaction between agents in a socio-technical system is

applied, based on the approach of distributed cognition, oriented to the knowledge management, in an

academic context at the upper university level. In particular, the analysis focuses on the interactions of

cognitive agents, that occur during the cycle of knowledge management (activities of using, creating,

distributing and sharing knowledge), and on how coordination, communication and collaboration is

affected, which are key aspects of group work. From the problems detected in the interactions, changes are

proposed in the design of the interface of the artefacts, where the interactions between agents in the

cognitive system are reflected.

1 INTRODUCTION

The collaborative contexts in the domain of CSCW

(Computer supported Collaborative Work) and

Computer-supported Knowledge Management have

cognitive and sociological dimensions that cannot be

neglected in the evaluative analysis of their designs.

The physical and social environment models the

cognitive processes since the vast majority of them

are mediated by interaction with other agents and

other artifacts. Computers connected through the

Internet are becoming key elements of these

interactions. Human beings generate cognitive

potential by creating and actively modifying the

environment in which cognitive operations are

performed (Feltrero Oreja, 2003). If it is admitted

that the user interface, as a key element of a

computer, is part of this cognitive environment, then

the user interface presents very relevant cognitive

characteristics that will determine the type of

activities and social relations mediated and enabled

by technologies.

Within the field of psychology, the Distributed

Cognition (CD) theory offers a framework for

analysis to describe human work systems or socio-

technical systems, in computational and information

terms, and is useful for the design of technology as

mediator of collaborative social activity. For this

reason the distributed cognition constitutes an

important theory for the field of Man-Computer

Interaction, and in particular for person-computer-

person interaction in the field of CSCW (Horsky,

2008).

The aspects of human cognition inherent to the

context of distributed work with active social agents

and tools are generally not taken into account when

designing and developing information technologies.

In this work we aim at the proposition of applying a

Framework for Analysis of Interactions between

Agents called MAIA, proposed by (Ferruzca

Navarro, 2008) for the evaluation of the usability of

a distributed work system, ReSU (Lescano et al,

2011), under the theoretical framework of the theory

of distributed cognition. This analysis would lead to

the redefinition of the components of the ReSU

Rearte C., Castillo H. and Sosa M.

Improvement of Interactions in Interactive Environments and Knowledge Management from the Distributed Cognition Approach.

DOI: 10.5220/0006497901460150

In Proceedings of the 9th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (KMIS 2017), pages 146-150

ISBN: 978-989-758-273-8

system, based on the foundations of distributed

cognition in the applied analysis framework. This

analysis will focus on detecting the inconsistencies

that may exist during the interactions occurring in

the process of knowledge management (use,

creation, distribution) and on discovering the nature

of these interactions through the application of an

adapted distributed cognition model. The aim is to

improve the technological tool in order to achieve a

better cognitive distribution that favors the inter-

actions involved in the management of knowledge

when the partakers work in a collaborative group.

2 MOTIVATION AND

BACKGROUND

This work is part of a larger research that tries to

determine the way in which the physical and social

environment models the cognitive processes of

people, when they work with artifacts or with other

agents in technology mediated environments. It

specially aims at enlightening the aspects of

distributed cognition and their incidence in contexts

of computer-mediated collaborative work and

knowledge management.

Taking into account this vision and

interdisciplinarity, we will pinpoint some proposals

that are more related to the treated topic.

In (Feltrero Oreja, 2003), a global perspective of

philosophical and cognitive type is developed for the

evaluation of the interfaces in the IHC field. It is

here considered that people create cognitive

potential through a dynamic transformation of the

environment in which cognitive operations are

performed. If the user interface is part of the

cognitive environment, the ability to autonomously

and actively configure the interface and to exploit all

its possibilities is revealed as a basic cognitive tool.

In this sense in (Feltrero Oreja, 2003), general

philosophical and axiological criteria are proposed

so that the design of the interfaces in the field of

human-computer interaction considers mechanisms

such as referential multimodality and dynamic

configuration, according to their own needs and

skills. It provides general value principles that guide

the design of interfaces.

Another analytical framework is the one

developed by (Hussain y De Brujin, 2010), based on

the CD approach to guide the evaluation of a

collaborative design task around an interactive table.

The theoretical approach is useful to understand how

people collaborate around the table by analyzing the

relationship between actors, artifacts and the

contexts in which interactions occur. It is possible to

identify the representations that flow through the

functional systems as objects of analysis, which

allow the researcher to reason about the design and

cognitive artifacts within the same conceptual

framework. This analytical framework provides a

deep understanding of how representations were

created, and how this framework contributes to

problem solving in a collaborative design

environment, ultimately identifying the design

features that best support collaborative activities. It

is also useful for defining the limits of the unit of

analysis and for guiding the analysis of the data.

In contrast to the particular-oriented approach,

that is generally used for the application of

distributed cognition, in (Nobarany et al, 2012)

AnalyticStream is proposed as a system to perform a

high-level analysis of the project situation and

identify the cognitive processes that can be

distributed through people to facilitate collaboration.

Based on the ideas of distributed cognition, a

process is designed to facilitate the recommendation

of relevant elements of analysis, as well as a

mechanism for the attention management that allows

users to have greater control over their flows of

shared activities. A study of the AnalyticStream

mixed methods shows that the suggestion of relevant

artifacts facilitates the discovery and consequently

the feasibility of reusing them, and provides

awareness of relevant aspects of the context on the

activities of other analysts.

More recently, in (Moran et al, 2011) they apply

the DCog (Distributed Cognition) approach to study

the interaction between agents (human beings or

artifacts) in a system, in the same way as the

collaboration that can be distributed during air traffic

control. The intention is to extract tacit knowledge

from observations of the cognitive system, which

can be derived from identification of trajectories and

the grouping of agent information actions/activities

into specific abstract processes. From this

perspective they analyze how information is

transferred through a system and identify specific

areas in which pervasive technology can be

introduced. However, they found difficulties related

to the contextualization, systematization and

excessive time that is taken in the analysis, and thus

realized the need for a tool to support this type of

analysis, guide the researcher, and reduce the

amount of time spent on analysis. They proposed a

prototype software, TITAN, designed to support

DCog analysis. This tool can be used to support the

identification and introduction of new pervasive

technologies to improve system information

processing.

Finally, the methodological framework DiCoT

(distributed cognition for teamwork) based on

distributed cognition is presented by (Vasiliou,

2015), to explain interactions and interdependencies

during collaborative learning activities in an ecology

of artifacts. They broaden the framework and

validate its applicability by understanding how

groups of students interact and collaborate, and

indicate aspects of redesigning to enhance a learning

environment supported by a physical and digital

artifact ecology.

Although CD theory has been frequently used to

study activities in various work scenarios,

(Edmondson y Russell, 2008) (Horsky, 2008)

(Mansour, 2009) (Nilsson et al, 2012) (Rajkomar

and Blandford, 2011) there are few applications

related to collaborative work and knowledge

management environments supported by computer

in the academic training.

3 PRELIMINARY RESULTS

This section explains the steps taken to apply the

Framework of Analysis of Interactions between

Agents in the analysis of the interactions that are

created during a process of knowledge management

(use, construction, distribution), based on the

perspective of distributed cognition.

A case of analysis is presented, where students

of a higher level solve problems in a group work and

collaborative way, mediated by ReSu, a tool oriented

to knowledge management. Different stages are

carried out in which the following results have been

achieved.

First stage: Representation of the ReSU system

from a cognitive approach

At this stage, ReSU is displayed as a cognitive

system in which structural agents (subjects, artifacts,

product, organization and environment) and

articulation agents (objectives and tasks) are

identified. By means of this identification procedure,

concepts of distributed cognition are implemented in

a structured way, and a conceptual and graphical

representation of the cognitive architecture of the

system is achieved.

Second Stage: Cognitive analysis unit’s

identification

Based on the agents that were identified in the

first stage, the aimed aspects to by studied, mainly

the interactions which occur in the process of

knowledge management are defined. Therefore, the

agents and the interactions between them (student-

pupil, student-teacher and pupil-artifact) are selected

as units of cognitive analysis, and is evaluated with

the aid of the use of the system

artifacts (forum,

Chat, wiki) if the interactions between agents

contribute to the construction, distribution and use of

knowledge.

The selection of agents as a cognitive unit is

consistent with the cognition distribution approach

centered on the subject, based on the observation of

the use of sign systems (semiotic mediation), the

evolution of the intersubjectivity processes

(construction of shared meanings), the role of

experience in human development, the coordination

between the internal structure (mind) and the

material structure (artifacts, environment), and

finally, the use of other human agents as cognitive

resources in the development of activities presented

as units of analysis (idea linked to the “proximal

development zone”).

Third stage: Data collection

To collect data on interactions, two

questionnaires one for the student and the other for

the teacher are designed. The purpose is to get

answers that help detect if the resources

implemented as knowledge management strategies

by ReSU are adequate to promote communication

and academic exchange between agents, i.e., if the

distribution of cognition on the system is favored.

On the other hand, we try find out the perception

they have about the effectiveness and efficiency with

which they have worked, and the ease and

satisfaction of using the artifacts. This information

will serve to identify the problems of usability of the

system. (Rearte et al, 2014)

In order to design the questions, constructs

(concepts, dimensions, factors or variables) are

taken into account from which it is expected to

obtain information. From the answers in the

questionnaires, we need to extract meaning units,

namely, fragments of text that represent indications

of some problem that occurred during the

interactions. Each meaning unit is categorized

according to the type of interaction involved

(student-pupil, student-teacher and student-artifact).

Fourth stage: Analysis of the units of

significance

Units of significance are illustrative and serve as

a guide to interpret problems or failures occurring

during interactions. Each unit of significance, which

represents some detected inconsistency or failure, is

represented by a causal diagram, representing the

probable issues and its causes. As an example, two

causal diagrams are presented, one obtained from

the answers given by the student and another from

the answer given by the teacher:

Issue

"Difficulty editing wiki pages"

Cause Because "the toolbar has few

editing options."

Cause Because "the wiki uses a

marked language (Wiki text),

for editing the pages."

To provide a solution to this problem, ReSU is

included in an "Improved tool bar" to edit wiki in a

simpler way. New features are added that facilitate

users to build pages and enrich them including

images, tables, lists, in a simple way, without the

need to use the proper marked language of the wiki.

Improving these aspects has as a result the

improvement of the usability of the tool, and

therefore it facilitates the student work.

Issue

"Lack of online evaluation

mechanisms in wiki

Cause: Because "It is not possible to

track the progress of the work

from the application."

Cause: Because "The tool does not

allow the teacher to give

feedback to the work group."

In this case is proposed to make changes in the

interface and to include some functionalities in the

system. In order to solve the detected problem, an

"Online monitoring" tool is included in the system,

which enables the teacher to follow up the process

while the groups are working. Using the tool, the

teacher visualizes the evolution of the work, and can

make recommendations or observations, which can

be taken into account by the group. This

improvement of the tool constitutes a scaffolding for

the activity of the group of students, allowing the

intervention of the teacher as mediator and

facilitator in the construction of knowledge. Under

the metaphor of scaffolds, the function of mediation

does not derive strictly from the technological stuff,

but from the support it offers to the teacher to

mediate the collaborative construction of knowledge

through scaffolds in space that is created as a

"proximal development zone".

Other modifications that have been made in the

tool, are related to the incorporation of a chat

included in the wiki, since many students considered

it uncomfortable to move out from the work

environment to communicate through the forum

with the other members of the group. A conflict

management mechanism is also implemented when

working simultaneously on a wiki.

At last, aspects of safety in the access to the area

of each group are considered.

From these changes, the verification of the

improved tool is proposed by replicating the

collaborative group work experiments, to check if

these new resources contribute favorably in the

cognition distribution of the system.

4 CONCLUSIONS

In this first part of the process, the framework of

analysis of interactions between agents has

facilitated the structured application of the main

concepts of distributed cognition in a system for the

formation of collaborative type and oriented to

knowledge management. Regarding the distributed

cognition approach, its usefulness has been proven

since it brings you towards a detailed level of

analysis, which can offer clues on how to change the

design of an technological artifact. As a pending

work, this should continue with the design of new

collaborative group work experiences to solve

problems, to check if it contributes to the cognitive

distribution in knowledge management cycles,

taking into account the new artifacts incorporated

into the system.

In carrying out this work, it has aroused different

concerns, which are considered as future works. We

would like first to deepen the determination, scope

and implications of the models of distributed

cognition in the collaborative and knowledge

management systems; emphasizing the aspects of

platform, proximal development zone, semiotics, as

aspects of interest for Distributed Cognition.

Secondly, to delineate a more systematized

prescriptive and proscriptive framework, to guide

the analysis, from the first stages (formulation of

guiding questions, selection of aspects and factors to

be analyzed) to the discovery of problems and their

causes, which will lead us to decisions for the

improvement and redesign of both cognitive and

physical artifacts. Thus to incorporate mechanisms

that facilitate the logical linkage of the data with the

criteria to interpret the results.

Finally, to generate automated tools that support

the analysis of the inherent aspects of distributed

cognition, in order to make the analysis more

accurate, to register the representational states that

are happening as the cognitive processes are

distributed. Ultimately, a way to reflect a cognitive

architecture and determine if it favors or

not the way

to distribute the cognition in environments supported

by technologies.

ACKNOWLEDGEMENTS

This work was partially supported by the Scientific

and Technological Research Council (CICyT)

Incorporated the Scientific Research and

Development Project to the Incentives Program of

the Secretariat for University Policies of the

Ministry of Education, Science and Technology of

the Republic of Argentina, Period 2016-2019.

REFERENCES

Edmondson W., Russell B. 2008. Projected Cognition -

extending Distributed Cognition for the study of

human interaction with computers. Interacting with

Computers, Elsevier Science Inc.Volume 20 Issue.

Feltrero Oreja, R. 2003 The Philosophy of Person-

Computer Interaction: Cognitive Dimensions and

Values. Disponible en: http://www.aipo.es/articulos/3/

03.pdf

Ferruzca Navarro, M. 2008 Estudio Teórico y evidencia

empírica de la aplicación Del Marco teórico de

Cognición Distribuida en la gestión de sistemas E-

Learning. Tesis doctoral. Universidad Politécnica de

Cataluña. España. Disponible en: http://www.tdx.cat/

handle/10803/6548.

Hussain N. y De Brujin O. 2010. An Analytical

Framework for the Evaluation of Collaborative Design

Around an Interactive Tabletop, Proceedings of the

11th International Conference of the NZ Chapter of

the ACM Special Interest Group on Human-Computer

Interaction.

Horsky J. 2008. Distributed Cognitive Resources: A

Framework for the Analysis of Human Interaction

with Complex Health Information Technology. VDM

Verlag Saarbrücken, Germany.

Lescano, G.; Pacheco Toledo, R. 2011. Practical

application of knowledge management approach: Web

system based on social applications as support for

collaborative work in higher academic areas. Degree

work. National University of Santiago del Estero.

Argentina.

Mansour O., 2009. Group Intelligence: A Distributed

Cognition Perspective. Proceedings of the 2009

International Conference on Intelligent Networking

and Collaborative Systems Pages 247-250 IEEE

Computer Society Washington, DC, USA. 20094

Moran S., Nakata K., Inoue S. 2011. A Distributed

Cognition Based Tool for Information Trajectory

Analysis. IEEE/WIC/ACM International Conferences

on Web Intelligence and Intelligent Agent Technology

Nilsson M., van Laere J., Susi T., Ziemke T., 2012.

Information fusion in practice: A distributed cognition

perspective on the active role of users. Journal

Information Fusion, Volume 13 Issue 1, 2012 Pages

60-78 Elsevier Science Publishers B. V. Amsterdam,

The Netherlands

Nobarany S., Haraty M., Fisher B., 2012. Facilitating the

Reuse Process in Distributed Collaboration: A

Distributed Cognition Approach, Proceedings of the

ACM 2012 conference on Computer Supported

Cooperative Work, New York, USA.

Rajkomar A., Blandford A., 2011. Distributed Cognition

for Evaluating Healthcare Technology. Proceeding

BCS-HCI '11 Proceedings of the 25th BCS Conference

on Human-Computer Interaction Pp 341-350, 2011.4

Rearte A. del C. y Castillo J., Sosa M. 2014. Distributed

Cognition applied to the empirical analysis of

Computer Supported Collaborative Knowledge

Management Interactions. XX Argentine Congress of

Computer Science. - 1st ed. - San Justo: National

University of La Matanza, Buenos Aires, Argentina.

Vasiliou C. 2015. Collaborative Learning In An Artifact

Ecology: A Distributed Cognition Perception.

Proceeding ITS '15 Proceedings of the 2015

International Conference on Interactive Tabletops &

Surfaces.