Mobile Sensemaking: Exploring Proximity and Mobile Applications in the Classroom

Abstract: We propose mobile sensemaking as a collaborative mechanism to explore and understand information in highly mobile and fluid situations, where people engage in multiple parallel, rapid and ad-hoc interactions, rather than participating in large highly-structured decision processes. Mobile sensemaking is explored in the classroom context, where it has been recognized that the traditional lectures should be reconstructed as active processes centered on collaborative activities. Mobile sensemaking relies on mobile computing devices and a proximity model, both organizing collaborative activities according to the domain context and physical proximity. The paper describes in detail the proposed proximity model and the developed mobile application.

Keywords: Mobile Computing, Computer Supported Learning, Collaborative Learning, Proximity.

Categories: H.4.0, H.5.3

1 Introduction

With the help of appropriate mobile technology and applications, teaching and learning procedures are expected to achieve higher levels of engagement, better adjustment to individual and group learning needs, higher learning rates, and better quality of time utilization and a better flexibility of teaching for the instructors.

Many educators agree that the main disadvantage of the traditional classroom lecture — the one placing teachers as the major focus of attention and most critical resources — is the reduced level of interactivity between teachers and students, and among the students themselves. The limited interaction possibilities in classroom lectures originates a set of problems regarding students’ attention and motivation, reduced teachers’ awareness of the actual learning accomplishments, and lack of flexibility for handling the necessary adjustments regarding the teacher’s and students’ goals.

From a pedagogic-psychological point of view, it has been considered that learning in the classroom should be reconstructed and redefined as an active process with more involvement of the student in meaningful learning activities [Ernest, 95], [Honebein, 96]. This reconstruction would include [Kafai, 96]: a) promoting students’ engagement in stimulating collaborative activities; b) increasing teachers’ awareness of students’ progress; and c) enriching the learning process with more sophisticated activities such as brainstorming, creative thinking, decision making, planning, and critical evaluation of the outcomes [Sass, 89].

Interacting with their peers by being engaged in collaborative learning activities also represents an opportunity for the learner to take hand in shaping the informational, communicational and learning process, rather than remaining a passive and individual recipient. As far as the success of interactivity in the classroom is concerned, empirical results indicate that: a) lectures are not generally ineffective, but are unsuitable to a global knowledge transfer [Gage, 96]; and b) the diverse learner-centered measures positively affect learning success [Hasan, 01].

Nevertheless, the classroom lecture remains as the most frequent teaching-learning scenario, since it has also important advantages compared to other settings. Especially important is the economic aspect regarding the teachers’ cognitive effort: only in a classroom lecture a teacher can economically deliver knowledge to a large number of students, regarding the resources involved and the time invested.

Our endeavor is to improve interactivity in the classroom while still keeping the learning process efficient in terms of resources and time. We have strong reasons to believe that mobile technology provides a technological platform capable to support the levels of interactivity required by the active learning process, and we are building software mechanisms to conserve the teachers’ effort in this process.

In this paper we show how wirelessly interconnected handheld computing devices may improve interactivity in the classroom involving university students in more sophisticated interactions than those expected in classic lectures, which in turn will foster collaborative learning. The focus of this technology is to improve sensemaking in the classroom, i.e. the students’ ability to collectively explore and understand information [Thilliez, 03] while shifting the teacher’s role to the backstage, performing supporting activities but not coordinating the assigned tasks.

2 Scenario

The task enfolds as follows. Each student receives one or two papers from the teacher and is encouraged to find out the main topics addressed by those papers. This individual task should then contribute to the collaborative effort. Students are expected to share their findings with others, identifying common topics, establishing relationships, and avoiding misjudgments. This should enfold in a paced and informal way, avoiding wasting time waiting for individual students to deliver their contributions, and in particular avoiding spending too much time discussing their divergences as a group. Instead, students are encouraged to engage in parallel negotiations with multiple parties to resolve their differences and reach consensus. Overall, the students assume the central role in the decision process, while the teacher is sent to backstage, coaching and encouraging students, assessing their accomplishments, although not coordinating the assigned task.

The fundamental aim of this task is to engage students in the sensemaking process. The sensemaking process was proposed by [Weick, 93] as a primary mechanism for organizations to explore and understand information. Sensemaking is an ongoing process aiming to create order and make retrospective sense about some event or collection of events. It has also been associated to preliminary decision-making activities like "understanding the situation" or "getting the picture" [Hasan, 01]. Sensemaking is also inherently collaborative [Larsoon, 03], meaning that the several mechanisms defined by sensemaking (ecological change, enactment, selection, retention) rely on the capabilities of a community of people to identify cues, update and share information, identify possible actions and provide feedback on those actions.

We argue sensemaking precisely captures the decision process defined by our working scenario. When students identify new main topics, they contribute to an ecological change. These new events may be sensed by other students, who enact their responses, looking for similarities, relationships, or even misjudgments. Then, collaboratively, they may try to make sense out of such events and construct a shared and coherent view. In summary, this scenario involves students exchanging information, moving around the classroom to engage in discussions with the other parties, negotiating common interests, and ultimately making sense of information.

3 Context and Proximity in the Proposed Scenario

Dix et al. [Dix, 00] describe four generic forms of context that influence interaction with mobile devices: infrastructure, system, domain and physical context. The infrastructure context addresses issues like the variability of services, user awareness of available services, or liveness of data. The system context is related to the management of feedback and feedthrough, support to distribution, and support to emergent behavior. The domain context considers the semantics of the application domain, e.g. the definition of the relationships between the mobile devices and their users, and how these relationships can be used to determine the application behavior. Finally, the physical context is related with the possibility that mobile devices are likely to be aware of their physical surroundings. For instance, the mobile devices may know that they are proximate to other devices (if some network connectivity is available) or in a specific classroom (e.g. if the classroom has a router installed).

Our approach explores two forms of context defined by [Dix, 00]: the domain and physical contexts. The domain context in our scenario is relatively complex because it combines individual and group work in a very fluid way. Students serendipitously move around the classroom forming temporary groups and holding ad-hoc interactions. The information about when groups were set up, who belonged to those groups and what interactions occurred characterizes the domain context in our scenario. This domain context should be maintained by technology to facilitate sensemaking, since it improves the retrospective understanding of the situation. The absence of domain context would represent an additional effort from the participants, who would have to search endlessly for hints about previous interactions with other students, the common topics that were found and decisions made.

We thus believe that the combination of proximity and context is a key aspect for supporting sensemaking in the classroom using handheld computing devices. We define two fundamental types of proximity contexts:
 

• Environmental proximity - The students perform their activities in the classroom. Environmental proximity contributes to define them as a group and to consolidate their expected behavior as group. Environmental proximity is thus associated to the production, sharing and sensing of topics in the classroom.
• Close proximity - The students engage together in very proximate face-to-face interactions, to avoid disturbing other students who may be engaged in their own interactions. Close proximity is associated to a face-to-face collaborative workspace, where two or more students share information and discuss about specific topics, their relevance and possible relationships.

Let us now discuss these matters in the physical context. According to [Thilliez, 03], proximity is relevant when users are close to each other and, according to their location handheld computing devices may support a differentiated set of services. Physical proximity is based on the communications networks established by handheld computing devices, which are formed dynamically by juxtaposition of wireless networks created on demand. Physical proximity defines a context identifying who was physically close to each other and what information was exchanged between them. Based on these notions, we may complete our definitions of environmental and close proximity:
 

• Environmental proximity - The students perform their activities in a confined physical space, the classroom, allowing establishing a communications network between all students’ handheld devices. This allows sensing topics in the classroom.
• Close proximity — When the students engage together in very proximate face-to-face interactions, their handheld devices will establish a communications network. This network is distinct from the one associated to environmental proximity, and allows sharing a workspace between proximate students.

4 Proximity Model for Mobile Sensemaking

• The CPC is automatically activated when two or more handheld devices are connected together at the very proximate physical level (e.g. using IRDA).
• The CPC will be active as long as there is physical connection between at least two devices.
• The students engaged in the same CPC automatically share their workspace and the information belonging to the shared workspace is also part of the CPC.
• The CPC evolves according to the participants and shared information. This allow for several students to get anytime in and out of the discussion.
• The CPC is automatically deactivated when physical connectivity is lost.
• The deactivated CPC will remain in the handheld devices for search and navigational purposes. This functionality supports retrospective sensemaking.

Focusing on the whole classroom, we also define an Environmental Proximity Context (EPC):
 

• The EPC is automatically activated when several handheld devices are interconnected at the physical level (e.g. using WiFi).
• The students engaged in the EPC automatically receive indications about the topics generated by other students that may be of interest to them. This functionality uses similarity text matching.
• The students might interact with their devices to request becoming proximate to students for which some similarity has been indicated to them.

Environmental proximity situation. The students are identified with letters from A to Z, while T represents the teacher. They all share the same classroom, and their handheld devices share the same (Wi-Fi) network. Therefore they are potentially engaged in the same EPC. However, not all students are effectively engaged in the EPC at a specific time, because they may be engaged in a close proximity situation (these are the cases of, e.g., RL, GO and KD). Within the EPC, when a student produces a topic, it is distributed to the other students’ handheld devices. The devices compare their current list of topics with the distributed topic and, if there is some similarity matching, the student will be notified. Note that unrelated topics are filtered out, but they may become related later on, when students change their list of topics. If a student wishes to discuss with the student that produced the topic, she will invoke an engagement protocol, which is described next.

• Case EPS1. Using the WiFi network, a student X (see figure 1.) searches for the topics other students have defined for the articles they have read. Once the student has found the topics she is interested in decides it is not necessary to try a face-to-face meeting. Nevertheless, the following situations may occur: a) student V has more detailed information about topics X is interested which V is willing to share, b) X can send information to V about the topic both are interested in, c) X and V are only interested in exchanging information about the topic but not on discussing about why they use them to characterize the article, and finally d) there is no interest in sharing information.
• Case EPS2. A student Q finds using the WiFi network that Z is willing to share information with him. In this case, it is necessary to activate the engaging process (request, accept, software defines hot-spot) in order to enable Q and Z find each other in the Hot-Spot Y. After that a CP1 situation may arise.

Close proximity situation. The students in this situation are face-to-face and share a CPC. Their handheld devices automatically establish a temporal ad-hoc network connection (IRDA). Furthermore, their devices will provide a shared workspace, where topics may be collaboratively edited and linked with other topics present in any one of the participants’ handheld devices. This allows effectively exchanging and sharing topics and links across multiple devices in an epidemic way, whenever students engage in new close proximity situations. We shall consider several possible scenarios within the close proximity situation:
 

• Case CPS1. Students R and L engage in a social face-to-face interaction in order to share, discuss, understand and relate the topics the application context has found they share. After this interaction following situation may occur: a) that R and L could define adequate relationships between some topics specified by each one for the articles they have read, b) that they found no relations between the common topics.

• Case CPS2. Students K and D are informed by their handhelds that they have topics in common for the articles they have read. However, they decide that a social interaction is not necessary and that is enough that: a) K sends information to D, b) D sends information to K or c) both exchange their information. This information is sent over the WiFi network and may correspond to the detail generated by each student for a certain topic and Hill be available only if there is mutual consent.
• Case CPS3. Students C and J are informed by their handhelds that they have not common topics on their lists so there is no need to engage in a social interaction. However, it is still possible that: a) The information J has is somehow relevant to C, who is willing to share it, b) the information C has is relevant for J, who is willing to share it, c) the information J and C have is relevant for each other and both are willing to share it, and d) the information J and C have is not relevant for the other so there is no transfer of information between them.

Disengagement protocol. The disengagement protocol occurs when one student considers that the face-to-face interaction is completed, and perhaps other students could be contacted. The disengagement occurs when the student moves away from the face-to-face interaction and the (IRDA) network connection is lost. Then, the student is again in the environmental proximity situation. As mentioned, the contextual information associated to the face-to-face interaction is preserved in the CPC.

5 Implementation of the Mobile Sensemaking Application

Also, as described in [Zurita, 06], the framework is able to recognize when to users engage in a face-to-face encounter, aligning their handheld devices. In this section we describe how these features were used to build the mobile sensemaking application. The application offers several User Interfaces (UIs) allowing the teacher to assess the classroom activity, and giving the students the ability to write topics associated with their assigned papers, link these topics with other topics, and engaging in collaborations with other students. Most interaction with these UIs is done with pen gestures, because it is the natural way for a user to control a handheld device.

Figure 2: The teacher UI displays the list of students and available papers. On the left side the list of students is displayed (which were found by the participant discovery mechanism of the application based on multicasting messages). On the right side, the list of the papers (identified by the author’s name and publication year) is displayed. A paper is assigned to a certain student by drawing a line with the stylus from a paper’s bullet to a student’s bullet (or vice-versa)

5.1 Papers distribution

To assign a paper to a student, the teacher must drag its document-icon over the student-icon. Dragging a student over a document-icon would also assign a paper to a student. These actions may be repeated several times, assigning multiple papers to a student and multiple students to a paper. Every time this is done, both icons will show an updated count of links over their icons: the document-icon will show how many students have been assigned to work with that paper, and the student-icon will show how many papers have been assigned to him/her (figure 2). The teacher can also randomly assign one paper to each student by clicking on the dice-icon, at the upper bound of the UI. Clicking this icon repeatedly assigns multiple papers to each student, ensuring every paper has a similar number of students assigned.

5.2 Paper reviewing and topics linking

The teacher’s UI for topics definition is normally empty. However, it allows the teacher to type or draw generic topics that may help students recognize what kind of sentences are meant to be considered as topics. When the teacher creates such topics, they appear in the students’ devices along with his/her own written topics. The topics created by the teacher are displayed using different colors and borders than those created by the students. Figure 3 shows topics created by the teacher and the student.

Figure 3: Topics definition and linking UI. Using the stylus, students can link papers with related topics. Icons show the user current links’ configuration, which may be public, private or available only during face-to-face encounters. Topics created by the teacher are displayed with a different border

5.3 Sharing privileges and information sharing

The system allows students to choose in which way they want to share generated knowledge. In this case, each link may be configured as "public", "face-to-face only" or "private". When a connection between a paper and a topic is configured as public, all students in the activity may access such information through the "Topic search screen" or "face-to-face discussion", both described next. If it is configured as face-to-face only, such information will be revealed when two students engage into a face-to-face discussion, allowing the unconfident student to talk about the idea with another participant. When a topic link is configured as private it won’t be available to other students under any interaction mode until the student changes its configuration. Students may configure a link access by double clicking it on the screen using the handheld stylus. When this occurs, a small floating palette will offer the three available states that the user can click. Each link between papers and topics displays a small icon representing its sharing configuration, as shown in figure 2. Links are created with "face-to-face only" privileges by default.

5.4 Related topic search and environmental sharing

Smart text matching algorithms simplify the search process by organizing the topics map according to the student’s interests. Topics similar to the current student’s ones are displayed closer to the center, drawn in darker color if their similarity reaches a high level. The participant distribution in the screen depends on overall topics likeness: other students may be located near the center when they have a high number of coincidences between his/her topics and current student’s ones.

Figure 4: The topics map UI. Other students’ similar topics are displayed in bold and darker color. Double clicking another student’s icon displays the interaction UI.

Originally, the screen is zoomed in order to display the closest participants only. The user can drag the screen to navigate through the entire list of participant holding and dragging the stylus. Also, the user may zoom in or out clicking the magnifier icons or dragging the zoom slider at the right of the screen. Finally, the user can double click another student’s icon when he/she is interested in this particular student’s topics or wants to invite him/her to a face-to-face encounter. Based on these simple pen-based gestures each student may browse all public topics.

5.5 Interacting with other students

 The bottom and the top of the interaction screen display both students’ topics collection. Topics from each student are horizontally sorted in order to be vertically aligned with topics from the other student based on text similitude. If their texts match exactly or above a certain peak, an automatic link between them is displayed. A student may manually link his/her topics with the other students’. To create a link between two topics he/she has to draw a line connecting their labels, in the same way as he/she linked the topics with the papers in the topics definition screen. Topic to topic links show an arrowhead according to which student created it. In case both students agree on such relation, having the two of them drawn the same link, the line will have arrowheads in both ends and get highlighted. Automatically created links always display as a two-way link. Finally, students may link their papers directly to the other users’ topics. Topic-paper’s links are created using the same link gesture available in the "topics definition" screen. By doing this, topic label will be relocated in the center of the screen, showing its links to papers of both students.

5.6 Engagement invitation

6 Discussion & Conclusions

The importance and potential of context in general and awareness in particular was discovered very early in the short history of the development of collaborative mobile applications. In [Kaasinen, 03] the author presents a works showing how context information can be used in different application areas, e.g. tourist guidance, exhibition guidance, e-mail, shopping, mobile network administration, medical care and office visitor information. In these studies, the location of the user is the main attribute used in the context-adaptation. In [Bardram, 04] the authors show the value of context information and social awareness for developing an application to support collaboration between experienced and novel doctors in a hospital. In [Tähti, 04] a mobile application which offers various services supporting office-type work which uses context-awareness, mainly information on position of the user and available services nearby. It seems there are no major contributions in the field of context-aware applications for supporting collaborative learning except for those dealing with participatory simulations, like the one described in [Klopfer, 05].

Acknowledgements

This paper was funded by Fondecyt 1050601.

References

[Bardram, 04] Bardram, J., Hansen T.: The AWARE Architecture: Supporting Context Mediated Social Awareness in Mobile Cooperation. Proceedings of the CSCW’04, November 6-10, Chicago, Illinois, USA (2004), 192-201

[Dey, 01] Dey A.: Understanding and Using Context. Personal and Ubiquitous Computing, 5, 2001. 4-7

[Dix, 00] Dix, A., Rodden, T., Davies, N., Trevor, J., Friday, A., Palfreyman, K.: Exploiting Space and Location as a Design Framework for Interactive Mobile Systems ACM Transactions on CHI (2000), 7.

[Ernest, 95] Ernest, P.: In Constructivism in education (Ed, Gale, L.) Hillsdale, NJ: Erlbaum, 1995. 459-486.

[Gage, 96] Gage N., Berliner D.: Educational Psychology., Houghton, Mifflin, Boston, MA, (1996)

[Guerrero, 06] Guerrero, L., Ochoa, S., Pino, J., Collazos C.: Selecting Computing Devices to support Mobile Collaboration. Group Decision and Negotiation 15(3), 2006. 243-271

[Hasan, 01] Hasan, H., Gould E.: Support for the Sense-Making Activity of Managers. Decision Support Systems, 31, 2001. 71-86

[Honebein, 96] Honebein P.: Seven goals for the design of Constructivist learning environments. In Constructivist learning environments New Jersey: Educational Technology Publications, 1996. 17-24

[Jippling, 01] Jippling, M., Dieter, S., Krikker, J., Sandro, S.: Using Handheld Computers in the Classroom: Laboratories and Collaboration with Handheld Machines. Proceedings of the 2001 SIGCSE, SIGCSE Technical Bulletin, Vol. 33, No. 1 (2001), 169-173.

[Kaasinen, 03] Kaasinen, E.: User Needs for Location-Aware Mobile Devices. Personal Ubiquitous Computing 7, 2003. 70-79

[Klopfer, 05] Klopfer, E., Yoon, S., Perry J.: Using Palm Technology in Participatory Simulations of Complex Systems: A New Take on Ubiquitous and Accessible Mobile, Computing Journal of Science Education and Technology, Vol. 14, No. 3, September (2005), 285-297

[Larsoon, 03] Larsson, A.:. Making sense of collaboration: the challenge of thinking together in global design teams. Proceedings of ACM SIGGROUP Sanibel Island, Florida (2003), 153-160

[Liu, 03] Liu, T., Wang, H., Liang, J., Chan, T., Ko, H., Yang, J.: Wireless and mobile technologies to enhance teaching and learning Journal of Computer Assisted Learning 19, 2003. 371-382

[Sass, 89] Sass, E.: Motivation in the College Classroom: What Students Tell Us. Teaching of Psychology, 16, 1989. 86-88

[Tähti, 04] Tähti, M., Rautio, V., Arhippainen, L.: Utilizing Context Awareness in Office Type Working Life. Proceedings of the MUM 2004 October 2729, 2004 College Park, Maryland USA (2004), 79-84

[Thilliez, 03] Thilliez, M., Delot, T., Lecomte, S., Bennani N.: Hybrid Peer-To-Peer Model in Proximity Applications. In Proceedings of the 17th International Conference on Advanced Information Networking and Applications (2003), 306

[Weick, 93] Weick, K: The collapse of sense-making in organizations. The Mann-Gulch disaster Administrative Science Quarterly, 38, 1993. 628-652

[Zurita, 04] Zurita, G., Nussbaum, M.: A constructivist mobile learning environment supported by a wireless handheld network. Journal of Computer Assisted Learning 20, 2004. 235-243

[Zurita, 05] Zurita, G., Baloian, N., Baytelman, F.: Handheld-Based Electronic Meeting Support. Lecture Notes in Computer Science, Vol. 3706/2005. Proceedings of the 11th International Workshop, CRIWG 2005, Porto de Galinhas, Brazil, September 25 - 29 (2005), 341-350

[Zurita, 06] Zurita, G., Baloian, N., Baytelman, F.: A Face-to-Face System for Supporting Mobile Collaborative Design using Sketches and Pen-based Gestures. In Proceedings of the 10th International Conference on Computer supported Collaborative Work in Design, May 2006, Nanjing, China. IEEE press (2006), 250-255
 

[Zurita, 07] Zurita, G., Baloian, N., Baytelman, F., Farías, A.: A Framework for motivating Participatory Simulations. Proceedings of the 11th International Conference on Computer supported Collaborative Work in Design IEEE press, Melbourne, Australia, May (2007), 1016-1021