Papers: Visitor Interactions with Digitized Artifacts

Introduction

The Kinematic Models for Design Digital Library (http://kmoddl.library.cornell.edu/) is an on-line collection of mechanical models and related resources for learning about and teaching the principles of kinematics and the history and theory of machines. KMODDL features two separate collections in two locations - the Reuleaux Collection of Mechanisms and Machines housed at Cornell University, and the Clark Collection of Mechanical Movements housed at the Museum of Science, Boston.

The library contains digital videos and simulations of the models, and offers the capability for the on-line collection to be printed into 3D tactile models using CAD simulations of the original artifacts. The KMODDL collection affords educators with access to three different versions of mechanical models: the original historical artifacts (which are on-site at Cornell University and the Museum of Science, Boston), computer simulations, and 3D tactile printouts. The variety provides educators with the opportunity to choose the format for presenting and using the models that best suit their needs.

Using the resources of KMODDL, a team of researchers and educators from Cornell University and the Museum of Science, Boston, conducted a research study to explore how the digital and physical experience can complement each other in learning about mechanical motion, and how the learning experience differs in two contexts: middle school classrooms and science museums. The learning experiences developed by the educators included the three different types of mechanical models: historical artifacts, 3D replicas, and computer simulations. Educators at the Museum of Science introduced the activities during August 2005, and educators at Cornell University introduced their activities to a local middle school classroom beginning in November 2005. The project will ultimately compare learning between the classroom and science museum contexts, but the current research study examines only the programs at the Museum of Science.

Research suggests that middle school students value models that are as close as possible to the real thing (Grosslight, Unger, & Jay, 1991; Harrison & Treagust, 1998; Treagust, Chittleborough, & Mamiala, 2002). This contrasts with the view of scientists who believe that in some cases (particularly when working with objects that are distant, such as planets, or processes that take considerably long periods of time) models may be preferred as they provide insights into dimensions or facets of the process that the original cannot. This evidence suggests that the visitors may prefer the original artifacts because they may believe they are closer to the “real thing.”

However, studies conducted with undergraduate students who utilized KMODDL tools in the classroom found that the preferred learning tool for the students (comparing the original artifact, computer simulation or 3D tactile printout) was not absolute and seemed to vary depending upon the context of use (Pan, Gay, Saylor, Hembrooke, & Henderson, 2004). This finding follows the framework established by activity theory, which moves the focus of the interaction from the object alone to the intersection of the learning group, the mediating artifacts, and the object or goal of the interaction (Engestrom, Miettinen, & Punamaki, 1999). It is through the intersection of these factors that changes in knowledge and attitude can occur.

Activity theory suggests, therefore, that the type of learning tool that visitors prefer (original artifact, computer simulation and tactile model), and what visitors learn through those tools may vary greatly depending on how the tools are used and who is included within the learning group. For this reason, this study examined visitor preference for the different types of learning tools across two programs that present very different contexts for use.

Educational Program Description

At the Museum of Science, educators in the School Programs and Interpretation departments incorporated the three versions of the mechanical models into the learning experience of two existing programs: Design Challenges and Exhibit Interpretation. The Design Challenge program (http://www.mos.org/doc/1417) introduces students and visitors to the engineering design cycle through hands-on activities that ask visitors to design, build, and test a prototype solution to a given problem. This program was aimed at visitors in grades four through ten. Educators in the Design Challenge Program invited visitors to use the original historical artifacts, 3D printouts, digital representations, and prototypes that other visitors had built to explore the engineering design process and inspire their design for a hand-stamping machine. Visitors synthesized gears, leavers, and pulleys into their own unique system to accomplish the task. The program was used daily throughout the summer months, and has been integrated into the rotation of Design Challenge programs that are presented in the museum (http://www.mos.org/doc/1853).

Exhibit Interpretations are informal science activities led by trained Museum of Science staff and volunteers at exhibit components and carts located throughout the institution. Interpreters provide hands-on, inquiry-based learning experiences with the goals of encouraging observation, questioning, and critical thinking, and helping to make science and technology interesting, fun, and relevant to visitors' lives. Interpreters provide activities aimed at a wide audience, including family groups and middle and high school students. This interpretation focused on helping visitors to recognize the machine elements that are common in everyday objects, such as clocks, eggbeaters, CD players, and music boxes. Visitors interacted with the historical artifacts, digital representations, 3D-printed replicas of the Clark models, toy gear models, and the real-life objects they were examining.

Research Goals

Research explored the following questions:

1. What effect does the integration of digital and physical experiences (provided by the physical artifacts, 3D replicas, and computer simulations) have on learning?
2. Which of the three models do the visitors prefer? Does this vary depending upon the context and type of activity?
3. Are the programs that utilize these models successful at achieving the educational outcomes as defined by the project teams?

Methods

Sampling Strategy

Only un-cued visitors who chose to participate in the programs were included in this study. In total, 79 visitors were observed and interviewed between the two programs: 35 for the Design Challenge program and 44 for the Exhibit Interpretation program. One visitor per visiting group was selected for observation. For observation and interview, observers selected the first child between the ages of 8 and 18 from each group to begin an interaction with the activity.

The target age range was 8 to 18, but the actual age range of individuals observed and interviewed was 8 to 17, with a mean age of 10.5. Of the 79 total visitors observed and interviewed between the two activities, 45 participated in the activity with their family, 10 participated with a field trip group, 10 participated in a group that included only other children under the age of 18, one participated with a home school group, and three visitors did not identify their visiting group. Thirty-eight of the visitors were female and 41 were male.

Data Collection and Analysis

Data collection methods employed included tracking and timing, which captured visitor attraction and engagement with the different learning tools, and post-activity interviews, which focused on the visitors’ perspective of the relative value of the different models. For tracking and timing data, one visitor per visiting group was observed as s/he participated in the activity. The amount of time the visitor spent interacting with each type of model and the number of times s/he chose to interact with each was recorded, as well as the length and number of times the visitor interacted with other parts of the activity. These observations also recorded who initiated the interaction with the model: the educator, the visitor being observed, or someone else in the visiting group.

Following participation in the program each visitor was interviewed. Whenever possible, these visitors were the subjects of the tracking and timing study, and individuals were interviewed at a distance from the educators running the program and their own group members to avoid biasing their responses. Most interview questions were similar across the two programs so that the interview responses could be compared. Other questions were tailored to specifically measure the educational goals of the programs.

Results

Tracking And Timing Of Design Challenges

In the Design Challenge program, 54.3% (19 of 35) of visitors interacted with some form of the Clark Models (the historical artifacts, 3D replicas, or computer simulations). The 3D replicas were the most frequently used form of the Clark Models, with 28.6% (10 of 35) of visitors using them, closely followed by the historical artifacts, which were used by 20% (7 of 35) of visitors. Just 5.7% (2 of 35) of visitors interacted with the computer simulations of the Clark Models. However, 68.6% (24 of 35) of visitors interacted with the prototypes that other visitors had developed, more that double the number that interacted with any one form of the Clark Models.

Figure 1: Frequency of Model Type Use - Design Challenge

Across all visitor interactions, the historical artifacts had the greatest duration of interaction of the Clark Models, with an average of 14 seconds. Although more people interacted with the 3D models, the average duration of interaction was much shorter (only four seconds). The average interaction duration for the computer simulations was two seconds. The average time interacting with all of the three forms of the Clark Models was 21 seconds. In contrast, the prototypes created by other visitors were both more widely used and had average interaction duration of 53 seconds. The bulk of the time at the Design Challenge activity was spent on creating and testing a prototype, with an average of 23:52 of the overall 26:18 seconds spent engaged in this part of the activity.

Overall, 90.7% of visitors’ time was spent on creating and testing their prototype, 3.4% of time was spent interacting with other visitors’ prototypes, and only 1.3% of their time was spent interacting with some form of the Clark Models.

Tracking and timing of Exhibit Interpretation

In the Exhibit Interpretation program, 77.3% (34 of 44) of visitors interacted with some form of the Clark Models (the historical artifacts, 3D replicas, or computer simulations). The historical artifacts were the most frequently used form of the Clark Models, with 61.4% (27 of 44) of visitors using them, followed by the 3D replicas, which were used by 50% (22 of 44) of visitors. Just 11.4% (5 of 44) of visitors interacted with the computer simulations of the Clark Models. Visitors used the toy gear models and the everyday objects containing gears more frequently than any of the Clarke Model variations, with 95.5% (42 of 44) of visitors using each.

Figure 2: Frequency of Model Type Use - Exhibit Interpretation

As with the Design Challenge program, the historical artifacts had the longest duration of interaction of the three types of Clark Models, with an average of 2 minutes, 12 seconds. With the Exhibit Interpretation program, the duration of interaction followed the frequency of use, with average an interaction of 1:00 with the 3D models and 17 seconds for the computer simulations. The average time interacting with all of the three forms of the Clark Models was 3:29. So, although the average interaction for the Exhibit Interpretation was shorter (10:00, compared to the average interaction of 26:18 for Design Challenges), the amount of time spent with all three forms of the Clark Models was greater.

Overall, visitors spent 40.4% of their time interacting with the real-world objects, 23.8% of their time interacting with the toy gear models, 21.9% with the historical artifacts, 9.9% with the 3D replicas, and 2.9% with the computer simulations.

Figure 3: Interaction Duration by Model Type

Visitor Vs. Staff Initiated Interactions

The visitor experience with the activities also varied with regard to who initiated the interactions with the various learning tools. In the Design Challenges activity, interactions with all model types were initiated by the visiting group 86% of the time, and by the Museum educators only 14% of the time. Most often, the visiting groups (and not the educators) initiated the interactions with each of the Clarke model variations. However, interactions with the different model types in the Exhibit Interpretation were as likely to be initiated by the educator as the visiting group. Visiting groups most frequently initiated the interaction with the computer simulations (66.7% of the time), followed by the toy gear models (60.2%), and 3D model replicas (50%), and less frequently with the real-world objects (43.1%), and the historical artifacts (42.5% of the time).

Visitor Preference Of Model Type

Visitors at both the Design Challenge and Exhibit Interpretation programs were asked which of the model types they learned the most from. Participants in the Design Challenge program overwhelmingly named the prototypes made by other visitors as the type that they learned the most from (27 of 35 visitors). When asked how these models help them learn, they were most likely to say that they could see how the object worked. For example, one nine-year-old female said, “I saw how the stamps worked, the motor and the gears, so it gave me ideas of what to do.” Another 12-year-old male said, “It gave me examples. I could take some parts of them to add it together to make my own machine.” Other responses given were that the prototypes were just like what the visitors had to do themselves, such as a 14-year-old female who said, “[It] showed what to do, very clearly,” or that they could touch them. In addition, four visitors said that they did not know or did not give a reason how it helped them learn.

The only other model type that was named by visitors in the Design Challenge activity as the one they learned the most from was the historical artifacts (by 5 of 35 visitors). These visitors stated that the models’ ability to show them how the object worked helped them to learn. The remainder of visitors could not or chose not to name a model type that they learned the most from.

Participants in the Exhibit Interpretation program were equally likely to name the toy gear models and the real-world objects as the learning tool that they learned the most from (both were named by 16 of 44 visitors). When asked how the toy gear models helped them to learn, visitors were most likely to say that they could see how the object worked. One eight-year-old female said, “Well it shows you how all the things work, and how it usually, how everything connects together.” Visitors also mentioned that they could manipulate the toy gear models themselves, such as the 10-year-old male who said, “You could interact and change things and move them yourself.” Visitors also said that they were easy to use and understand, and that it was like something they had used before.

Visitors who named the real-world objects as the object type they learned the most from were again likely to say that they could see how it worked. One 10-year-old female named the eggbeater and said, “It helped me to learn how the gears work to move it.” Visitors also mentioned that they hadn’t done or seen something like that before, that it was like something they used or had, or it was fun to use.

The historical artifacts (4 visitors) and 3D models (3 visitors) were much less likely to be chosen by visitors at the Exhibit Interpretation activity as the object type that they learned the most from, and no visitors chose the digital simulations. Visitors who chose the historical artifacts were likely to say that they helped them to learn because they could control the movement of the models or they could easily see how the models moved. They also mentioned that they were interesting or complex. Visitors who chose the 3D models also said that they could easily see how the model worked or chose them because they could manipulate the model themselves.

Figure 4: Preferred Model Type - Design Challenge and Interpretation

Visitors at both programs were also asked which model type they learned the least from. The visitors at the Design Challenges named the 3D model copies most frequently (18 of 37 visitors). When asked what made them difficult to learn from, visitors most often said that they were not related to the activity that they were doing, such as an 11-year-old male who said, “It doesn't really move like how the [hand-stamping] machine has to move.” Three visitors also said that they didn’t understand how the 3D model copies worked, and three did not know why they were difficult to learn from. Two visitors said that they did not see them, and visitors also mentioned that they were not interesting, they couldn’t manipulate them, or they were broken.

Eight Design Challenge visitors named the historical artifacts as the most difficult to learn from. Four of these visitors did not see the models. Two said that they were not like the activity they were doing. Others stated that the artifacts were too difficult to understand, too hard to use to see how it worked, or were broken.

Visitors at the Design Challenge also named the digital simulations as difficult to learn from (6 of 37). They were most likely to say that they had not seen them (3 visitors) or that they were not related to the activity they were doing (2 visitors). Visitors also said they were too difficult to see.

Visitors at the Exhibit Interpretation were also asked which type of object was the most difficult to learn from, and most often named the real-world objects (12 of 44). However, six of these 12 visitors also named the real-world objects as the easiest to learn from. In each of these cases, the visitor named a different specific object when asked what type of model was the easiest or most difficult to learn from. When asked what made these objects difficult to learn from, visitors were most likely to say that it was too difficult or complicated (“Some of the real-life ones, they were too complicated for me to understand, like the CD player.”), or they could not see how it worked (“They're tinier and it's harder to see into with the little gears.”).

The next most frequently named model type that was named by visitors as difficult to learn from at the Exhibit Interpretation was the digital simulations or the toy gear models (both 9 of 44 visitors). Visitors who found the digital simulations difficult to learn from most frequently said that this was because they could not manipulate the digital simulations, such as the 13-year-old male that said, “You can't do anything with it. It's just the same thing over and over.” Several visitors also did not see the digital simulations, did not find them interesting, or were not able to see how they worked.

Those who found the toy gear models difficult to learn from cited that they felt they had seen things like them before; did not find them interesting; thought they were “for kids”; found them too difficult; couldn’t manipulate them; or didn’t understand how they worked.

Very few visitors at the Exhibit Interpretation named the historical or 3D models as most difficult to learn from (3 and 2, respectively). The visitors who found the historical artifacts difficult to learn from said that they were too complicated, they couldn’t touch them, or didn’t understand how they worked. The visitors that found the 3D models difficult to learn from thought they were not interesting.

Figure 5: Difficult to Learn from Models - Design Challenge and Exhibit Interpretation

Overall, visitors at both programs said with overwhelming frequency that they learned the most from objects where they could clearly see how the mechanism worked. Visitors at the Design Challenge program also mentioned with some frequency objects that were like the task they needed to do, while this was not mentioned at all at the Exhibit Interpretation because that activity did not require visitors to use gears and mechanisms to complete a task. Although both groups of visitors said that the objects they could manipulate made learning easier, this was mentioned with greater frequency by visitors at the Exhibit Interpretation than those at Design Challenges (8 versus 2 times, respectively). Visitors at Exhibit Interpretation also mentioned ease of use, familiarity of the object, and how interesting the object was, while visitors at the Design Challenge activity did not mention these factors.

Visitor Learning

During the interview, visitors at both activities were asked to name something new that they learned. Visitors at the Design Challenge activity most frequently said that they learned the process of building - how to build with gears to accomplish a task, how to build a hand stamper, how to connect gears, the role of experimentation and improvement of a design, or how to use a specific gear or mechanism to accomplish their goal. Several visitors had affective responses, such as “Building with gears is fun!” or “I can build with gears.” Just one visitor mentioned learning about how gears function.

In contrast, none of the visitors at the Exhibit Interpretation said that they learned about the process of building or had affective responses. Many visitors said that they learned things about how gears function (such as that gear size affects its speed), about gears in general or a specific gear or mechanism, and about the wide range of gear types. Many also said that they learned about the gears in everyday objects. Visitors who said they learned the most from the historical artifacts or 3D models were more likely to say that they learned how a specific gear or mechanism works than visitors who learned the most from the toy gears or real-world objects.

Discussion

This study examined the effect that integration of digital and physical experiences had on learning. Although both activities had similar digital and physical artifacts, they were used with different frequency and resulted in different types of learning. According to activity theory, changes in the subject (visiting group), object (or goal), or mediating artifacts (including the educator, physical location, and original or digital artifacts) result in different outcomes (Engestrom, Miettinen & Punamaki, 1999). The different goals of the Design Challenge and Exhibit Interpretation activities led to different usage of the models and different learning by the visitor.

The Design Challenge program had a clearly defined goal: design and build a working hand-stamping machine. This goal resulted in tightly focused interactions with the models, with visitors overwhelmingly saying that they learned more from model they perceived as being most relevant to the problem they were trying to solve (the prototypes built by other visitors). This was the model type that the highest percentage of Design Challenge participants interacted with and that had the longest interaction duration. Visitor learning was also tightly focused, and mainly related to the building process they engaged in. In addition, visitors were very self-directed in their learning, initiating 86% of the interactions with all model types.

In contrast, the Exhibit Interpretation activity had a more general focus, with the educational goal of visitors recognizing components of the historical artifacts and other mechanisms in everyday objects. This broad goal led to a wider variety of visitor experiences, including a more varied use of models, broader range in visitor versus educator initiation of interactions, a wider spread in the type of learning tools visitors learned the most from, and more varied visitor learning.

In both activities, the digital simulations were not named as the type of model that visitors learned the most from; they also had the shortest duration of visitor interaction. Visitor who saw the models indicated that this was because they were not applicable to the task they were trying to complete in the case of the Design Challenge, or that the visitor could not manipulate them. Team members at Cornell University are currently developing a program that will allow a user to interact with and change digital simulations of gears and mechanisms. This may address the issue of manipulation and visitor control. In the case of Design Challenges, visitors do not show the inclination to interact with any model types that do not directly relate to their task, so digital (or other) models must be carefully selected both to apply to the goal of the interaction and to be relevant to the other mediating artifacts that are a part of the activity. The use of digital simulations also poses logistical issues that can serve as barriers to mobile programs that are not permanent exhibition components (setup with extension cords, availability of power sources on museum floors, etc.). These issues have served as barriers to using digital simulations in the long-term adoption of mobile programs. Technical changes must be made to the delivery mechanism to address these needs.

Although the type of learning tool visitors had a preference for and the way that the visitors engaged with these learning tools varied across the two programs, we do see similarities in the types of attributes discussed when describing their preferred model. Across both the Exhibit Interpretation and the Design Challenge program, visitors stated that the learning tool they learned the most from was the one where they could see what was happening and how the gears worked. In Design Challenges, the visitors made reference to the fact that the other visitors’ prototypes helped them to best see how the their eventual designs might work, while at Exhibit Interpretation visitors stated that the real-world artifacts and toys allowed them to see how gears function. This preference for models that allow seeing how things work is not surprising; this sentiment in echoed in other research that explores middle school students’ conceptions of models (Treagust et al., 2002).

Another similarity in the visitors’ preferences for the different learning tools is that they preferred the learning tool that best met the goal and direction of the program. When describing the tool that they felt helped them to learn the most, visitors participating in the Design Challenge activity referred to the utility of the model in helping them to design their own hand-stamping machine. For this reason, an overwhelming percentage of the visitors preferred the other visitors’ prototypes as these prototypes best helped them to brainstorm potential designs., rather than any version of the Clarke models whether they were digitally or otherwise represented. In contrast, visitors participating in the Exhibit Interpretation program described in their interview how their preferred learning tool helped them to see how the gears in everyday objects work and function. For this reason, these visitors tended to prefer the real-world objects, the toy gears, and some forms of the Clark models when the connection to everyday objects was clearly made. This suggests that museum designers and educators should carefully consider the utility and function of the object representations when determining which learning tools to incorporate into a program as this may be more important than whether or not the object is “real” or a digital reproduction.

The same physical and digital objects when presented in activities with different goals can produce widely varied visitor experiences and learning. Regardless of an activity’s goal, models are used by visitors and identified as useful learning tools based on their applicability to the activity’s stated goals. Successful use of digital artifacts in museum programs requires taking this into account when designers decide whether or not digital reproductions of artifacts will be meaningful additions to an activity, as well as when specific objects to be represented are selected.

Acknowledgements

Funding for this project was provided by the Institute of Museum and Library Services (IMLS). Many thanks to Kizer Walker, John Saylor, Bing Pan and Helene Hembrook at Cornell University, Lydia Beall and Giselle Ellis in the Design Challenge department at the Museum of Science, Suzanne Spring and Angela Damery in the Interpretation Department at the Museum of Science, and Elissa Chin and Susanna Coit in the Research Department at the Museum of Science.

References

Engestrom, Y., R. Miettinen, & R.L. Punamaki, (Eds.) (1999). Perspectives on activity theory. Cambridge, U. K.: Cambridge University Press.

Grosslight, L., C. Unger, & E. Jay (1991). Understanding models and their use in science conceptions of middle and high school students and experts. Journal of Research in Science Teaching, 28(9), 799-822.

Harrison, A. G., & D.F. Treagust (1998). Modelling in science lessons: Are there better ways to learn with models? School Science and Mathematics, 98(8), 420-428.

Pan, B., G.K. Gay, J. Saylor, H.A. Hembrooke, & D. Henderson (2004). Usability, learning and subjective experience: user evaluation of K-MODDL in an undergraduate class. Paper presented at the Fourth ACM/IEEE Joint Conference on Digital Libraries (JCDL '04), New York.

Treagust, D. F., G. Chittleborough, & T.L. Mamiala (2002). Students' understanding of the role of scientific models in learning science. International Journal of Science Education, 24(4), 357-368.

Cite as:

Lindgren-Streicher A. and Reich C.,Visitor Interactions with Digitized Artifacts, in J. Trant and D. Bearman (eds.). Museums and the Web 2006: Proceedings, Toronto: Archives & Museum Informatics, published March 1, 2006 at http://www.archimuse.com/mw2006/papers/lindgren/lindgren.html

• Workshops
• Sessions
• Speakers
• Interactions
• Demonstrations
• Events
• Exhibits
• Best of the Web

• produced by
  
• Join our mailing list
• Search A&MI
  

published: March 27, 2006
analytics code updated: Oct. 28, 2010

Archives & Museum Informatics, 158 Lee Avenue, Toronto, Ontario, M4E 2P3 Canada
Telephone: +1 416 691 2516 | Fax: +1 416 352 6025 | E-mail:
Copyright © 2006 - Archives & Museum Informatics - All rights reserved.