Can Diagrams of your Knowledge Help you Learn? - Kyung Kim

Kyung Kim

You probably already know that knowledge is not stored willy-nilly in our minds. Rather, it is structured.

How knowledge is structured in the human mind fascinates Dr. Kyung Kim, a CISLL affiliate in the Department of Educational Technology, Research and Assessment, housed in NIU’s College of Education.

“Knowledge structure refers to how the concepts in a domain are related to one another in memory. We believe knowledge structures can be represented in a network form that can be depicted visually,” explains Kim.

To get an idea of how a knowledge structure can be depicted visually, imagine a diagram consisting of a dozen or so circles haphazardly placed on a page, with lines connecting some of the circles.

If the topic of the structure is the “heart”, then the individual circles might represent particular concepts related to the heart, such as “blood” “pulmonary valve”, “lung”, and so on.

A line between any two concepts indicates that they are conceptually connected in the person’s memory. Some concepts may be linked to many other concepts, whereas others may be only linked to one.

This sort of visual network of a student’s knowledge organization can help instructors assess the learning of their students.

For example, if a student knows that the pulmonary valve pumps oxygen-rich blood to the lungs, then that student would connect the concepts “pulmonary valve” and “lungs” if asked to draw their network. A lack of a line, or other deviations from an “ideal” network, would indicate that student’s knowledge is incomplete and may include misconceptions.

However, having students draw out their knowledge so that the instructor can provide timely and accurate feedback is not very efficient. Imagine being an instructor with hundreds of knowledge maps to grade every night, and you can see why. Instructors get tired and may make mistakes, and it might be days before students receive feedback.

In addition, most students are more used to writing essays than creating these types of concept maps.

In response to these and other educational needs and constraints, Kim developed a computer application called GIKS, which stands for Graphical Interface of Knowledge Structure.

Based on written input (e.g., a summary essay or a transcribed conversation), GIKS uses various existing computerized analytic tools (e.g., ALA-reader, Pathfinder) that together create a visual representation of the organization of preselected concepts mentioned in the input.

That visual representation would be similar to the circles and lines mentioned earlier.

GIKS allows an instructor (or researcher) to input a “master” text that provides the information of the to-be-learned knowledge structure (e.g., a textbook passage). A student (or research participant) can write and input a summary essay about the topic. This essay would reflect the student’s knowledge. Then GIKS produces visual representations of both the master text and the student’s essay and compares them.

“GIKS gives the student immediate feedback in an easy-to-understand visual way. It gives them information as to which concepts were correctly included, as well as missing concepts, and correct and incorrect links among the concepts,” explains Kim.

With GIKS, students do not need to wait for the instructor to provide feedback. Feedback is instantaneous.

Since the tool is web-based, it is ideal for online courses, and because it uses text as input, it can be used in many content domains. Thus far, Kim has studied learning in physics, biology, geography, and foreign language.

What is also nice about GIKS is that it allows a student to input a summary essay at different points in time. This allows the student (and possibly the instructor or researcher) to see how knowledge changes as a new domain is learned.

“In addition to being useful for assessment, we have also shown that students learn by receiving feedback via GIKS,” said Kim.

In a publication with co-authors Roy Clarianay and NIU professor Yanghee Kim, tenth-grade students in an online physics course who used GIKS as a reflective tool included more relevant connections and fewer irrelevant connections in their summary essays than did students who were given explanatory videos or test questions, other known ways to increase learning.

Excitingly, GIKS is being acclaimed by educational technology researchers as well as other academic communities.

Kim was recently awarded the 2019 AECT Young Researcher Award from the Association for Educational Communications and Technology, the most prestigious association in Educational Technology. The award recognized a GIKS study that analyzed participants’ text messages to explore the dynamics of online collaborative problem solving. His dissertation, which used GIKS to explore the influence of text structures on readers’ knowledge structures, also won the 2019 Timothy & Cynthia Shanahan Outstanding Dissertation Award from the International Literacy Association. A study that applied GIKS to geography was recently selected for the 2019 Journal of Geography Award from the National Council for Geographic Education.

Kim is now interested in augmenting GIKS with speech-recognition capability. This feature would allow him to examine how individuals’ knowledge structures, as revealed through their speech, can be harnessed and used effectively in real-world collaborative work settings.

“Groups spend too much time discussing information that everyone in the group already knows. However, there is little research on how unshared information can play a role in collaborative learning. Having GIKS analyze both text and speech would be an ideal way to help groups know what other members know and don’t know. Perhaps by emphasizing unique knowledge held by an individual, group learning could be enhanced,” volunteered Kim.

In sum, the answer to the question “Can diagrams of your knowledge help you learn?” is a resounding “Yes”—provided the diagrams were created by GIKS, of course.

If you are interested in learning more about GIKS, you can contact Professor Kim at kkim2@niu.edu.

About the author: Keith Millis is on CISLL’s executive board and is a professor of psychology at NIU.

How Does a Bilingual Environment Affect Language Development? – Milijana Buac

Milijana Buac

Language is a fundamental aspect of being human. Being able to understand and produce language allows one to create strong social and emotional bonds with friends and family and to fully participate in society.

It is therefore not surprising that when parents receive news that their young toddler may have a language disorder, they become worried and concerned.

According to Dr. Milijana Buac, this news may be problematic if the child is reared in a bilingual environment. “In my work, I noticed that the language-assessment tools were biased against bilingual children. They are biased because they were not normed using bilingual children but monolingual children.”

This is a problem because often the bilingual children score lower on these assessments when compared to monolinguals peers, sparking concerns of a delay.

“A low score by a bilingual child may not indicate a language disorder but just a language difference, reflecting the fact that they are learning two languages.” explained Buac.

A CISLL affiliate, Buac is on the faculty of the Speech-Language Pathology Unit housed within the Allied Health and Communicative Disorders Division within NIU’s College of Health and Human Sciences. Buac is a speech-language pathologist by training.

Using biased language-assessment tools is a problem potentially affecting many lives. In fact, 22% of children in the United States speak a language other than English at home.

Buac studies how a bilingual context affects language and cognitive development and how to make such language assessment tools less biased.

As one example of her research on language development, her dissertation used an eye-tracking visual world paradigm to examine how monolingual and bilingual children learn new words from non-native input.

In the visual world paradigm, participants are shown four unfamiliar objects on a screen. Each object is presented with a spoken presentation of a new “name” for each. The new name is a made-up word, so that it is unfamiliar to all participants. Later, participants are shown two of the objects and hear the word that was presented with one of them. Using an eye-tracker, the researcher measures whether and how quickly the child looks to the correctly-paired object. A shorter response delay and higher proportion of looks to the correct object indicates that the child has learned the new word.

In her dissertation, Buac was interested in how accents may impact the learning of new words. She presented the “names” in the visual world paradigm to typically-developing children in three different accents: native English, Spanish-accented English, and Korean-accented English.

The accents approximate how mono- and bilingual children may learn new words at home. English-Spanish bilingual children would be familiar with both native English and Spanish accents, whereas the monolingual children would only be familiar with the native English accent. Both types of speakers would be less familiar with the Korean accent.

She found that children living in a monolingual native English family learned new words best when presented by a native English-accented speaker, followed by a Spanish-accented English speaker, followed by a Korean-accented English speaker. However, children living in a bilingual environment showed no difference in learning between the native English and Spanish accents, and both of these led to higher learning than words presented in a Korean accent.

“The results showed that the familiarity of the accent matters. If you are familiar with the accent, then you learn new words faster than if you are not,” explained Buac. “I looked at learning because previous research has looked at how accents affect the time to process language but not on how accents affect learning.”

Buac is extending this research by looking at word learning in children who have a language impairment.

“The reason why I wanted to start with typically-developing children is to get a baseline. Now I would like to see if the accents are a greater impediment for children with a language disorder. A primary theory of language impairment assumes that the impaired individual has trouble processing spoken input. I would expect that children with a language disorder would be hindered even more than usual when presented with new words in an unfamiliar accent,” elaborated Buac.

As an expert in language disorders, Buac gets many questions from concerned parents.

In particular, she is often asked by parents of a child in a bilingual environment who have been newly diagnosed with a language disorder which language should be spoken at home.

“I tell them to speak which language they are most comfortable with. Never take away a language, because if a child is able to learn one language, they are going to be able to learn another. Parents often think it is better to stick to one language, let’s say English. But that could hinder the child if no one else in the family speaks English,” explained Buac.

As a new faculty member at NIU, Buac is busy planning future experiments and writing grants to fund research on how bilingual contexts affect language and cognitive development.

About the author: Keith Millis is on CISLL’s executive board and is a professor of psychology at NIU.

Even Brief Stints Abroad Help Students Learn a New Language- Mandy Faretta-Stutenberg

Mandy Faretta-Stutenberg

¿Hablas español? If your answer is “No”—would you like to?

Learning a new language opens up a whole new world of places, friends, experiences, and stories.

But we all know that learning a new language is not easy. That is why the courageous among us who do want to learn a new language often enroll in a foreign language (FL) course. The nice thing about taking a FL course is that you will most certainly learn—provided you pay attention in class, do the homework, and practice, of course. Watching foreign films over and over again with crossed fingers simply may not cut it.

But conventional wisdom says that the best way to learn a new language is to speak it every day with native speakers, surrounded by the language—that is, in an immersion setting.

Taking a three-hour-a-week course at your local college or university is likely not enough. Many students seeking “fluency” in a second language enroll in a study-abroad program, in which they take FL courses in a country where that language is spoken.

According to Dr. Mandy Faretta-Stutenberg, director of the Spanish Basic Language Program at NIU and FL learning researcher, many factors contribute to whether students will learn more studying abroad than studying at home. Moreover, it may be the case that study abroad impacts certain aspects of language development more than others.

“The research shows students who study abroad experience significant learning gains in measures of fluency—things like rate of speech and number of pauses—and also in pronunciation,” explained Faretta-Stutenberg.

However, according to Faretta-Stutenberg, the existing research is limited in two important ways.

The first is that the effect of immersion on the full spectrum of language learning is not well understood.

“Although fluency and pronunciation have been studied, very few studies have measured accuracy with regards to grammatical precision. For instance, do students who study abroad use verb tenses or other agreement structures accurately? Another understudied area is complexity. Students may be speaking more quickly, but are they using more complex or advanced structures than students who study a FL in their home country?” said Faretta-Stutenberg.

She argues that it is important to study the full gamut of language skills—fluency, accuracy, and complexity—because there are trade-offs between them. For example, someone might have high fluency but low accuracy. Someone else might be highly accurate but slow. If a researcher does not include the full range of skills in a single study then those trade-offs are not documented, and our understanding of the role of learning context in language development will be incomplete.

The second limitation of the existing research is that studies have been based primarily on study-abroad programs that last between a semester and a year. However, Faretta-Stutenberg points out that around 60% of study-abroad programs in the United States are classified as “short-term”—around 3-8 weeks in length—which falls substantially short of a semester, not to mention a full year.

In a recent study, she and her research team sought to address both of these limitations.

“We wondered whether there would be significant learning gains in a five-week study-abroad experience. Students expect that they will learn the language within that time frame but, frankly, the research has not been done,” explained Faretta-Stutenberg.

Faretta-Stutenberg collaborates with Janire Zalbidea at Temple University, and with Bernard Issa and Harriet Bowden at the University of Tennessee at Knoxville.

Working with Drs. Issa and Bowden, Faretta-Stutenberg set out to find out what students learn when they study abroad for five weeks in Spain. Because initial proficiency and experience with a language are important predictors of whether or not students experience learning gains, students were divided into two groups. Students in one group only had two semesters of Spanish under their belt, whereas students in the other group had taken six semesters before they studied abroad.

Both before departing and upon their return to the U.S., students in both groups completed a version of the “elicited imitation task.” In the task, students listen to sentences in Spanish and repeat them aloud to the best of their ability. The sentences become increasingly longer and more complex, and learner accuracy is scored using a rubric, which results in an independent measure of overall proficiency.

Once the students arrived in Spain, they were tested using a grammaticality judgment task and a lexical decision task. The first required the students to judge whether sentences were grammatically correct and the second required a judgment of whether a series of letters formed a true word in Spanish. They were tested again on these measures right before they came home.

“These tests measured various aspects of accuracy. The grammaticality judgment task measured students’ sensitivity to violations of morphosyntax, while the lexical decision task measured their word knowledge. We were interested in going beyond measures of fluency in order to contribute to our understanding of the linguistic impacts of study abroad,” elaborated Faretta-Stutenberg.

Using hierarchical regression, the researchers held constant initial scores on the grammaticality judgment and lexical decision tasks to determine whether performance on elicited imitation pre-departure would predict learning gains on those tasks.

When they saw the results, Faretta-Stutenberg was stunned. “I was skeptical whether we would see learning gains in such a small amount of time, but we did. It was exciting!” she exclaimed.

They found that early-stage learners (the two-semester students) showed significant gains on both grammar and word knowledge, whereas the later-stage learners only showed significant gains on word knowledge.

“The different types of gains in the two groups probably reflected the focus of language instruction in the courses they were taking over there,” explained Faretta-Stutenberg.

Although the study found significant learning gains, Faretta-Stutenberg emphasizes that there is a lot of variability among learners and there are several factors at play.

For example, her research has found that greater reported use of the target language while abroad predicts gains in some language abilities. “This relationship between using the language and language gains supports the claim that immersion may increase learning—but also indicates that learner actions play an important role,” said Faretta-Stutenberg.

Faretta-Stutenberg is now focusing on other research questions, such as how the study-abroad experience impacts learning once students return to their home universities, and how long the learning effects in different domains last.

As for the rest of us, we might feel a bit better pulling out our checkbooks to pay for a five-week study-abroad experience. According to Faretta-Stutenberg’s research, it is likely that we (or our loved ones) can learn more of a language in a little over a month than you’d think.

¡Excelente!



About the author: Keith Millis is on CISLL’s executive board and is a professor of psychology at NIU.