Allison J. Jaeger

If it's hard to read, it changes how long you do it: Reading time as an explanation for perceptual fluency effects on judgment

Perceptual manipulations, such as changes in font type or figure-ground contrast, have been shown to increase judgments of difficulty or effort related to the presented material. Previous theory has suggested that this is the result of changes in online processing or perhaps the post-hoc influence of perceived difficulty recalled at the time of judgment. These two experiments seek to examine by which mechanism (or both) the fluency effect is produced. Results indicate that disfluency does in fact change in situ reading behavior, and this change significantly mediates judgments. Eye movement analyses corroborate this suggestion and observe a difference in how people read a disfluent presentation. These findings support the notion that readers are using perceptual cues in their reading experiences to change how they interact with the material, which in turn produces the observed biases.

The Cambridge Handbook of Multimedia Learning: The Individual Differences in Working Memory Capacity Principle in Multimedia Learning

Effective use of the working memory system is critical for successful learning, and this assumption has motivated much of the work on multimedia instruction. Interestingly, the limited capacity of human working memory has been invoked as part of explanations for both advantages and disadvantages of multimedia learning in comparison with learning from text or pictures alone. This chapter reviews several lines of reasoning that have guided explorations of the role of working memory in multimedia learning, including approaches that have emphasized the modality-specifi c buffer system and the potential for overloading the limited resources that are available to learners; as well as a newer approach that considers working memory capacity as an individual differences variable representing attentional control. What Is the Individual Differences in Working Memory Capacity Principle? Learning from multimedia is a higher-order cognitive process that relies on many subprocesses to be successful. For the purposes of this volume, multimedia comprehension is defi ned as learning from a combination of words and images. As such, it requires all of the processes involved in developing comprehension from written text or spoken discourse, in addition to the processes needed to interpret and represent information from images, diagrams, graphs, animations, or other forms of visualizations, as well as the processes necessary to alternate between and integrate multiple representations. Since the fi rst studies on learning from multimedia, the Individual Differences in Working Memory Capacity 599 construct of working memory and the nature of the working memory system has been cited as one of the primary motivations for using multimedia instruction. Interestingly, the limited capacity of human working memory has been invoked to help explain both advantages and disadvantages of multimedia learning in comparison with learning from text or pictures alone. This chapter describes several approaches to exploring the role of working memory in multimedia learning, including those that have emphasized the modality-specifi c buffer system and the potential for overloading the limited resources that are available to learners. As an alternative, a newer approach considering working memory capacity as an individual differences variable representing attentional control is discussed in greater detail. Working Memory as a System with Modality-Specifi c Buffers In much of the work on multimedia learning, the construct of working memory refers to the functioning of the multiple component system originally proposed by Baddeley and Hitch ( 1974 ). This model defi nes working memory as the mental workspace used for the short-term storage and manipulation of information required for diverse cognitive tasks. The working memory system was initially described as having three main subsystems: the visuospatial sketchpad for holding and manipulating visual-spatial information; the phonological loop for maintaining and rehearsing verbal information; and the central executive, which is an attentional control system involved in the coordination of performance on separate tasks, including selective attention, retrieval from long-term memory, set shifting, and inhibition, as well as serving as the system that takes over when either buffer is overloaded. A common approach within the multimedia literature has been to determine how different forms of presentation may enable readers to maximize performance via the use of both buffers. This tradition has largely used selective interference/dual-task paradigms, and investigations of selective defi cits in groups diagnosed with specifi c disabilities, as a means to support the existence and independence of the components of the working memory system and to explore the utility of multiple modes of presentation (see Chapters 9 and 11 on the modality principle for a more complete description of this approach). The Limited Capacity of Working Memory Because multimedia learning requires processing from multiple sources, codes, channels, or modalities of information, one of the earliest concerns was its relation to the limited capacity of the human working memory system. In particular, there has been a great deal of concern about how having to attend to multiple information sources may put the learner under load. Many studies on multimedia learning have found that when learners are Wiley, Sanchez, and Jaeger 600 given more information, including additional information that should be helpful for their understanding, they actually learn less (not more). Many of these studies have been situated in the literature using cognitive load theory as a construct to explain performance. Cognitive load theory ( Sweller, 1988 ) originally emerged from a line of research on mathematical problem solving. In this work, it was found that naive students were able to learn mathematical principles more effectively from worked examples than from less-structured problem-solving attempts. The explanation offered for these results was that the demands of engaging in acts of problem solving, while also learning from those acts, were overwhelming and required too many resources for naive learners to both perform the necessary operations and abstract the principles. In other words, the working memory system was overloaded by attempting to learn from problem solving. This overloading concern was formalized as cognitive load theory and has since been applied to many other educational contexts, including, most prominently, multimedia instruction (see Chapter 2 for a more complete description of this approach). Given this research, there has been a long tradition of assuming that poor learning outcomes obtained in multimedia contexts are due to learners being placed under load by the materials, meaning that the amount of information they are given, as a function of how information is presented or included in particular multimedia contexts, overwhelms the capacity of their working memory system. In this literature, the amount of load is frequently discussed as a characteristic or a property of the learning materials themselves . Working Memory Capacity as an Individual Differences Variable An alternative approach examines working memory involvement by considering the effects of individual differences in working memory capacity (WMC). In this approach, WMC is generally considered to be a trait of individuals in relation to their ability to use their working memory system. In other words, the main construct of interest in this approach is the central executive, or the ability to control one’s attentional resources. Complex span tasks have been explicitly designed to assess the functioning of the central executive; that is, they test for the maintenance of information in immediate memory and retrieval from secondary memory in the face of interference from an ongoing processing task ( Daneman & Carpenter , 1980 ; Unsworth & Engle , 2007 ). As opposed to simple span tasks that do not involve an intervening processing component, complex span tasks involve both a memory storage component and a processing component. For example, in the operation span task (OSpan), learners participate in trials consisting of verifying a math equation and then remembering a word or letter presented after the mathematical verifi cation task. These trials are presented in sets of two to seven items. At the end of each set of trials, a participant is asked to recall the to-be-remembered items. The following is an example set of trials during the OSpan task: Individual Differences in Working Memory Capacity 601 While simple memory span tasks (such as remembering a string of words or digits or sequences of spatial locations) without an additional intervening processing task are considered to assess primary memory, or the capacity of the buffers in the working memory system, the presence of the processing task in complex span tasks (like the equation verifi cation task in OSpan) renders it a measure of the central executive system, or the ability to allocate or control one’s attention. In complex span tasks, participants are required to maintain, update, and retrieve to-be-remembered information from secondary memory, engage in active processing in a secondary task, switch back and forth between different task sets, and avoid proactive interference from previous trials and from intervening stimuli. Most modern versions of complex span tasks are based on the reading span task (RSpan) originally developed by Daneman and Carpenter ( 1980 ). Although in the initial version of the reading span task sentences served as stimuli for both the processing and the memory components of the task (participants were asked to verify sentences and to remember their last words), newer versions have improved upon the design of the task by making the processing and memory elements independent. Newer versions of complex span tasks also use many different processing tasks as the intervening stimuli other than just sentence processing (e.g., verifying math equations, making symmetry judgments, reordering lists; see Conway et al., 2005 ) so that they might represent a more generalized measure of cognitive ability. Performance on complex span tasks thus provides an individual differences measure that represents the central executive, or the ability to control one’s attention. This estimate of generalized ability is particularly robust when a composite performance measure is derived from performance on multiple complex span tasks. The individual differences approach to WMC has become more popular over the past 20 years, a trend driven largely by the development of and advances in complex span tasks to tap this construct, as well as an increasing interest in the role of individual differences in executive functioning and attentional control in many cognitive tasks ( Engle , 2002 ; Oberauer , 2009 ). Much research has co

Different Approaches to Assessing the Quality of Explanations Following a Multiple-Document Inquiry Activity in Science

This article describes several approaches to assessing student understanding using written explanations that students generate as part of a multiple-document inquiry activity on a scientific topic (global warming). The current work attempts to capture the causal structure of student explanations as a way to detect the quality of the students’ mental models and understanding of the topic by combining approaches from Cognitive Science and Artificial Intelligence, and applying them to Education. First, several attributes of the explanations are explored by hand coding and leveraging existing technologies (LSA and Coh-Metrix). Then, we describe an approach for inferring the quality of the explanations using a novel, two-phase machine-learning approach for detecting causal relations and the causal chains that are present within student essays. The results demonstrate the benefits of using a machine-learning approach for detecting content, but also highlight the promise of hybrid methods that combine ML, LSA and Coh-Metrix approaches for detecting student understanding. Opportunities to use automated approaches as part of Intelligent Tutoring Systems that provide feedback toward improving student explanations and understanding are discussed.

The Roles of Working Memory and Cognitive Load in Geoscience Learning

ABSTRACT Working memory is a cognitive system that allows for the simultaneous storage and processing of active information. While working memory has been implicated as an important element for success in many science, technology, engineering, and mathematics (STEM) fields, its specific role in geoscience learning is not fully understood. The major goal of this article is to examine the potential role that working memory plays in successful geoscience learning. We start by reviewing two popular approaches to studying working memory in science learning—the individual differences approach and the cognitive load approach—and consider how these two approaches have been utilized in geosciences education research. Next, we highlight examples of various activities and curricular materials that have been used in geoscience classrooms in an effort to improve student learning and offload working memory resources, including using concept sketches and providing varying levels of scaffolding. We outline recommendations about how to structure geoscience classrooms and labs to maximize student learning and suggest potential avenues for future research aimed at investigating the role of working memory in geoscience learning.

Leveling the playing field: Grounding learning with embedded simulations in geoscience

Although desktop simulations can be useful in representing scientific phenomena during inquiry activities, they do not allow students to embody or contextualize the spatial aspects of those phenomena. One learning technology that does attempt to combine embodiment and grounded experience to support learning in science is Embedded Phenomena. The objective of this research was to investigate the effectiveness of a classroom-based Embedded Phenomena activity for learning in geoscience, and to investigate whether individual differences in spatial skills had an impact on the effectiveness. The simulated scientific phenomenon was earthquakes, and 44 fifth grade (10-year old) students learned from a unit containing both content instruction and simulations. In the embedded condition, 15 earthquake events were simulated within the classroom space and students enacted the computation of epicenters with strings and their bodies. Students in the non-embedded condition received the same content instruction and did the same activities, but the epicenter computations were done with maps instead of with students’ bodies. Students in the embedded condition showed greater learning gains overall. Further, the Embedded Phenomena activity attenuated the effect of individual differences in spatial skills on learning in science such that low spatial individuals performed as well as high spatial individuals in the embedded condition.

Sketching and Summarizing to Reduce Memory for Seductive Details in Science Text.

Seductive details refers to interesting pieces of information within an expository text that are only tangentially related to the target concept (Garner, Gillingham, & White, 1989). When the presence of this information results in reduced comprehension, this is called the seductive details effect. Previous work has found the seductive details effect to be resistant to reduction via various instructional manipulations. One avenue that has not been investigated as a tool for reducing the seductive details effect is having students generate sketches. A growing body of research suggests that sketching activities are beneficial for science learning and, moreover, that sketching can improve learning from science text (Ainsworth, Prain, & Tytler, 2011; Van Meter, 2001). The goal of the present research was to investigate the impact of sketching as compared to generating summaries or thinking silently on recall and comprehension of a text that included seductive details. Across two studies, the seductive details effect was replicated; generating sketches did not eliminate it. In Experiment 2, students compared their sketches and summaries to correct ones and were asked to identify differences between them. Results indicated that participants in the summary group recalled the most core concepts and demonstrated the highest comprehension. These results suggest that sketching may not be effective for eliminating the seductive details and that having students generate summaries with feedback may be more successful. These findings inform the design of scaffolding to support learning from naturalistic science text with its distracting details.