Effects of Redundancy and Paraphrasing in University Lessons: Multitasking and Cognitive Load in Written-Spoken PowerPoint Presentation

This paper is about the use of a widespread teaching tool: the slide presentation used in face-to-face, systempaced university lessons. It is produced by lecturers to support students’ comprehension during listening; nevertheless it poses elaboration requests to the audience which should be taken into consideration at the planning stage and in formulating its verbal content. The paper reports the results of a survey conducted with 163 University students who were asked to listen to a lecture accompanied by a PowerPoint presentation, prepared according to the most frequent formats. The written presentation had 3 degrees of concision/ redundancy: it had a fully redundant with the oral message, partially redundant (main points in key words), or had a different linguistic form (paraphrase of the message). Furthermore, information in written text and spoken message could have had the same order or they could be scrambled. The results showed that, subjectively, students judged comprehensible every kind of presentation. However, learning tests demonstrated that paraphrasing negatively affected learning, while changes in the order of presentation did not, at least in the synthetic main point – key word presentations. The study suggested that the concise, only partially redundant, presentation is the one which leads to better results, both in the ordered and in the scrambled version. DOI: 10.4018/jdldc.2012070101 2 International Journal of Digital Literacy and Digital Competence, 3(3), 1-11, July-September 2012 Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. schools, ppt e wp became standards – as they are available on almost every pc, they are easy to be used and do not ask for a specific training. However, they are not exempt from criticisms. About ppt, for example, it was said that it fragments thought, makes it impossible to present data with efficacy, forces to use templates which don’t correspond to the text’s conceptual structure. As Tuftee wrote (2003) http://www. wired.com/wired/archive/11.09/ppt2.html) the PowerPoint style routinely disrupts, dominates, and trivializes content. Parallel to these criticisms published on the web, on the same web we may find guidelines, advices and also researches on the conditions which may have a positive or negative effect on comprehension of presented information. Conditions which may regard very obvious deficiencies such as poor readability, due to insufficient size of characters or to the lowcontrast text-background. But they may also concern aspects more complex and less intuitive, such as the redundancy between text and message information, the interference between written and spoken information, the request to implement complex search to connect the sentences heard with a specific part of the text presented on the screen. Some of these aspects are under consideration in this paper, which has the aim to examine the effect of redundancy and concision, of changes in linguistic formulation and order of presentation. First we will describe a framework which tries to explain the ppt producer’s and receiver’s intentions and objectives. Next we will identify some of the factors that may affect ppt effectiveness and will describe the research conducted. DOES POWERPOINT PROMOTE LEARNING? As a starting point, it seems appropriate to analyze the point of view of teachers and learners about the usefulness of PowerPoint. From a recent study of Cantoia et al. (2011) we obtain a description of the intentions and objectives of the teachers. The sample interviewed in their survey – a group of University teachers – states that they use it with the intention of promoting the understanding of their lessons. The interviewed teachers also said to prefer concise formats, partially redundant with the spoken message, as these formats should facilitate the identification of the structure of the lesson. The aim seems to be to give a cognitive guidance, by which, according to Richard Mayer, one wants to make sure that the audience members build appropriate knowledge in their memory (Atkinson, 2004). With regard to opinions and behaviors of students, it is possible to refer to researches showing that students prefer classes where there is a presentation with ppt compared to transparencies or absence of every presentation tool (Bartsch & Cobern, 2003; Blokszijl & Naeff, 2004). In fact they find classes which use ppt more compelling, clear and organized (Apperson, Laws & Scepansky, 2004). They also prefer when content is organized in a synthetic way with the use of graphs, diagrams and bullet points (Cantoia et al., 2011). Other studies also show that students learn more when the lecture is supported by a PowerPoint presentation than when it is not (Blokzijl & Andeweg, 2005; Paoletti, Rigutti, & Guglielmelli, 2008), although the results are not conclusive in all researches (Savoy et al., 2009). The conditional success of the ppt can be explained by the fact that it can act as an Advance Organizers (Ausubel, 1962; Mannes & Kintsch, 1987) when it provides a schematic and ordered overview of the content. In these conditions, the presentation may help in the identification of relevant information and of the structure and organization of the lesson (Clarke, 1992; Stull & Mayer, 2007). When it takes other forms, the result does not seem as favorable, instead. Thus, it appears that there is some degree of agreement between lecturers and learners on the optimal form of presentation: one that shows the structure organization of the lesson, with short, concise sentences or key points. The effectiveness of this form of presentation during International Journal of Digital Literacy and Digital Competence, 3(3), 1-11, July-September 2012 3 Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. classes in which there is a double presentation (written text on the screen and spoken message) seems to be confirmed, for example, by the literature on The Redundancy Effect (Adesope & Nesbit, 2012; Mayer & Johnson, 2008; Paoletti & Rigutti, 2009) that we will describe later. But this type of presentation is not the only format that can be observed in classrooms and was described in the literature on the subject. The literature (Farkas, 2005; 2006; Bohec & Jamet, 2008; Paoletti, 2012) refers to many types of presentation, including three which we will consider here as prototypical formats of text on slides: 1. The already described concise text: an outline of main points which summarizes key information. 2. A redundant text, which reproduces verbatim the oral message as it is read by the speaker. 3. A set of sentences and phrases which are not read verbatim, but are modified by the lecturer during the presentation: the lecturer re-implements/re-formulate the text on the slide, choosing a different verbal formulation, a loose paraphrase. In this study, these formats were also encountered during a series of observations and informal interviews conducted in a pilot study. During this pilot study we discerned two interesting factors. First we noticed that the verbal-spoken formulation of the presenter may be more or less similar to the text written on the screen. The text on screen may be read verbatim, may be paraphrased closely (by glossing/expanding the main points) or may be paraphrased loosely (producing a very different text). Second we observed that, with the first and last kind of presentation (concise text, not verbatim sentences) sometimes the speaker gives the information a different order, moving inside the slide and among slides. In this study we examined five different forms of presentation, assuming a positive effect on ease of processing and learning of similarity between written text and spoken message and of conciseness. More in particular, the factors we have studied are: 1. The conciseness of the written text. 2. The maintenance of the same order / sequence of topics. 3. The degree of correspondence between written and spoken linguistic formulation. These are factors that affect the ability to process information, call into question the limits of our working memory and the inability to perform multiple linguistic tasks at the same time, as demonstrated, for example, by scholars of the Cognitive Load Theory. COGNITIVE LOAD THEORY (CLT) VERBAL REDUNDANCY AND CONCISION Cognitive load is a theoretical construct that refers to the demands placed on the limited capacity of working memory as learners process instructional information. Cognitive load researchers have used the term redundancy effect to refer to situations in which learning is hindered by the presentation of identical information in different formats (Chandler & Sweller, 1991; Kalyuga, Chandler, & Sweller, 1999; 2001; Mayer, 2005; Sweller, 2005). Learning can be impaired because learners expend cognitive resources to elaborate and integrate a verbal text and a redundant diagram (Chandler & Sweller, 1991) or concurrent animation, narration, and on-screen text (Moreno & Mayer, 2002). In a related line of reasoning it was also predicted that concurrent spoken– written presentations might inhibit learning by inducing extraneous processing. The reference to a recent meta-analysis by Adesope and Nesbit allows us to understand what consequences may have, in the context under consideration here a University lecture to listen to the message and read the text on the screen (Adesope & Nesbit, 2012). Data analysis in this meta-analysis investigated the effects of spoken-only, written-only, and spoken–written 4 International Journal of Digital Literacy and Digital Competence, 3(3), 1-11, July-September 2012 Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. presentations on learning, retention and transfer, in 57 studies, mostly on postsecondary students. The analysis performed lead to the conclusion that there are advantages when the presentation is redundant (written and spoken) and not just spoken. In other words, the expectation of students and teachers that it is beneficial to accompany the auditory presentation with a deck of slides is confirmed. The effect is particularly significant when the presentation on the slide is concise (lowredundancy), when the learner is a light reader (but the effect is also present with good readers). In addition, the double presentation is particularly advantageous when it is continuous and system-paced, as in face-to-face lessons (the teacher talks and the learner listens and cannot stop, rewind and replay the presentation). The learner in this case (low-redundant writtenspoken presentation) would benefit from the permanence of the written text to retrieve pieces of information already processes if he has lost trace of them because of distraction or did not understood them (Adesope & Nesbit, 2012). And if so far we talked about text-andmessage redundancy, a similar effect also occurs when the presentation is threefold: message, text on the screen and image, as demonstrated by a research of Mayer and Johnson (2008). While large text next to an animation poses processing problems, short texts help to process the animation. SIMILARITY BETWEEN SPOKEN-WRITTEN TEXT: SCRAMBLING AND CHANGES OF LINGUISTIC FORMULATION When text and message are similar and redundant, because the same information is given in two formats, one can assume that one is supporting the other and that the double processing is not burdensome, indeed is beneficial (Principle of Multimedia Learning, most often verified with text and figure; Mayer, 2001). But what happens when the text and the message do not look similar because they don’t have the same linguistic formulation and / or the same order / sequence of presentation? In our empirical observation we found that these changes in the spoken presentation (Paraphrasing and Scrambling) are frequent and are caused by many reasons the ppt was prepared in advance, for other circumstances and reasons, it was prepared for distant students, who could not listen to the oral message none of which is concerned with ease of processing. Paraphrasing and scrambling are likely to be two factors which impose a cognitive load and prevent a smooth processing. In both cases, it is required to perform a search on the information displayed on the screen, while listening, to find the corresponding information (i.e., looking for the segment of the text corresponding to the spoken segment, evaluating the correspondence ...) and then a match between sources. It is not impossible nor infrequent to carry out these multiple processing and indeed in other research settings these are text manipulations intentionally used to improve retention. For example, it has been suggested that changing the order of information could be a factor that prevents the superficial processing of the text, because it increases deep processing and then memory (Lockhart, Craik, & Jacoby, 1976). However this request seems to have a negative impact on the understanding of the less skilled readers and of those who have low prior knowledge (Mannes & Kintsch, 1987). Changes in the linguistic formulation (paraphrasing) demand the learner to process two stimuli with the same content but with two different linguistic forms. We know that we can do it, because during the processing of a text, we normally lose its surface form and retain the cognitive content of the text, the mental representation of its meaning (Castelfranchi & Parisi, 1980). Recalling this content we cannot reproduce it verbatim: with the exact words, the exact syntax. The problem is that in a situation in which the learner wants to process a sentence (heard) and its paraphrase (written), he/she is located in front of two stimuli that compete for his/her limited attentional and processing resources and interfere. International Journal of Digital Literacy and Digital Competence, 3(3), 1-11, July-September 2012 5 Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. The processing of scrambled and of paraphrased information is generally feasible, but it requires cognitive resources. It is possible that there is a breaking point, a threshold, beyond which the ppt presentation no longer favors listening and beyond which trying to use the text on the slide while listening to the message makes it more difficult to process the information. What we will try to assess is whether conciseness can at least partially reduce the processing difficulties, due to processing of paraphrased information and of scrambled information. By reducing the size of text segments that must be analyzed during listening, and therefore the amount of written text, conciseness might increase available working memory space and reduce cognitive load. HYPOTHESIS AND EXPECTATIONS OF THE RESEARCH In our expectations, the concise text which presents an outline of key points is the preferable presentation, because it shows which pieces of information are most relevant and how they are structured. A fully redundant deck of slides (like a prompter) is a format that does not create interference between listening and reading and so it may have some advantages compared to the third format – non verbatim sentences – which could be the worst condition, because it requires the elaboration of two interfering sources of information. Finally, processing may be negatively affected by a different ordering of topics between ppt and message. We are accustomed with this kind of request: a speaker may in fact decide, during the oral presentation, to give a different emphasis and priority to some pieces of information by changing the order of presentation of topics. The same happens when the author of a book tells the story with flashbacks and changes in perspective (Kintch et al., 1977). However this may be a factor that makes it difficult to perform search and match activities in order to integrate message and text, especially when the text is verbose. The listener may try to shift attention from one source to the other. We guess that this attempt can only succeed up to a certain limit that is up to a certain amount of information. The focus of this paper is therefore on the