Larkin, J.H. & Simon, H. A. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11, 65-99. 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... [18, 19, 21] Diagrammatic representation and reasoning became recently a field of intensive research, starting from basic theoretical works on the fundamentals of the use of diagrams for representation of knowledge and formal reasoning, concerning both its psychological and cognitive sources [9, 26] and mathematical foundations [3, 10, 20, 40] Applications in various fields are also devised and investigated, like in qualitative analysis of physical systems [16, 24] visual programming, graphical interfaces, and mathematics [4, 9, 10, 17, 19, 20, 21] In this paper we will argue that the ....

J.H. Larkin, H.A. Simon, Why a diagram is (sometimes) worth ten thousand words, Cognitive Science 11 (1987) 65 99.

....on maintaining some form of equivalence in work. Algorithmic optimization. Algorithmic optimization relies on the fact that differences in encoding or procedure in can create differences in performance without changing the essential outcome i.e. without changing the function being computed [13]. For succinctness, we shall gloss over the many possible variations in terminology (e.g. data structure instead of encoding , algorithm instead of procedure , etc. Normally, differences in encoding and procedure go hand in hand. For instance, for two different encodings of some data, ....

....computing science, which normally fixes the function being computed and the underlying computational infrastructure [2] Performance can be optimized according to a variety of measures (e.g. speed, memory usage, etc. In changing data encoding, the information content is presumed to be unchanged [13]. In computing science, a familiar example is the change from singly linked lists to doubly linked lists. Such a change can make various list operations simpler and quicker to perform. In cognitive support terms, optimization normally means the introduction of artifact designs that reduce ....

[Article contains additional citation context not shown here]

J. H. Larkin and H. A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1):65--99, 1987.

.... A large body of literature supports the view that the way in which information is represented, external to the user has a fundamental effect on the users ability to perceive relationships within the information, and to reason about the information with respect to some task or activity [6]. Hutchins [5] goes further, and suggests that in collaborative tasks the appropriate unit of analysis is the distributed cognitive system of human and system actors, technological artefacts and external information representations. Thus the task of controlling the flight of an aeroplane is ....

J.H. Larkin and H.A. Simon. Why a Diagram is (Sometimes) Worth Ten Thousand Words. Cognitive Science 11:65-99, 1987.

....are understood, the ability to construct GLIF representations that are equivalent to the original text guideline is crucial. Specifically, there are two perspectives from which the original guideline and the GLIF encoded guideline may be judged as equivalent: informationally and computationally [10]. If all of the information that can be inferred from the original guideline can also be inferred from the GLIF encoded guideline, and vice versa, then the two guidelines are said to be informationally equivalent, where they both contain the same concepts and relations. On the other hand, if all ....

Larkin J, Simon H. Why a diagram is (sometimes) worth ten thousand words. Cog Sci 1987; 11 (1): 65-99.

....represented in some programming language; it is a document representing the set of activ ities the computer should perform to solve the problem. One important characteristic of a language is its expressive power: what concepts can be represented and how easy it is to work with this representation [Larkin and Simon, 1987]. I am using expressive power in a loose sense. It can mean different things in different contexts. For example, for a knowledge representation application, the expressive power of a representation would be characterized by properties like representational adequacy, inferential adequacy, ....

J.H. Larkin and H.A. Simon. Why a diagram is (sometimes) worth ten thousand words. Co#nitive Science, 11(1):65 99, January 1987.

....are understood, the ability to construct GLIF representations that are equivalent to the original text guideline is crucial. Specifically, there are two perspectives from which the original guideline and the GLIF encoded guideline may be judged as equivalent: informationally and computationally[10]. If all of the information that can be inferred from the original guideline can also be inferred from the GLIF encoded guideline, and vice versa, then the two guidelines are said to be informationally equivalent. Thus, both must contain the same concepts and relations to be informationally ....

Larkin J, Simon H. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science 1987;11(1):65-99.

....the relationship between map encoding schemes and processing activities and user tasks. We also looked at processing activities and user tasks where map based visualizations were not preferred even though the domain is spatially oriented. 4. 1 Cognitive Processing Based Analysis Previous research [2 4, 6] has shown that e#ective visualizations allow users to substitute quick perceptual inferences for more di#cult logic inferences. Thus the display of information is dependent upon the cognitive processing required by a user s task. To analyze the data collected during our user study we classified ....

Jill H. Larkin and Herbert A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1):65--99, 1987.

....visualization utility is dependent not only upon the task for which a visualization is created, but also on the abilities of a user to interpret the presented visual display. 2.2. 1 Cognition Research into the relationships between visualization and cognition was pioneered by Larkin and Simon [Larkin Simon 87] whose work has become the foundation of most of today s research e#orts. The theoretical basis behind this approach stems from a fundamental principle of cognitive science: the representation of the problem provided to a problem solver can a#ect his, her, or its task performance [Woods 95] ....

....We then discuss the criteria with which we judge the utility of a visualization. Finally, we put it all together into a three step approach for e#ective visualization design. 4. 1 Overview Perhaps one reason why the visual representation of information has been shown to not always be e#ective [Larkin Simon 87, Petre 95, Vessey 91] is that very little research has focused on the factors that make a visualization useful. Only in the last 10 15 years, as the amount and availability of data has dramatically increased, have researchers begun to focus on the utility of visual designs. Yet we are still ....

[Article contains additional citation context not shown here]

J. H. Larkin and H. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1):65--99, January-March 1987.

....make comparison to related problems and existing solutions in later chapters. From this analysis, the areas in the GRAIL browsing tool where improvement can be made are drawn out. 2.1 A Framework for Information Visualisation Design The design of visualisation is complex process. Many frameworks [14, 162, 163, 111, 85, 41, 42, 91, 73, 138, 176], cognitive models [99, 135, 123, 133] taxonomies [100, 98, 177, 84, 146, 181, 180] design principles [169, 86] and metrics [160, 96, 21] have already been proposed in relation to graphical information processing in general, and in 36 2.1. A FRAMEWORK FOR INFORMATION VISUALISATION DESIGN 37 some ....

....benefit of using external representations, and in particular graphical ones, is the ability to 2.1. A FRAMEWORK FOR INFORMATION VISUALISATION DESIGN 38 shift some internal cognitive processes and information into the external, in most cases changing the nature of information processing tasks [14, 91, 178, 138] (termed Computational off loading in [138] The benefit is brought about through Re representation and Graphical constraining. Re representation refers to the process of looking into the information under a different perspective or abstraction, thereby filtering away irrelevant information for ....

[Article contains additional citation context not shown here]

LARKIN, J. H., AND SIMON, H. A. Why a Diagram is (Sometimes) Worth Ten Thousand Words. Cognitive Science 11 (1987), 65--99.

....Card, Mackinlay, and Schneiderman [2] de ne Information Visualization as: the use of computer supported, interactive, visual representation of abstract data to amplify cognition , where cognition is de ned as the acquisition or use of knowledge . Apart from anecdotal evidences, formal studies [17] have 2 also supported the notion that information visualization can help in cognition. Visualization takes advantage of the perceptual capabilities of the user to help in cognition. It helps the user in understanding information in several ways [2] it allows the user to recognize patterns ....

J. Larkin and H. A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1):65-99, 1987.

....us to replace All As are 17 A B C Figure 2.3: Euler s Circles: All As are Bs, All Bs are Cs Bs with Either all As are Bs, or no A is C . Thus, we cannot use Euler s Circles to infer that (C n B) A = One of the earliest accounts dedicated speci cally to diagrammatic representations is [LS87] In this paper, Larkin and Simon claim that diagrams help solving problems, by facilitating more ecient search and identi cation procedures than sentential representations do. The claim is that diagrams can group together the problem elements which are used together in inference steps, thereby ....

....is type referential. A direct consequence of token referentiality is that a referent can only be picked by one symbol; prominent consequences of this, such as the concentration of information related to one element, and the ease of search brought forth by it, were already noted by Larkin Simon [LS87] as mentioned above. Stenning draws a stronger, intuitively appealing, but in my opinion, unjusti ed, conclusion: If a referent can only be picked out by one symbol on a diagram, then it cannot be picked out by two logically unrelatable ways. The complexity of reasoning in type referential ....

Jill Larkin and Herbert Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11:65-99, 1987.

....without any additional support. Multiple representations have been hypothesized to help because they provide redundancy at different levels of abstraction [6] pre sent complementary aspects of the subject matter [2] influence the allocation of attention [16] induce distinct cognitive processes [10], or provide a better fit to individual learning styles. Salzman, et al. 17] recently provided support for this idea in a conceptual learning task involving electric fields. In their study, high school students who interacted with both an exocentered view and an endocentered view of an electrical ....

Larkin, J., Simon, H., Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11, 1987, pp. 65-99.

....as an external memory aid, and this supports the claim here that explicit representation of planning by a graphical representation of goals and subgoals would be an effective support tool. They also noted that the key to successful solution of a problem was an appropriate labelling of a diagram [Larkin Simon 1987]. In statics and dynamics problems there are often cases of instantiation of general rules where explicit instantiation would have both diagnostic and supportive advantages. One example is the resolution of a force into two perpendicular directions. This is a nontrivial action for a novice who ....

Larkin J.H. and Simon H.A. [1987]. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science 11 65-99.

....to corresponding objects in the represented world is laconic, is to imply that distinct objects in the diagram refer to distinct objects in the represented world. This property, typically referred to as icon identity, very often holds of diagrams. Indeed, in the seminal work of Larkin and Simon [14] it is taken as one of the major de ning characteristics which di erentiates diagrammatic from textual representations. Diagrams are typically token referential systems, whereas texts are typically token referential. In token referential systems di erent tokens of the same type will refer to ....

J. H. Larkin and H. A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11:65-99, 1987.

....picture to solve this sort of problem; they nd it impossible to complete without it. Another reason why it is bene cial to consider diagrammatic representations is because of possible computational gains that come from di erent representations. There are good discussions regarding this point in [LS95] and [Slo95] A large part of the study of computer science de nitely involves the data structure used to represent what is being reasoned about. Computational eciency depends partially on how good this representation is. By nding a better representation, the computational eciency and ease at ....

....system based on Venn 4 Qin and Simon (in [QS95] actually found close links between the quality and accuracy of diagrams drawn by people and their understanding of the concepts in Einstein s 1905 paper on special relativity, which was published with no diagrams. 5 This example was also shown in [LS95]. 1 INTRODUCTION 7 x Figure 4: Diagram used as visual aid in geometry example. diagrams that is sound and complete for some speci c form of syllogistic reasoning. This is an amazing step in the right direction as it shows us that some diagrams have already been taken seriously and proven to be ....

Jill H. Larkin and Herbert A. Simon. Why a Diagram is (Sometimes) Worth Ten Thousand Words. In Glasgow et al. [GCN95], chapter 3, pages 69-110.

....examined the inherent constraints of diagrams (topological, geometric, spatial and so forth) to explicate their computational benefits. The second dimension which has been studied (particularly from an HCI perspective) concerns features and properties which impact upon the cognition of the user [4, 7, 16, 19, 29]. Recently a careful examination [9] of analogies (and disanalogies) between typical text based languages and diagrammatic languages has sought to unify the above two dimensions of naturalness . The examination was quite revealing about both similarities and di#erences in the textual and ....

J. H. Larkin and H. A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11:65--99, 1987.

.... particular areas of interest and expertise [Leigh et al., 1996; Smith and Mariani, 1997; Gutwin and Greenberg, 1998, Jaa Aro and Snowdon, this volume] Recent work in providing multiple representations to enhance learning have implied that this is a non trivial problem [Bibby and Payne, 1993; Larkin and Simon, 1987]. It has been shown that students perform better in tests when they learn a concept given more than one representation than students given only a single representation do [Salzman et al., 1998] However it has also been shown that this is not necessarily always the case; another study has found ....

Larkin, J., Simon, H. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 1987; 11: 65-99.

.... Previous studies of diagrammatic representations have typically sought to explicate either computational benefits of diagrams through analysis of inherent constraints [9, 60, 123, 124, 125] or (from an HCI perspective) features and properties which impact upon the cognition of the user [39, 43, 85, 109, 129]. Studies such as [41, 130, 148] have indicated that the most e#ective representations are those which are well matched to what they represent, in the context of particular reasoning tasks. Furthermore, it has been demonstrated in [43, 110, 112] that pragmatic features of diagrammatic ....

J H Larkin and H A Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11:65--99, 1987.

....Relationships When they are solving problems, human beings use both internal representations, stored in their brains, and external representations, recorded on a paper, on a blackboard, or on some other medium. Larkin and Simon, Why a Diagram is (Sometimes) Worth Ten Thousand Words [375], pg. 66. Aristotle speculated that one of the reasons that the base 10 numbering system is so prevalent is because our 10 fingers are so useful for counting [588] We now know that many different number systems have been developed throughout history which are based on counting other physical ....

....number of relevant works have continued in this traditional vein and have still managed to take into account some of the impact of artifacts on cognition. Examples of this sort of work include the foundational work by Larkin et al. on displaybased problem solving [374] and diagrammatic efficiency [115, 375], the work by Card et al. 94] on HCI modeling, and the work by Brooks [71 73, 76] on program writing and comprehension. In most of this sort of work, the external world is explicitly (but simply) treated as an external memory whose currently perceivable contents can appear in working memory. ....

[Article contains additional citation context not shown here]

Larkin, J. H., and Simon, H. A. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1), 1987, pp. 65--99.

No context found.

Larkin, J.H. & Simon, H. A. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11, 65-99. 1987.

No context found.

Jill H. Larkin and Herbert A. Simon. Why a diagram is (sometimes) worth ten thousand words? Cognitive Science, 11(1):65--100, January--March 1987.

No context found.

Larkin, J., and Simon, H.A. Why a Diagram is (Sometimes) Worth Ten Thousand Words. In Cognitive Science, Vol. 11, No. 1, pp. 6599, 1987. SKAFFA

No context found.

J.H. Larkin and A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11(1):65--99, 1987.

No context found.

Larkin JH, Simon HA. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science 1987;11(1):65-99.

No context found.

J.H. Larkin and H.A. Simon. Why a diagram is (sometimes) worth ten thousand words. Cognitive Science, 11, 1987.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC