[[_Anotacions]] tags:: #marc_teoric #Distributed_Cognition # Mangalaraj2014 # Distributed cognition in software design: an experimental investigation of the role of design patterns and collaboration ## Anotacions [[Mangalaraj2014]] "Hollan et al. (2000, p. 177) refer to distributed cognition as "a broader conception that includes phenomena that emerge in social interactions as well as interactions between people and structure in their environments."" ([George Mangalaraj et al 2014:2](zotero://open-pdf/library/items/F2AX5TB3?page=2)) _Utilitza Hollan per a definir la cognició distribuïda ([note on p.2](zotero://open-pdf/library/items/F2AX5TB3?page=2))_ "Socially distributed cognition arises from the dynamic exchange and processing of information between two or more members of a group engaged in a problem-solving task. Structurally distributed cognition, on the other hand, is an outcome of the processing of information interlaced between internal cognitive processes and external artifacts that contain information related to the task at hand." ([George Mangalaraj et al 2014:2](zotero://open-pdf/library/items/F2AX5TB3?page=2)) _Definició de Cognició distribuïda Socialment i Definició de Cognició distribuïda Estructuralment. ([note on p.2](zotero://open-pdf/library/items/F2AX5TB3?page=2))_ "engender. Thus, distributed cognition theory provides an ideal framework to empirically investigate alternative cognitive systems that software development practices" ([George Mangalaraj et al 2014:2](zotero://open-pdf/library/items/F2AX5TB3?page=2)) _El marc que necessitem. ([note on p.2](zotero://open-pdf/library/items/F2AX5TB3?page=2))_ "In light of the preceding discussion, our study makes a notable contribution to the literature by addressing the following questions: 1. How would a socially distributed cognitive system comprising a collaborating pair perform vis-à-vis individuals in a software design task? 2. How would a structurally enhanced cognitive system affect the performance of software designers? 3. When comparing the performance of a low performing software designer with that of a collaborating pair, would the availability of an external cognitive artifact to the former compensate for the absence of a social actor?" ([George Mangalaraj et al 2014:3](zotero://open-pdf/library/items/F2AX5TB3?page=3)) _Tres qüestions que es donen resposta en aquest estudi ([note on p.3](zotero://open-pdf/library/items/F2AX5TB3?page=3))_ "Distributed cognition theory calls into question the premise that internal processes alone give rise to cognition, and contends that the social context and the artifacts present in the environment result in a cognitive system that transcends individual cognition (Flor and Hutchins 1991). Cognitive processes may be distributed in three ways (Hollan et al. 2000). First, they may be socially distributed among actors engaged in a collaborative activity. Individuals in a group actively organize themselves and coordinate their efforts to accomplish tasks. Dynamic processing of information resulting from their rich interactions gives rise to a complex system with cognitive processes that are not only singularly different but also cannot be easily anticipated from any constituent member's internal processes. Second, a cognitive architecture with properties distinct from those of its components may emerge from the interactions between internal cognitive processes (within individuals) and cognitive artifacts or structures (e.g., external media containing taskrelated information) in the work environment. This is referred to as embodied (Hollan et al. 2000) or structural (Hansen and Lyytinen 2009) cognition. Third, there is a temporal aspect of distributed cognition that suggests that experience from past cognitive events can have a bearing on subsequent ones (e.g., Hansen and Lyytinen 2009; Hollan et al. 2000)." ([George Mangalaraj et al 2014:3](zotero://open-pdf/library/items/F2AX5TB3?page=3)) _3 idees: 1.- El grup millora l'acompliment de les tasques (no son processos cognitius independents). 2.- Analitzar les interaccins tan amb d'altres persones com amb artefactes cognitius o estructures. 3.- Els events cognitius passats poden tenir efecte en els futurs. ([note on p.3](zotero://open-pdf/library/items/F2AX5TB3?page=3))_ "our research examines the performance of four cognitive architectures resulting from the interplay between various levels of social distribution (pair versus individual) and structural distribution (with and without design patterns)." ([George Mangalaraj et al 2014:3](zotero://open-pdf/library/items/F2AX5TB3?page=3)) _Aquesta investigació estudia 4 estructures: Treball en parella o individual i Amb patrons de disseny i sense. ([note on p.3](zotero://open-pdf/library/items/F2AX5TB3?page=3))_ "Table 1. Conceptual Foundations of the Study Conceptual Foundation Description Relevance to Our Study Distributed A cognitive system extends beyond an individual's Four cognitive arrangements (presence/absence of cognition mind. Specifically, cognition may be distributed a partner versus availability/nonavailability of an (Socially across individuals, or between individuals and taskexternal artifact) were evaluated. Paired designing versus related artifacts in the environment. The human provided the socially distributed context while design structurally cognitive system may be extended by adding a patterns served as cognitive artifacts with the distributed partner (social distribution) or by adding an external potential to influence task outcomes. cognition) artifact (structural distribution). Distributed Cognition and the Software Design Task Structuring The problem space for a design task is typically Design patterns as external representational media the problem unstructured because of incomplete information (i.e., structurally distributed cognition) have the space and/or lack of clarity in requirements. Providing potential to facilitate opportunistic designs, evoke access to task-related information can enhance the cognitive processes relevant to the task at hand, chances of transforming an ambiguous design and fill informational gaps to reduce the inherent problem into a well-structured one, and facilitate its ambiguity of design problems. Thus, they can affect solution. both the quality of the solution and the time taken to solve the problem by focusing attention on a limited range of feasible alternatives. Likewise, socially distributed cognition in the event of pairing can augment the pool of available knowledge and bridge any gaps that might exist in the problem space. Pairing in Performance of pairs may be enhanced (hindered) Consistent with prior research in the pair software by process gains (losses). The comparative programming domain, our study compares the design performance of pairs and individuals is an enduring performance of pairs with that of individuals in the stream of research. context of software design." ([George Mangalaraj et al 2014:4](zotero://open-pdf/library/items/F2AX5TB3?page=4)) "The intent and the application context constitute declarative knowledge, whereas the guidelines on how to use a pattern in a given situation determine the procedural knowledge (Kohls and Scheiter 2008)." ([George Mangalaraj et al 2014:4](zotero://open-pdf/library/items/F2AX5TB3?page=4)) "Further, distributed cognition theory suggests that the conception of a pair should be expanded to embrace a richer cognitive system that springs forth from the interactions between members of a pair as well as from the interplay between the social actors and the environment in which they are situated (Flor and Hutchins 1991; Hollan et al. 2000; Rosen et al. 2009)." ([George Mangalaraj et al 2014:6](zotero://open-pdf/library/items/F2AX5TB3?page=6)) _Parella i entorn de treball. ([note on p.6](zotero://open-pdf/library/items/F2AX5TB3?page=6))_ "One of the aims of our study is to investigate if the findings of pair programming studies hold up to scrutiny when the task type is significantly different. Such an examination is absolutely necessary for us to establish empirical regularity of our findings with regard to pairs and individuals, which, in turn, is essential for developing a theoretical framework of paired software developmen" ([George Mangalaraj et al 2014:8](zotero://open-pdf/library/items/F2AX5TB3?page=8)) "Distributed Cognition in Software Design The following gaps in the literature are apparent from our earlier discussion: (1) There is little empirical research on the efficacy of pairs versus individuals on design-centric tasks. (2) Rigorous empirical work on design patterns is not only limited, but also the inconsistent findings in the extant literature offer inadequate guidance to academics and practitioners. In addition, these studies provide no insight into the viability of patterns in a real design context. (3) To the best of our knowledge, there exists no designcentric study that empirically examines the performance of cognitive systems resulting from the interactions among social actors and external artifacts in a problem solving environment." ([George Mangalaraj et al 2014:8](zotero://open-pdf/library/items/F2AX5TB3?page=8)) _L'objectiu principal d'aquest estudi és donar exemples empírics sobre com funciona la cognició distribuïda en el disseny de software, ja que no hi ha molts estudis realitzats al respecte. ([note on p.8](zotero://open-pdf/library/items/F2AX5TB3?page=8))_ _Definició de la teoria de la cognició distribuïda de Hutchins ([note on p.8](zotero://open-pdf/library/items/F2AX5TB3?page=8))_ "The theory of distributed cognition suggests that artifacts, whether they are outcomes of the design process or those that are part of the design environment, are external representational media that can interact with mental processes of designers to confer unique cognitive properties (Flor and Hutchins 1991)." ([George Mangalaraj et al 2014:8](zotero://open-pdf/library/items/F2AX5TB3?page=8)) "In our study, external cognitive artifacts in the form of design patterns containing information relevant to the design problem were made available to some pairs and individuals." ([George Mangalaraj et al 2014:8](zotero://open-pdf/library/items/F2AX5TB3?page=8)) _Aqui es veu la importància de dissenyar bé, tant l'experiència gamificada, com l'eina de treball (ellaboratori), ja que formen part del processos mental dels estudiants per conferir propietats cognitives uniques. ([note on p.9](zotero://open-pdf/library/items/F2AX5TB3?page=9))_ "lem into a well-structured one to facilitate its easy solution. Design patterns may be viewed as external representations that have the potential to constrain search, to structure the problem, and facilitate a quicker and better solution. According to Simon (1973), the crux of problem solving is to find appropriate information to transform an ill-defined prob-" ([George Mangalaraj et al 2014:9](zotero://open-pdf/library/items/F2AX5TB3?page=9)) "Batra's (2005) argument that design patterns not only enhance the search for correct operators, but also stimulate existing operators in novel ways. Based on these arguments, we predict H1: Quality of the design solution will be higher when design patterns are used during software design. H2: Time taken to complete a design task will be shorter when design patterns are used" ([George Mangalaraj et al 2014:10](zotero://open-pdf/library/items/F2AX5TB3?page=10)) _La importància dels patrons de disseny. ([note on p.10](zotero://open-pdf/library/items/F2AX5TB3?page=10))_ "inadequate communication and coordination" ([George Mangalaraj et al 2014:10](zotero://open-pdf/library/items/F2AX5TB3?page=10)) _principals problemes en processos cognitius distribuïts. ([note on p.10](zotero://open-pdf/library/items/F2AX5TB3?page=10))_ _segon ([note on p.10](zotero://open-pdf/library/items/F2AX5TB3?page=10))_ "Therefore, we hypothesize H3: Solution quality of a collaborating pair will be higher than that of the second-best member of a nominal pair in a software design task." ([George Mangalaraj et al 2014:10](zotero://open-pdf/library/items/F2AX5TB3?page=10)) _en el cas de la creació d'un videojoc, les FEP i DAM tenen tasques completament diferenciades. No hi hauria un membre bo i altre dolent (segon millor membre) ([note on p.10](zotero://open-pdf/library/items/F2AX5TB3?page=10))_ _El temps que es triga a resoldre un problema és més gran en la parella que de forma individual (tant el primer com el milllor segon membre) degut a la necessitat de resoldre els conflicte, testejar errors, la presentació, discussió, rebuig i acceptació d'idees, presa de decisió col·lectiva. ([note on p.10](zotero://open-pdf/library/items/F2AX5TB3?page=10))_ "Impact of Pair Designing on Task Completion Time As discussed earlier, a design task is unitary, judgmental, and lacks demonstrability. Small-group research has found that groups take longer than individuals while working on unitary tasks (Hare 1976), that is, tasks that cannot be further subdivided among group members (Steiner 1972). The additional time taken by groups has been attributed to error checking (Hare 1976); time spent in communicating and resolving conflicts among team members (Brodbeck and Greitemeyer 2000); presentation, discussion, rejection, and acceptance of a large number of possible answers (Klugman 1944); and time needed to reach a collective decision (Brodbeck and Greitemeyer 2000). Pairs working on a software design task that is not easily divisible must develop a single solution representing their collective decision. Further, the lack of demonstrability of software design solutions is likely to increase the time required to reconcile design alternatives. In a similar vein, Canfora et al. (2007) found that pairs took more time than individuals to complete a design task in the context of software maintenance." ([George Mangalaraj et al 2014:10](zotero://open-pdf/library/items/F2AX5TB3?page=10)) "Specifically, the availability of design patterns can evoke relevant knowledge structures and schemas that can lend structure to the problem and lead to a better solution vis- à-vis a simple-individual cognitive system (Kohls and Scheiter 2008)." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) "In both cases, the external artifact injects additional knowledge into the cognitive system." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) _Funció de ellaboratori ([note on p.11](zotero://open-pdf/library/items/F2AX5TB3?page=11))_ "Do individuals and pairs benefit equally from the availability of design patterns? If, however, there is a differential benefit to individuals, would the recipient of this benefit be the individual with the lowest skills" ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) _ellaboratori s'hauria de centrar més en les FEP. No perquè tinguin les habilitats més baixes, sinó perquè estan molt més allunyades del projecte. Haurien de tenir patrons molt concrets. ([note on p.11](zotero://open-pdf/library/items/F2AX5TB3?page=11))_ "A team of paired designers, by virtue of having two persons in the team, has access to additional knowledge, skill, and expertise of the second member of the team. Thus, the expertise encoded in design patterns may be somewhat redundant with that already available to the best individual and the pair, and represents more new information for the second-best member. We, therefore, expect the second-best member to benefit most by this new information and help him/her close the performance gap with the best member and the pair. Hence, we hypothesize H5a: When design patterns are available, the performance gap in solution quality between the secondbest member of a nominal pair and the best member of the nominal pair and the pair will be reduced in contrast to when design patterns are not used." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) "H5b: Solution quality of the second-best member of a nominal pair with design patterns will be higher than that of the collaborating pair without design patterns in a software design task." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) _Aquesta hipòtesi mostra la importància dels patrons de disseny ([note on p.11](zotero://open-pdf/library/items/F2AX5TB3?page=11))_ "H5c: Task completion time of the second-best member of a nominal pair with design patterns will be lower than that of the collaborating pairs without design patterns in a software design task." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) "H6: Task satisfaction will be higher when design patterns are used in a software design task." ([George Mangalaraj et al 2014:11](zotero://open-pdf/library/items/F2AX5TB3?page=11)) "H7: The average level of task satisfaction among collaborating pairs will be higher when compared with the average level of task satisfaction of nominal pairs" ([George Mangalaraj et al 2014:12](zotero://open-pdf/library/items/F2AX5TB3?page=12)) "four design patterns, namely, Adapter, Memento, Observer, and Strategy (see Gamma et al. for a description of these patterns)" ([George Mangalaraj et al 2014:13](zotero://open-pdf/library/items/F2AX5TB3?page=13)) _4 dissenys de patrons. Possible ampliació!!! ([note on p.13](zotero://open-pdf/library/items/F2AX5TB3?page=13))_ "Our study compared the performance of a socially distributed cognitive system with individual cognitive systems at two levels of performance, namely, best and second-best members of a nominal pair. The results show that, in terms of software quality, a socially distributed cognitive system outperforms a low-performing individual system, but operates at the level of a high-performing individual system. Further, our study demonstrates that a structurally enhanced cognitive system delivers superior performance with regard to both quality and time. In addition, a low-performing individual cognitive system augmented by structural cognition outperforms a socially distributed cognitive system in terms of time, while delivering comparable performance with regard to quality." ([George Mangalaraj et al 2014:20](zotero://open-pdf/library/items/F2AX5TB3?page=20)) _Conclusions de l'estudi. ([note on p.20](zotero://open-pdf/library/items/F2AX5TB3?page=20))_ "Our results suggest that the performance of the second-best individual augmented with design patterns, a form of codified knowledge, is comparable to that of a collaborating pair not using design patterns." ([George Mangalaraj et al 2014:21](zotero://open-pdf/library/items/F2AX5TB3?page=21)) _Afegir estructures al formulari de sol·licitud de joc. ([note on p.21](zotero://open-pdf/library/items/F2AX5TB3?page=21))_ "In this context, our finding that lowperforming individuals aided by design patterns produce designs of comparable quality in less time vis-à-vis collaborating pairs unassisted by patterns is particularly insightful." ([George Mangalaraj et al 2014:22](zotero://open-pdf/library/items/F2AX5TB3?page=22)) "Pattern use in a design task is a three-step process: selection, adaptation, and use. The designer selects one or more patterns appropriate for the design task at hand, adapts them suitably, and uses them in the design." ([George Mangalaraj et al 2014:22](zotero://open-pdf/library/items/F2AX5TB3?page=22)) _Ús de patrons. ([note on p.22](zotero://open-pdf/library/items/F2AX5TB3?page=22))_ "To this end, future studies may investigate the effects of factors such as team size, team diversity, difference in the levels of expertise between team members, and personality types of collaborators on software development group outcomes." ([George Mangalaraj et al 2014:23](zotero://open-pdf/library/items/F2AX5TB3?page=23)) _Amb aquests items es podrà analitzar com funciona els desensenvolupadors de ellaboratori. ([note on p.23](zotero://open-pdf/library/items/F2AX5TB3?page=23))_ "this research demonstrates that the insertion of task-related external artifacts can alter the cognitive performance of a system." ([George Mangalaraj et al 2014:23](zotero://open-pdf/library/items/F2AX5TB3?page=23)) "Further, the results suggest that increased structural distribution (i.e., second-best individual with design patterns) can adequately compensate for the absence of a collaborating partner. Future studies on software development may seek to affirm these findings while elucidating the nature of the distribution of cognition across social actors and external representational media." ([George Mangalaraj et al 2014:23](zotero://open-pdf/library/items/F2AX5TB3?page=23)) ## Conceptes Marc Teòric ### [[023. Aprenentatge Col·laboratiu]] ### [[Estudi de la Cognicio Distribuida]] ### [[Distributed Cognition]] ### [[Socially Distributed Cognition]] ### [[Embodied Cognition]] ### [[Structurally Distributed Cognition]] ### [[Culture and Cognition]] ### [[Design Patterns]] ## Conceptes Propis ### [[043. Aprenentatge Col·laboratiu]] ### [[Aprenentatge Col·laboratiu Alumnes-Professors]] #### [[Aprenentatge Col·laboratiu entre Alumnes]] ### [[Gamificació com a Infraestructura Frontera]] #### [[Presentació Aprenentatge Col·laboratiu]] #### [[Videojoc, Web o APP com a Objectes Fronteres]]