No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Automation and schema acquisition in learning elementary computer programming: Implications for the design of practice
Abstract
Two complementary processes may be distinguished in learning a complex cognitive skill such as computer programming. First, automation offers task-specific procedures that may directly control programming behavior, second, schema acquisition offers cognitive structures that provide analogies in new problem situations. The goal of this paper is to explore what the nature of these processes can teach us for a more effective design of practice. The authors argue that conventional training strategies in elementary programming provide little guidance to the learner and offer little opportunities for mindful abstraction, which results in suboptimal automation and schema acquisition. Practice is considered to be most beneficial to learning outcomes and transfer under strict conditions, in particular, a heavy emphasis on the use of worked examples during practice and the assignment of programming tasks that demand mindful abstraction from these examples.
... In addition, software development is complex and non-routine (Faegri et al., 2010), so developers have to exert greater efforts in programming. With high cognitive demands in computer programming (Van Merriënboer & Paas, 1990), using a variety of skills to perform additional tasks would probably provoke work stress that lead to job burnout. We then theorize: ...
... Automation encompasses knowledge compilation in that knowledge are organized in a manner that is ready to be applied in a form of task-specific procedures (Van Merriënboer & Paas, 1990). With knowledge compilation, working memory does not have to store and process a large volume of acquired knowledge, because knowledge can be retrieved from and processed by automated processes (Anderson, 1987;Van Merriënboer & Paas, 1990). ...
... Automation encompasses knowledge compilation in that knowledge are organized in a manner that is ready to be applied in a form of task-specific procedures (Van Merriënboer & Paas, 1990). With knowledge compilation, working memory does not have to store and process a large volume of acquired knowledge, because knowledge can be retrieved from and processed by automated processes (Anderson, 1987;Van Merriënboer & Paas, 1990). Accordingly, automation would reduce developers' cognitive demands and attenuates developers' burdens of skill building based upon knowledge acquisition. ...
Prior studies have claimed that, with evolving technologies, continuity processes in Development and Operations (DevOps) needs to be re-evaluated consistently. In this context, this study examines a relatively new continuity process -- continuous software security. With regulatory compliance and ubiquitous cyberattacks, it becomes increasingly important to integrate security into DevOps. Presently, DevOps developers assume the role of systems operators and software developers to facilitate Continuous Integration/Continuous Deployment (CI/CD). Assuming multiple roles involves role transitioning that requires developers to psychologically disengage from their current role and engage in another. Gradually, this may cause mental exhaustion. Therefore, implementing continuous security may add more complexity to DevOps, instigating more stress to software developers. Given this concern, we draw on Role Transition Theory and Cognitive Load Theory to examine whether integrating security into DevOps would aggravate developers’ job burnout, which would eventually undermine continuous security practices.
... Unfortunately, it is well known that people forget, meaning that the neural pathways that constitute memories decay or are displaced (Della Salla, 2010). Memory loss can be averted if the thought is embedded in schema and if used repeatedly (Sweller, 2016;Van Merrienboer & Pass, 1990). Hattie and Zierer (2019, p. 82) defined deliberate practice as conscious practice that was challenging, varied and regular and had a positive effect on learning (d = 0.49) due to the strengthening of long-term memory. ...
... Cognitive load theorists (e.g., Sweller, 2016;Van Merrienboer & Pass, 1990) provide a rationale for emphasising conceptual understanding in the first instance. First, if the subject matter is taught conceptually, it can be integrated into existing schema and thus be more easily remembered. ...
... In this sense, procedure is not the same as blindly following a set of steps towards a solution, but rather that the steps are well embedded in long-term memory (2019) recommend deliberate practice to develop fluency and, like cognitive load theorists, suggested that conscious, varied, spaced and regular practice fostered long-term memory retention. The ready recall of key facts and procedures is thought to reduce cognitive load (Van Merrienboer & Pass, 1990) and thereby enhance problem-solving. Rather than considering learning mathematics as a simple hierarchical progression, it is a progression through cycles of the form illustrated in Fig. 11. ...
An important topic of study in secondary mathematics is non-linear functions, including quadratic equations. In this study, findings from 25 Year 11 students indicated that difficulties with critical prerequisite concepts such as algebraic conventions impeded students’ success in understanding and working with quadratics. Analysis of student errors identified misconceptions associated with the null factor law, and the nature of quadratic equations. This paper proposes that these findings are a result of limited timeframes nominated for learning quadratic topics outlined in the enacted curriculum. The implication of this is that the enactment of the Australian Curriculum: Mathematics F-10 did not support the development of conceptual understanding or procedural fluency with key mathematical concepts for these students. Without purposeful attention to prerequisite knowledge, and suitable time allocated to develop understanding and fluency, students’ proficiency with topics such as quadratics is negatively influenced. A mastery approach to the hierarchically organised curriculum is supported by findings of this study.
... The authors also found that explicit story structure was beneficial for students to understand and recall stories effectively (Gordon & Braun, 1983). These studies support Van Merrienboer and Paas's (1990) conclusion that detailed procedural knowledge is implicit and not easy to verbalize in instructional talk. Those authors investigated visualized worked examples as an alternative way to communicate detailed procedural knowledge and identified that worked examples as concrete schema helped students map their new solutions in problem-solving and shortened the learning process for automation (Van Merrienboer & Paas, 1990). ...
... These studies support Van Merrienboer and Paas's (1990) conclusion that detailed procedural knowledge is implicit and not easy to verbalize in instructional talk. Those authors investigated visualized worked examples as an alternative way to communicate detailed procedural knowledge and identified that worked examples as concrete schema helped students map their new solutions in problem-solving and shortened the learning process for automation (Van Merrienboer & Paas, 1990). ...
People are increasingly aware that metacognition can help us solve problems more effectively. Scholars believe that students’ metacognition can be facilitated by promoting the interplay between the unlimited long work memory for holding schemas and limited work memory for processing ongoing activities. In this study, a workshop designed teacher mediation method between English language teachers and students is implemented that focuses on instructional talk supported by a visualized thinking framework in learning English grammar. Consequent to student’s automation from retrieving schemas to engage in the self-regulatory process of English grammar problem-solving, this section involves three English language teachers and students in their taught classes in a secondary school in China. The teachers were equipped with an explicit approach to teacher mediation through researcher-designed workshops. As the teacher mediation approach was implemented in English language classrooms, the researchers made meaning of their experiences through the dialogue and interaction between teachers and students by revealing the essence in the mediation process: dialogic assessment, cognitive scheme construction, automated schema retrieval, and metacognitive schema construction. Through teachers’ English pedagogical grammar practice in the three example cases, metacognitive teachers and learners are expected to be cultivated in an interdisciplinary view.
... Didactical support is supposed to help people learn, understand and remember the learning content. It aims to enhance transfer-of-training through a process of generalization and/or abstraction, from knowledge of the task at hand to a higher level of knowledge, for instance, of the general principles of a domain or general procedures for performing a class or even several classes of tasks (Van Merriënboer & Paas, 1990). Thus, the learner has the (conceptual) knowledge available to interpret what particular actions may be effective or not in what situations (Van Merriënboer, 1997). ...
... We suppose that the MS Flight Simulator group had a certain level of understanding and flight skill enabling them to choose effective courses of actions in some flight tasks. This counts particularly in more generic elements of flight tasks in which performance is more dependent on general knowledge of the flight domain or more general flight procedures (Van Merriënboer & Paas, 1990). This transfer of overall flight skills may be considered as far-transfer (Brown, 1989;Brown & Campione, 1981;Flavell, 1976). ...
Aim. The objective of this study was to collect evidence of transfer-of-training to professional performance provided by two stand-alone PC-based flight games.
Background. These realistic games, Falcon 4.0 (F-16 specific) and Microsoft Flight Simulator (civil aircraft), are designed for entertainment purposes, lacking any purposeful or explicit instructional support.
Method. This quasi-experimental study used three pre-existing groups of gamers (n = 37; Falcon 4.0 gamers, Microsoft Flight Simulator gamers and control group: gamers without flight game experience) that performed three typical F-16 flight tasks in a high-fidelity fixed-base flight simulator.
Results. The Falcon 4.0 gamers performed substantially better on almost all tasks compared to the control group, and to a lesser degree to Microsoft Flight Simulator gamers. The Falcon 4.0 group showed near- and far-transfer on almost all flight performance measures: the game had prepared them for the generic and specific military aspects of the test flight tasks. Performance of the Microsoft Flight Simulator gamers indicated only far-transfer, i.e., transfer of more generic flight skills from the game to the test flight tasks.
Conclusion. Both near- and far-transfer of job related competences may occur by playing realistic entertainment games.
... Learning OOP skills require fundamental cognitive processes, and the cognitive capacity plays a mediating factor through out a learning process. Therefore, many studies underlined the importance of WM in computer problem solving (Merriënboer and Paas, 1990 In general, cognitive load related studies condense on the extraneous and/or germane cognitive load rather than the intrinsic cognitive load. The primary reason would be their manipulability by the instructional settings, tools, and strategies. ...
... In this study, learners' unconscious automated skills possibly helped them to solve the experimental problems. This facilitated their cognitive performance by freeing up processing resources that might be devoted to various controlled processes (Merriënboer and Paas, 1990). However, the OOP course and the lecture followed a traditional way for teaching programming to novice learners. ...
Teaching object-oriented programming (OOP) is a difficult task, especially to the beginners. First-time learners also find it difficult to understand. Although there is a considerable amount of study on the cognitive dimension, a few study points out its physiological meaning. Moreover, it has been suggested that neuroscientific studies and methods can enable educational researchers gain an insight into brain and cognitive processes as well. Therefore, this experimental study explored the previously learned OOP skills in terms of cognitive load. By using the functional near-infrared spectroscopy (fNIRS) method, we measured the cognitive load of participants when executing OOP tasks. The average oxygenation changes in prefrontal cortex of the brain represented their total cognitive response to a set of OOP tasks. There were two research questions investigated by this study. The first research question explored whether the average oxygenation changed according to the participants' academic achievements and task completion status. The second research question was for identifying the instant changes in the oxygenations to find out which programming tasks were more contributing to the oxygenation. Later, we compared the findings with experts' judgments. We observed that the fNIRS system was an effective and promising technology for monitoring cognitive tasks both in classrooms and in experimental environments.
... It has been suggested that programmers who know the solution to a problem write their solution in a linear manner, while solving a new problem is done using means-ends analysis with the use of existing related schemas [15,51]. Over time and through practice, accumulation and evolution of schemas allow programmers to solve problems more fluently, and also to learn to solve new problems with more ease [51,58]. ...
... If we refer to Sweller (2006), the use of schemata mobilizes less cognitive resources, thus allowing the system to go further the limited capacity of the working memory, while limiting cognitive overload. In this perspective, acquisition of abilities or knowledge correspond to the elaboration of a new schemata, or the automation of an existing schemata (Anderson, 1983;Pollock et al., 2002;van Merriënboer & Paas, 1990). These statements underline the major importance of the organization of knowledge in the memory and its influence on reading comprehension depending of the type of text, narrative or expository. ...
In this study, we compared the effects of two media (Interactive Whiteboards and Paper) on both expository and narrative texts reading comprehension among 5th grade children of primary school. Two texts were constructed, according to the same controlled hierarchical structure. Comprehension was assessed by a multiple-choice questionnaire including three types of questions (surface, semantics, inferential). Results of the comprehension test revealed no difference between the two supports. Regardless of support, we found better performances for the narrative text, as well as an interaction between Text and Question factors, revealing that children had more difficulties to elaborate inferences when reading the expository text. These results are in line with previous findings underlying that texts with a similar structure, with a single-page presentation elicit similar performances on paper and electronic devices. They also provide interesting perspectives about the use and impact of Interactive Whiteboards during reading activities or lessons in classrooms.
... Aus dem Expertise-Umkehr-Prinzip lässt sich schlussfolgern, dass Lernumgebungen nicht für alle Lerner einheitlich gestaltet, sondern an den jeweiligen Expertisegrad angepasst sein sollten, damit Lernziele effektiv und effizient erreicht werden können (Kalyuga, 2007(Kalyuga, , 2014. Paas, 1990;van Merriënboer, Schuurman, de Croock, & Paas, 2002); auch als Fading Guidance Strategy bezeichnet (z. B. Renkl, Atkinson, & Große, 2004 (Sweller et al., 1998). ...
... Although this may seem counterintuitive, tests have demonstrated that studying complete examples facilitates learning more effectively than actually solving the equivalent problems [7]. Additionally, in many cases, a variation of worked examples, balanced with assignments, has been used and studied [8]. Students can be urged to complete the solution, which is only possible with the careful study of the partial example provided in the completion task. ...
... When studying the worked example, the learner has only to determine how the example goes from one step to the next-a very reduced search space which is a means-end search (i.e., they know the result and must only find a path to get to that one end). This instructional strategy reduces the amount of cognitive processing required from the learner (Sweller, 2011). ...
Background
Programming a computer is an increasingly valuable skill, but dropout and failure rates in introductory programming courses are regularly as high as 50%. Like many fields, programming requires students to learn complex problem-solving procedures from instructors who tend to have tacit knowledge about low-level procedures that they have automatized. The subgoal learning framework has been used in programming and other fields to breakdown procedural problem solving into smaller pieces that novices can grasp more easily, but it has only been used in short-term interventions. In this study, the subgoal learning framework was implemented throughout a semester-long introductory programming course to explore its longitudinal effects. Of 265 students in multiple sections of the course, half received subgoal-oriented instruction while the other half received typical instruction.
Results
Learning subgoals consistently improved performance on quizzes, which were formative and given within a week of learning a new procedure, but not on exams, which were summative. While exam performance was not statistically better, the subgoal group had lower variance in exam scores and fewer students dropped or failed the course than in the control group. To better understand the learning process, we examined students’ responses to open-ended questions that asked them to explain the problem-solving process. Furthermore, we explored characteristics of learners to determine how subgoal learning affected students at risk of dropout or failure.
Conclusions
Students in an introductory programming course performed better on initial assessments when they received instructions that used our intervention, subgoal labels. Though the students did not perform better than the control group on exams on average, they were less likely to get failing grades or to drop the course. Overall, subgoal labels seemed especially effective for students who might otherwise struggle to pass or complete the course.
... Since the 1950s, researchers wanted to determine which variables are more effective in predicting computer programming performances (i.e., Alspaugh, 1972;Bergersen & Gustafsson, 2011;Merrienboer & Paas, 1990;Rowan, 1957), mainly using the expert-novice paradigm (Lin, Wu, Hou, Lin, Yang, & Chang, 2016). Among those variables are gender, personality, intelligence, attitude towards computers, experience, level of comfort, background in mathematics, courses taken, the playing of games (Charlton & Birkett, 1999;Wilson, 2002), academic background and psychological factors (Bergin & Reilly, 2006), cognitive, behavioral and attitudinal factors (i.e., deRaadt et al., 2005), study habits (Willman, Linden, Kaila, Rajala, Laakso, & Salakoski, 2015) and cognitive skills (i.e., Bergersen & Gustafsson, 2011). ...
... Another is Constructionism http://constructivist.info/14/3/234.griffin to use completion problems; here students are given well-structured code and asked to complete missing sections of it (Deimel & Moffat 1982;Van Merriënboer & Paas 1990). Completion problems help bridge the gap between understanding code and writing code. ...
Context • Constructionism, Papert's pedagogy and learning theory, involves experiential learning where students engage in exploration, create things that are personally meaningful, and share them with others. This approach is quite motivating, evidenced by the popularity of maker spaces, hackathons, and educational technologies that promote creative computing. With constructionism, the learner's choice is important. This means that learning is often serendipitous. It also means that people often abandon their designs when obstacles arise. This is problematic in learning environments where coverage of key concepts is necessary, practice to develop skills is essential, and persistence with troubleshooting errors is required. > Problem • How can teachers and instructional designers complement a constructionist approach with one that addresses its limitations? I introduce de-construction-ism, a pedagogy and learning theory that emphasizes learning from taking things apart. It is inspired by reverse engineering, cognitive load theory, practice theory, and theories of learning from errors and negative knowledge. This approach is applicable to computer science, as described here, and other disciplines. > Method • I report on a design-based research experiment, where university students interacted with Python practice problems during weekly labs. The designs of the individual problems, and series of problem sets, were based on a model for de-construction developed by the author. > Results • The experiment serves as a successful proof of concept for implementing practice problems designed with a de-constructionist approach. Few technical difficulties arose, and the students enjoyed the learning experience. A few revisions to the model for de-construction were warranted. > Implications • I provide teachers and instructional designers with a simple, practical way to think about instruction , in terms of construction and de-construction. The de-constructionist approach involves ample, effective practice with taking apart well-built examples, some with intentional bugs. > Constructivist content • I propose a pedagogy that is opposite yet complementary to constructionism. > Key words • Constructionism, de-constructionism, bugs, intentional bugs.
... Time and effort can be saved through the reuse of tried and trusted methods. The same is also true in solving unseen problems, for which established algorithms can satisfy some sub-goals of the typical hierarchical goal structures characterizing computation problems (Van Merrienboer and Paas, 1990). Expert problem solvers can often draw upon hundreds of algorithms from their mental repository Automated generators of examples (Brooks, 1977) that enables them to come up with solutions almost mechanically. ...
Purpose
The purpose of this study is to investigate students’ decisions in example-based instruction within a novel self-regulated learning context. The novelty was the use of automated generators of worked examples and problem-solving exercises instead of a few handcrafted ones. According to the cognitive load theory, when students are in control of their learning, they demonstrate different preferences in selecting worked examples or problem solving exercises for maximizing their learning. An unlimited supply of examples and exercises, however, offers unprecedented degree of flexibility that should alter the decisions of students in scheduling the instruction.
Design/methodology/approach
ASolver, an online learning environment augmented with such generators for studying computer algorithms in an operating systems course, was developed as the experimental platform. Students’ decisions related to choosing worked examples or problem-solving exercises were logged and analyzed.
Findings
Results show that students had a tendency to attempt many exercises and examples, especially when performance measurement events were impending. Strong students had greater appetite for both exercises and examples than weak students, and they were found to be more adventurous and less bothered by scaffolding. On the other hand, weak students were found to be more timid or unmotivated. They need support in the form of procedural scaffolding to guide the learning.
Originality/value
This study was one of the first to introduce automated example generators for studying an operating systems course and investigate students’ behaviors in such learning environments.
... A key signifier of this transition is the student's understanding and use of "schemas" [17], which transfer knowledge from a declarative form to a more procedural form, reducing the cognitive load of programming [8]. As well as the explicit statement of schemas in teaching material, examination of worked examples and practice are key elements of developing programming skill [15] [7]. ...
A significant difficulty in teaching programming lies in the transition from novice to intermediate programmer, characterised by the assimilation and use of schemas of standard programming approaches. A significant factor assisting this transition is practice with tasks which develop this schema use. We describe the Summer of Code, a two-week activity for part-time, distance-learning students which gave them some additional programming practice. We analysed their submissions, forum postings, and results of a terminal survey. We found learners were keen to share and discuss their solutions and persevered with individual problems and the challenge overall. 93% respondents rated the activity 3 or better on a 5-point Likert scale (n=58). However, a quarter of participants, mainly those who described themselves as average or poor programmers , felt less confident in their abilities after the activity, though half of these students liked the activity overall. 54% of all participants said the greatest challenge was developing a general approach to the problems, such as selecting appropriate data structures. This is corroborated by forum comments, where students greatly appreciated "think aloud" presentations by faculty tackling the problems. These results strongly suggest that students would benefit from more open-ended practice, where they have to select and design their own solutions to a range of problems.
... Therefore, learning with WE is usually combined with different forms of cognitive activation and additional instructional support. One such form of activation is the presentation of incomplete WE (Stark, 1999;Van Merrienboer & Paas, 1990). Incomplete WE require the learner to complete some of the solution steps by themselves, that is, one or more solution steps have been omitted. ...
The aim of this study is to investigate whether worked examples are effective in fostering psychology students’ explanation competence. Explanation competence is a context-specific cognitive disposition that enables a person to construct a causal model of an observable psychological phenomenon by drawing on psychological theories. We set up a training intervention using worked examples to demonstrate how the observed psychological phenomenon (e.g., cognitive dissonance) is represented in an explanation. Instructional support was implemented using a fading procedure. We investigated the effects of worked examples on explanation competence using a sample of psychology students (n = 46) from a German university. In an experimental between-group pre- and post-tests design, the participants in the experimental condition received the training intervention, whereas the participants in a control condition did not receive the intervention. The experimental and control groups did not differ in their explanation competence before the training intervention. Participants in the experimental condition had a significant higher explanation competence after the training intervention than the participants in the control condition. Thus, our results indicate that worked examples effectively foster psychology students’ explanation competence. Considerations on how the results could be implemented in actual teaching settings are provided.
... Since the advent of computers and their use in different life contexts, as industries, residences and educational institutions, the way they can help individuals to improve their routine has been a concern [7]. Mainly in education area, this concern reflects the emergence of several educational theories that support the adoption of technological solutions [8], [9], [10], such as the 21th century theories [11]. This is a direct consequence of the creation of new technologies and their adoption in the classroom for traditional, blended or distance education [12]. ...
... An important aspect is that the part-complete examples are carefully designed as they have to contain enough "clues" in the code to guide the students in their completion. It is suggested that this method facilitates both automation, students having blueprints available for mapping to new problem situations, and schemata acquisition as they are forced to mindfully abstract these from the incomplete programs (Van Merrienboer & Paas, 1990). ...
The purpose of this paper is to present a new software tool that has been developed, the purpose of which is to help novices learn programming. The tool supports what is know as the “completion” method of learning to program. It begins by discussing the difficulties that students face when learning to program and the use of part-complete solutions as a teaching and learning method. CORT has been developed to support this use of part-complete solutions and its features are outlined. When used by a student, a part-complete solution to a given programming problem is displayed in one window and possible lines of code that can be used to complete the solution are displayed within another window. The lines can easily be moved between the windows in order to complete the solution, the solution then being transferred to the target programming environment for testing purposes. Preliminary feedback from students indicates that CORT is easy to use and perceived to be helping them in their learning of programming. Three different methods of using CORT have been identified and these will be the subject of future research.
... It is also an open question whether our account is suitable for explaining the transfer effect of retrieval practice, as it was shown that retrieval practice enhanced transfer to a new knowledge domain compared with study practice ( Butler, 2010). Although we did not investigate the issue of transfer and its relationship to automatization, there are studies that found causal relationship between the features of automatization during training and the transfer to similar problems in simple algebra transformation problems ( Cooper & Sweller, 1987) and computer programming ( Van Merrienboer & Paas, 1990). Altogether, the previous findings point out that retrieval practice leads to a diminishing involvement of attentional control in declarative retrieval and preserves long-term knowledge through fast and automatized processing of specific cue-target associations. ...
... It is also an open question whether our account is suitable for explaining the transfer effect of retrieval practice, as it was shown that retrieval practice enhanced transfer to a new knowledge domain compared with study practice (Butler, 2010). Although we did not investigate the issue of transfer and its relationship to automatization, there are studies that found causal relationship between the features of automatization during training and the transfer to similar problems in simple algebra transformation problems (Cooper & Sweller, 1987) and computer programming (Van Merrienboer & Paas, 1990). ...
The “testing effect” refers to the striking phenomenon that repeated retrieval practice is one of the most effective learning strategies, and certainly more advantageous for long-term learning, than additional restudying of the same information. How retrieval can boost the retention of memories is still without unanimous explanation. In 3 experiments, focusing on the reaction time (RT) of retrieval, we showed that RT of retrieval during retrieval practice followed a power function speed up that typically characterizes automaticity and skill learning. More important, it was found that the measure of goodness of fit to this power function was associated with long-term recall success. Here we suggest that the automatization of retrieval is an explanatory component of the testing effect. As a consequence, retrieval-based learning has the properties characteristic of skill learning: diminishing involvement of attentional processes, faster processing, resistance to interference effects, and lower forgetting rate.
... Therefore, learning with WE is usually combined with different forms of cognitive activation and additional instructional support. One such form of activation is the presentation of incomplete WE (Stark, 1999;Van Merrienboer & Paas, 1990). Incomplete WE require the learner to complete some of the solution steps by themselves, that is, one or more solution steps have been omitted. ...
Die Erklärung von Verhalten ist nicht nur ein definierendes Merkmal der Psychologie als Wissenschaft, sondern auch ein wesentliches Merkmal wissenschaftlicher Kompetenz im Fach Psychologie (Dietrich et al., 2015). Wissenschaftliche Erklärungen beantworten unter Rückgriff auf eine psychologische Theorie die Frage, warum ein bestimmter Sachverhalt, z.B. kognitive Dissonanz nach einer Entscheidung, aufgetreten ist, indem die Bedingungen angegeben werden, die zu dessen Auftreten geführt haben (Bierhoff & Petermann, 2014; Ohlsson, 2010). In diesem Sinn kann eine Erklärung als die Anwendung einer psychologischen Theorie auf eine Situation im Sinne einer Ausarbeitung dessen, was die Theorie implizit oder explizit über diese Situation aussagt, verstanden werden (Ohlsson, 1992; Wagner et al., 2014). Die inhaltliche Vermittlung psychologischer Theorien steht zu Beginn des Psychologiestudiums im Mittelpunkt (vgl. Birke et al., 2016). Problematisch hierbei ist, dass Wissen über Theorien nicht hinreichend für deren Anwendung auf eine konkrete Situation ist. Theorien enthalten kausale Aussagen über einen gegebenen Inhaltsbereich, allerdings keine Aussagen, wie sich die abstrakten Elemente einer Theorie in einer konkreten Situation darstellen können (Ohlsson, 1992). Daher liegt es nahe, die Anwendung psychologischer Theorien gesondert zu fördern. Vor diesem Hintergrund wurde eine integrierte Lernumgebung (Reinmann & Mandl, 2006) zur Förderung der Erklärungskompetenz Psychologiestudierender entworfen. Im ersten, instruktionsorientierten Teil der Lernumgebung wurde den Probanden deklaratives Wissen über Inhalte und Struktur wissenschaftlicher Erklärungen sowie Wissen über die Kriterien, was eine gute Erklärung auszeichnet (Bierhoff & Petermann, 2013; Westermann, 2000), vermittelt. Im zweiten, problemorientierten Teil wurde die Erstellung einer Erklärung anhand zweier beispielhafter Situationen und einer vorgegebenen psychologischen Theorie (Theorie der kognitiven Dissonanz bzw. Leistungsmotivationstheorie) dargestellt. In beiden Teilen wurden die Probanden durch Elaborationsprompts instruktional unterstützt. An der Studie nahmen 46 Psychologiestudierende teil (MAlter=21.5 [sd=2.63]; 12 männlich; NEG=30; NKG=16). In einem Pre-Post-Test-Design wurden deklaratives Wissen über Erklärungen (deklaratives Erklärungswissen) sowie die Kompetenz zur Erstellung einer eigenen Erklärung (angewandtes Erklärungswissen) erfasst, wobei Pre- und Posttest identisch waren. Deklaratives Erklärungswissen wurde mittels eines MC-Tests erfasst (max. 12 Punkte). Angewandtes Erklärungswissen wurde mittels der Vorgabe einer Situationsbeschreibung erhoben, die mithilfe vorgegebener Theorien erklärt werden musste. Kodiert wurde anschließend, ob die erstellte Erklärung die in der Lernumgebung vermittelten Merkmale einer wissenschaftlichen Erklärung erfüllt (max. 23 Punkte). Hinsichtlich des deklarativen Erklärungswissens ergab eine ANOVA mit Messwiederholung einen signifikanten Effekt des Messzeitpunktes (F[1,44]=68.97, p<.001, ηp2=.61), einen signifikanten Effekt der Lernumgebung (F[1,44]=11.81, p=.002, ηp2=.20) sowie einen signifikanten Interaktionseffekt (F[1,44]=51.17, p<.001, ηp2=.54.), der auf einen Wissenszuwachs der EG beim deklarativen Erklärungswissen zurückzuführen ist. Zum Vortest gab es keine Unterschied zwischen EG und KG im deklarativen Erklärungswissen (t[44]=-0.92, n.s, MEG=3.10 [sd=2.39], MKG=3.75 [sd=2.08]), während sich im Nachtest ein hochsignifikanter Effekt zeigte (t[44]=8.62, p<.001, d=2.67, MEG=8.13 [sd=1.61], MKG=4.12 [sd=1.26]). Ein ähnliches Bild ergab sich beim angewandten Erklärungswissen. Eine ANOVA mit Messwiederholung ergab einen signifikanten Effekt des Messzeitpunktes (F[1,44]=7.52, p=.009, ηp2=.15), einen signifikanten Effekt der Lernumgebung (F[1,44]=6.18, p=.017, ηp2=.12) und ebenso einen signifikanten Interaktionseffekt (F[1,44]=9.75, p=.003, ηp2=.18) der auf einen Wissenszuwachs in der EG zurückzuführen ist. Im Vortest gab es keinen signifikanten Unterschied der Gruppen (t[44]=0.42, n.s, MEG=4.17 [sd=2.02], MKG=3.88 [sd=2.58]), während sich Experimental und Kontrollgruppe im Nachtest signifikant unterschieden (t[44]=3.22, p=.002, d=0.99 MEG=7.07 [sd=3.56], MKG=3.69 [sd=3.07]). Die vorliegenden Befunde sprechen somit für die Lernwirksamkeit der Lernumgebung. Sowohl deklaratives als auch angewandtes Erklärungswissen konnte gefördert werden. Allerdings zeigt der im Sinne der Effektstärke zwar starke, aber absolut betrachtet kleine Effekt beim angewandten Erklärungswissen, dass zusätzliche instruktionale Maßnahmen in die Lernumgebung integriert werden müssen. Im Sinne des Cognitve-Apprenticeship-Ansatzes (Collins et al., 1989) könnte die Lernumgebung um eine coaching Phase ergänzt werden, in der die Erstellung eigener Erklärungen systematisch eingeübt wird. Zudem bleibt offen, inwieweit ein Transfer auf eine andere Situation bzw. andere psychologische Theorien erfolgt und wie stabil und nachhaltig die erzielten Lernerfolge sind.
... According to Winne and Hadwin's model (1998), any type of instructional support or scaffolding (i.e., provided by a human tutor or a computer) that is designed to guide or sustain students' learning is considered to be an aspect of the environment characteristics ( or task conditions) and may affect the way students engage in the learning tasks. Instructional support has been operationalized in different ways, either as worked-out examples and completion problems (Van Merriënboer & Paas, 1989;Van Merriënboer, Kirschner, & Kester, 2003;Renkl, 1997;Renkl & Atkinson, 2003) or as conceptual , procedural , and metacognitive scaffolds (e.g., Azevedo & Hadwin, 2005;Bannert & Reimann, 2012). ...
In the context of fast technological development and the widespread use of learning technologies in education, the need for students to regulate their own learning processes has become increasingly important (Bannert & Reimann, 2012; Winters, Greene, & Costich, 2008). Research in the last decade has revealed that students have to possess specific self-regulated learning (SRL) abilities in order to learn successfully in computer-based learning environments (CBLEs; Azevedo, 2009; Bannert, Hildebrand, & Mengelkamp, 2009). CBLEs contain different kinds of informational resources (e.g., texts, graphics, help tools) and therefore provide various opportunities for students to improve their learning. In addition, many CBLEs provide a high degree of learner control, allowing students to take full responsibility for the entire learning process. In other words, in learner-controlled CBLEs students can choose their own learning activities based on their personal needs and preferences (Williams, 1996).
19. Design and provide physical examples and Problems (the transition from worked - examples study to problem solving) in the light of Cognitive Skills Acquisition Theory, Cognitive Load Theory, Expertise reversal effect and Self-Explanation Effect Phenomenon.
Computer science pedagogy, especially in the higher education and vocational training context, has long-favored the hands-on practice provided by programming tasks due to the belief that this leads to better performance on hands-on tasks at work. This assumption, however, has not been experimentally tested against other modes of engagement such as worked example-based reflection. While theory suggests that example-based reflection could be better for conceptual learning, the concern is that the lack of practice will leave students unable to implement the learned concepts in practice, thus leaving them unprepared for work. In this article, therefore, we experimentally contrast programming practice with example-based reflection to observe their differential impact on conceptual learning and performance on a hands-on task in the context of a collaborative programming project. The industry paradigm of Mob Programming, adapted for use in an online and instructional context, is used to structure the collaboration. Keeping with the prevailing view held in pedagogy, we hypothesize that example-based reflection will lead to better conceptual learning but will be detrimental to hands-on task performance. Results support that reflection leads to conceptual learning. Additionally, however, reflection does not pose an impediment to hands-on task performance. We discuss possible explanations for this effect, thus providing an improved understanding of prior theory in this new computer science education context. We also discuss implications for the pedagogy of software engineering education, in light of this new evidence, that impacts student learning as well as work performance in the future.
The chapters in this book are based on selected peer reviewed research papers presented at the 11th biennial Networked Learning Conference (NLC) 2018 held in Zagreb and were chosen as exemplars of cutting edge research on networked learning. The chapters are organized into three main sections: 1) Aspects of mobility for Networked Learning in a global world, 2) Use and misuse of algorithms and learning analytics, 3) Understanding and empowering learners. The three main sections are flanked by chapters which introduce and reflect on Networked Learning as epistemic practice. The concluding chapter draws out perspectives from the chapters and discusses emerging issues.
The book focuses on the nature of learning and interactions as an important characteristic sought out by researchers and practitioners in this field.
This chapter explores cognitive load theory (CLT) in the context of networked learning (NL). CLT supports NL practitioners’ efforts to understand and eliminate barriers to learning in NL situations. The premise is that by recognising unnecessary cognitive load in NL, educators can improve learners’ abilities to acquire and develop schema and, in doing so, support learning in NL situations. The chapter is structured into three main sections. The first section provides the background to the exploration of CLT in the context of NL. It includes an overview of CLT and its development, an overview of NL and a definition of the problem. The second section explores common features of NL and identifies potential sources of cognitive load in NL situations. It is organised according to the key features of NL ‘architecture’: the learning environment, learning tasks and learner activity. The third section identifies a potential research agenda to guide further explorations of CLT in NL including: research into technical aspects of NL to improve the presentation of information and computer interfaces, research into the use of instructional design techniques sympathetic to CLT and specifically targeting NL and engagement tasks and research to understand learning to learn online in NL from a CLT perspective.
The need for methods, techniques and approaches that we can develop high-level thinking skills in
important activities increases day by day in order to achieve effective use of technology and change in
information and communication technologies. In particular, the diversity, complexity of technical
skills and to gain technical skills required to be learned in schools and through applications in industry
is important. Teaching the programming as a technical skill during instructional design process (ID)
and how effective and meaningful teaching can be taught is an important problem. Thus, instructional
design models have been developed for the solution of learning problems in systemic, systematic and
appropriate learning conditions and especially for the development of technical skills (van
Merriënboer (1997). The instructional design model (4C/ID) activity mentioned here can be used for
teaching the importance of instructional and technological stages by combining and supporting another
multimedia project design, development and evaluation model. This study presents technical skills
only by pointing to the future developers and designers of programming that an instructional design
approach can be used to develop other programming skills. In addition, through ten steps proposed for
complex learning (van Merriënboer & Kirschner 2007) and steps in achieving complex cognitive,
high-level, algorithm based limited coding, technical skills, it is to provide a new different approach
to program developers, instructors and designers by planning and discussing the design of the process
within a basic frame as to be in four stages (van Merriënboer & Kirschner 2007). The purpose of this
study is to adapt the principles of the model for teaching technical skills by using four-component
instructional design model (4C/ID) within software programming. In this study, theoretical framework
for teaching complex technical skills, learning theories and problem solving in programming are
given. The relationships between components of 4C/ID model presented for teaching programming
skills. At the end of study, the ID model components and their applications for future programming
skills were indicated.
Cambridge Core - Education, History, Theory - The Cambridge Handbook of Computing Education Research - edited by Sally A. Fincher
The Cambridge Handbook of Computing Education Research - edited by Sally A. Fincher February 2019
While many online resources teach basic web development, few are designed to help novices learn the CSS concepts and design patterns experts use to implement complex visual features. Professional webpages embed these design patterns and could serve as rich learning materials, but their stylesheets are complex and difficult for novices to understand. This paper presents Ply, a CSS inspection tool that helps novices use their visual intuition to make sense of professional webpages. We introduce a new visual relevance testing technique to identify properties that have visual effects on the page, which Ply uses to hide visually irrelevant code and surface unintuitive relationships between properties. In user studies, Ply helped novice developers replicate complex web features 50% faster than those using Chrome Developer Tools, and allowed novices to recognize and explain unfamiliar concepts. These results show that visual inspection tools can support learning from complex professional webpages, even for novice developers.
Students’ perception of the costs of engaging in learning has only recently been the focus of empirical study. Perceived costs include effort cost, opportunity cost, and psychological cost. This study focuses on the implications of psychological cost – the perceived negative psychological consequence of participating in a learning task –for learning behavior, performance, and the potentially greater prevalence of cost for women learning math. Research questions include: (1) Does perceived psychological cost of engaging in a calculus course predict undergraduates’ behavior in the learning management system (LMS)? (2) Does psychological cost of engaging in a calculus course predict students’ academic performance? (3) Which digital learning behaviors predict final exam score? (4) Do females perceive greater psychological cost than males? (5) Are there differences in course achievement by gender? And (6) Do male and females’ digital learning behaviors differ? Contrasting theory and prior findings, psychological cost did not predict learning behavior or course performance. Students’ use of policy documents and a tool to organize study sessions predicted final exam performance. Females perceived greater psychological costs than males when studying calculus, consistent with prior research. Female students also scored lower than males on the final exam. Results suggest that costs may differ by gender and may mediate gender differences in performance.
This work presents the design and architecture of an educational tool for learning to code. The CodeLab tool is based on skill practice and assessment and is targeted for non-STEM students to develop computational thinking. The tool is designed to provide a lab experience and environment based on exercises to practice through a conversational interface.
This chapter presents a descriptive model of the instructional processes that may be distinguished in computer-based learning environments for teaching elementary computer programming. First, the claim is made that instructional processes fundamentally differ for recurrent component skills involved in programming (i.e., skills that are performed in a highly similar way over various problem situations, such as using the programming environment or translating an algorithm into code) and for non-recurrent skills involved in programming (i e, skills that require a variable performance over problem situations, such as analyzing a programming problem or designing an algorithm). Second, the claim is made that instructional processes also fundamentally differ for (a) practice, (b) the presentation of information, and (c) feedback and reaction. For each of the resulting six categories of the model, design principles are presented that one should take into account in the design of computer-based learning environments. Based on these principles, a survey is made of some well-known computer-based learning environments for elementary programming; it is concluded that the instructional models which underlie these systems are often incomplete and overly simplified.
This chapter addresses the differential effects on learning outcomes of two instructional strategies in an introductory course on solving database query problems using a relational database. In both strategies a problem-solving method was explicitly presented to grade 10 high school students. One strategy involved a top-down approach in solving the problems, while in the other strategy problems were approached in a bottom-up fashion. Although the top-down approach is considered to be superior in solving problems in general, it is hypothesized that for novices this is only the case if they possess relatively much relevant knowledge and skills. For querying a relational database logical reasoning, knowledge of set theory and the ability to identify relevance and sufficiency of data are supposed to be important. It is hypothesized that for high-ability students (with respect to the knowledge and skills mentioned) the top-down approach will be superior to the bottom-up approach, but for low-ability students the bottom-up approach will yield better results. For low-ability students the data supports the hypothesis; for high-ability students however, presenting different problem-solving strategies did not result in significant differences in learning outcomes. After discussing these results, plans for further research will be presented.
This chapter is concerned with instructional models, that is, implemented process models of instruction that offer some articulation of the didactic expertise involved. Three basic problems in the implementation of instructional models are discussed. Possible solutions to those problems are illustrated by a description of CASCO, a computer-based learning environment for introductory computer programming. Our focus is on the application of fuzzy set theory and fuzzy logic in the implementation of so-called Fuzzy Logic Instructional Models (FLIM’s). To illustrate this approach, an in-depth discussion is provided of CASCO’s Problem Selection model.
Programming plans are an important component of programming expertise. This study examined whether rule based elaboration (RBE) of worked examples, an instructional technique that uses rules to elaborate on plans in worked examples, could be used to promote the acquisition of plans by novices. Subjects were 102 undergraduates enrolled in introductory programming courses. In order to evaluate the effectiveness of RBE, it was compared to the use of conventional problem solving and the use of worked examples. The findings appear to suggest that the use of RBE was more successful than either the use of worked examples or conventional problem solving in promoting the learning of plans.
The empirical evidence described in this book indicates that instructional designs and procedures that are cognitively optimal for less knowledgeable learners may not be optimal for more advanced learners. Instructional designers or instructors need to evaluate accurately the learner levels of expertise to design or select optimal instructional procedures and formats. Frequently, learners need to be assessed in real time during an instructional session in order to adjust the design of further instruction appropriately. Traditional testing procedures may not be suitable for this purpose. The following chapters describe a cognitive load approach to the development of rapid schema-based tests of learner expertise. The proposed methods of cognitive diagnosis will be based on contemporary knowledge of human cognitive architecture and will be further used as means of optimizing cognitive load in learner-tailored computer-based learning environments.
Teaching functional programming as a second programming paradigm is often difficult as students can have strong preconceptions about programming. When most of these preconceived ideas fail to be confirmed, functional programming may be seen as an unnecessarily difficult topic. A typical topic that causes such difficulties is the language of types employed by many modern functional languages. In this paper, we focus on addressing this difficulty through the use of step-by-step calculations of type expressions. The outcome of the study is an elaboration of a worked example format and a methodical approach for teaching types to beginner functional programmers.
This paper reports on the re-design of a computer programming class for students of mechanical engineering. The content was re-designed using Cognitive Load Theory; the delivery was redesigned using on-line technologies. Student learning was objectively assessed; it improved and the drop-out rate reduced. A previous paper reported on greatly improved student attitudes and instructor reviews. This paper reports on objective data: comparing student performance on identical final exams. Note is made of improved learning by non-traditional engineering students. This paper also reports on two additional teaching strategies that were deployed to improve learning. Finally, this work points to the next step in this evolving redesign.
In this article, we propose a new way to assess children's acquisition of debugging skills in a LOGO environment. The assessment procedure is based on an explicit and precise model (in the form of a computer simulation) of good debugging skills. The model has four stages: 1) evaluating the program's planned and actual outcomes to determine that debugging is necessary, 2) identifying the bug by using descriptions of the discrepancy between the planned and actual outcomes to propose potential bugs, 3) locating the bug by using clues about the structure of the program to narrow the search, and 4) correcting the bug and retesting the program. We describe model-based measurements of the LOGO debugging skills actually acquired by students in a “typical” LOGO graphics course. Nine seven- to nine-year-olds were given twenty-four hours of LOGO training over a three-week period. Students learned the editing and command generation skills prerequisite to debugging but were not able to interpret commands and use clues to identify, locate, and correct bugs. We conclude by discussing objectives for teaching the model's debugging skills directly.
This review discusses recent literature on computer programming. It focuses on psychological studies of programming and selected issues related to instruction in programming. The purpose is to inform computer educators about the nature of the cognitive processes involved in programming, and the potential benefits to be gained from learning programming. Concerns relevant to pre-college students of programming are emphasized. The literature reviewed includes empirical studies of programming, rational analyses of the programming task by computer scientists, and expository essays by educators and psychologists. The literature is presented to illuminate three major issues: 1) what are the cognitive demands of programming and what are the possible cognitive outcomes? 2) how does instruction influence learning? 3) who benefits, and in what ways, from instruction?
Analogical reasoning is a powerful mechanism for exploiting past experience in planning and problem solving. This chapter outlines a theory of analogical problem solving based on an extension to means-ends analysis. An analogical transformation process is developed to extract knowledge from past successful problem-solving situations that bear a strong similarity to the current problem. Then, the investigation focuses on exploiting and extending the analogical reasoning model to generate useful exemplary solutions to related problems from which more general plans can be induced and refined. Starting with a general analogical inference engine, problem-solving experience is, in essence, compiled incrementally into effective procedures that solve various classes of problems in an increasingly reliable and direct manner.
We present an analysis and simulation model of verbal protocols of two college students (SS and AD) and one 8-year-old child (JP) learning to program recursive functions. The model is formalized as a production system capable of acquiring new production rules based on problem-solving experience. The model and protocols suggest: (a) that problem solving by analogy to worked-out examples is frequent in initial attempts by novices to write recursive functions; (b) different representations of examples are used to guide problem solving by analogy; and (c) performance on later problems reflects the particular representations used in problem solving by analogy on earlier problems. The protocols and simulations suggest that learning is facilitated by using abstract representations of the structure of recursion examples to guide initial coding attempts.
The avowed purpose of most programming courses is to teach problem solving as well as to teach programming. How this purpose can be achieved is the topic of the first part of this paper. Drawing on recent research examining the cognitive demands of programming and the behavior of programmers, a chain of cognitive accomplishments culminating in increased problem solving skill is described. Whether and how this purpose is being achieved is the topic of the second part of this paper. Results of an integrated set of studies of middle school programming classes show how students start on this chain and how far they progress in introductory courses. These studies illustrate how "exemplary" instruction moves students much further along the chain than does "typical" instruction. Examination of the background characteristics of students in these courses reveals that general ability and out-of-school computer access are more associated with progress along the chain in typical classes than in exemplary classes.
Derivational analogy, a method of solving problems based on the transfer of past experience to new problem situations, is discussed in the context of other general approaches to problem solving. The experience transfer process consists of recreating lines of reasoning, including decision sequences and accompanying justifications, that proved effective in solving particular problems requiring similar initial analysis. The role of derivational analogy in case-based reasoning and in automated expertise acquisition is discussed. (Author)
Means–ends analysis is a mechanism that is assumed to operate when people solve transformation problems. Its use is affected by the extent to which the goal is clearly specified to the problem solver as a problem state and by the extent to which learning occurs during a problem-solving episode. Five maze-tracing experiments were conducted with 116 undergraduates in which the finish point of the maze could be presented either as a specific location or in more general terms. The latter prevented the use of conventional means–ends analysis. Results indicate that on the particular maze configuration used, the nonspecific goal resulted in fewer errors and more rapid learning of the structure of the problem. Under conditions that facilitated the use of means–ends analyses, knowledge of the goal location rendered the problem insoluble. General results were replicated with the use of numerical problems. Implications for the generality of means–ends analysis as a problem-solving mechanism are discussed. (11 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Tested the 2-process theory of detection, search, and attention presented by the current authors (1977) in a series of experiments. The studies (a) demonstrate the qualitative difference between 2 modes of information processing: automatic detection and controlled search; (b) trace the course of the learning of automatic detection, of categories, and of automatic-attention responses; and (c) show the dependence of automatic detection on attending responses and demonstrate how such responses interrupt controlled processing and interfere with the focusing of attention. The learning of categories is shown to improve controlled search performance. A general framework for human information processing is proposed. The framework emphasizes the roles of automatic and controlled processing. The theory is compared to and contrasted with extant models of search and attention. (31/2 p ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Investigated the distinctive strategies employed by expert and novice problem solvers (forward-chaining and means–ends, respectively) in 7 experiments using 14 mathematics graduates and 162 9–12 yr olds. Exp I studied the course of development of expertise using a subset of kinematics problems. Ss demonstrated the switch from a means–ends to a forward-chaining strategy. This was associated with the conventional concomitants of expertise such as a decrease in the number of moves required for solution. Ss appeared to categorize problems according to the order in which equations would be required. Exps II and III tested the hypothesis that the means–ends strategies used by novices retarded the acquisition of appropriate schemata. The use of nonspecific rather than specific goals was found to enhance the acquisition of expertise, the number of moves required for solution, and the number of equations written without substitutions. Exps IV and V, using geometry problems, duplicated the enhanced rate of strategy alteration found with reduced goal specificity. Results of Exps VI and VII indicated that reduced goal specificity also enhanced the rate at which problem solvers induced appropriate problem categories. It is concluded that in circumstances in which the primary reason for presenting problems is to assist problem solvers in acquiring knowledge concerning problem structure, the use of conventional problems solved by means–ends analysis may not be maximally efficient. (20 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Describes 3 experiments in which a total of 124 undergraduates were taught the concept of a binomial probability using methods that emphasized (a) calculating with the formula, or (b) the meanings of the variables in the formula. Learning outcomes were tested using 4 kinds of items, including calculation for new problems and questions about general properties of the formula. Large interactions in transfer performance were obtained in 3 cases, indicating that the 2 methods produced structurally different learning outcomes. Results are interpreted in relation to a hypothesis that cognitive structures can vary in the connectedness that components have with each other (internal connectedness) and with other elements of the S's knowledge (external connectedness). (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Hypothesized that schema acquisition would precede rule automation and that it would have a strong effect on problems similar to initial acquisition problems. We further hypothesized that rule automation would have its primary effect on transfer and that the use of worked examples could facilitate both transfer and performance on similar problems. Experiments 1 and 2 contained simple algebra transformation problems involving the changing of the subject of an equation. The results indicated that subjects whose training included a heavy emphasis on worked examples or an extended acquisition period were better able to solve both similar and transfer problems than were those subjects trained with conventional problems. In Experiment 3, the use of verbal protocols gave some support to the hypotheses. Experiment 4, using algebra word problems, yielded data supporting the hypotheses. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Evidence is accumulating that the means–ends problem-solving strategies used conventionally by novice problem solvers are relatively ineffective as vehicles for the acquisition of schemata characteristic of experts. It is suggested that a means–ends strategy places a heavy load on cognitive processing capacity and that this load retards knowledge acquisition. A series of 3 experiments using trigonometry problems and a total of 20 10th-grade and 42 9th-grade students as Ss was carried out. The problem goal was modified with the intention of disrupting the strategy used by novices. It was hypothesized that the development of adequate cognitive representations of the sine, cosine, and tangent ratios would be enhanced as a consequence. Results indicate that preventing novice problem solvers from using means–ends analysis resulted in fewer mathematical errors both during acquisition and on subsequent problems, including transfer problems. This provided some evidence for the contention that a means–ends strategy places a heavy load on cognitive processing capacity, which retards knowledge acquisition. (13 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
A 2-process theory of human information processing is proposed and applied to detection, search, and attention phenomena. Automatic processing is activation of a learned sequence of elements in long-term memory that is initiated by appropriate inputs and then proceeds automatically--without S control, without stressing the capacity limitations of the system, and without necessarily demanding attention. Controlled processing is a temporary activation of a sequence of elements that can be set up quickly and easily but requires attention, is capacity-limited (usually serial in nature), and is controlled by the S. A series of studies, with approximately 8 Ss, using both reaction time and accuracy measures is presented, which traces these concepts in the form of automatic detection and controlled search through the areas of detection, search, and attention. Results in these areas are shown to arise from common mechanisms. Automatic detection is shown to develop following consistent mapping of stimuli to responses over trials. Controlled search was utilized in varied-mapping paradigms, and in the present studies, it took the form of serial, terminating search. (60 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Cognitive skills are encoded by a set of productions, which is organized according to a hierarchical goal structure. People solve problems in new domains by applying weak problem-solving procedures to declarative knowledge they have about this domain. From these initial problem solutions, production rules are compiled that are specific to that domain and that use of the knowledge. Numerous experimental results may be predicted from this conception of skill organization and skill acquisition. These include predictions about transfer among skills, differential improvement on problem types, effects of working memory limitations, and applications to instruction. The theory implies that all varieties of skill acquisition, including those typically regarded as inductive, conform to this characterization. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Proposes a framework for skill acquisition that includes 2 major stages in the development of a cognitive skill: (1) a declarative stage in which facts about the skill domain are interpreted and (2) a procedural stage in which the domain knowledge is directly embodied in procedures for performing the skill. This general framework has been instantiated in the ACT system in which facts are encoded in a propositional network and procedures are encoded as productions. Knowledge compilation is the process by which the skill transits from the declarative stage to the procedural stage. It consists of the subprocesses of composition, which collapses sequences of productions into single productions, and proceduralization, which embeds factual knowledge into productions. Once proceduralized, further learning processes operate on the skill to make the productions more selective in their range of applications. These processes include generalization, discrimination, and strengthening of productions. Comparisons are made to similar concepts from previous learning theories. How these learning mechanisms apply to produce the power law speedup in processing time with practice is discussed. (62 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Transformation problems are usually solved by a means–ends strategy that involves reducing differences between a current problem state and a goal or subgoal. However, when an easily learned rule governing the pattern of moves is available, it should be possible to induce Ss to switch to a history-cued strategy involving the extrapolation of a learned sequence of moves, as occurs during serial-pattern learning. This may be accomplished in appropriate problems by presenting subgoals that should be attained during problem solving. In the present study, 2 experiments with 128 9th and 10th graders investigated the effects of subgoal density and location on serial-pattern learning during problem solving. It was found that when subgoal location was appropriate to the serial pattern, Ss were more likely to use a history-cued strategy resulting in enhanced rule induction and transfer with increased subgoal density. When subgoal location was not appropriate to the serial pattern, Ss tended to continue using a means–ends strategy resulting in considerably reduced rule induction and transfer. (16 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
This article offers an examination of instructional strategies and tactics for the design of introductory computer programming courses in high school. We distinguish the Expert, Spiral and Reading approach as groups of instructional strategies that mainly differ in their general design plan to control students' processing load. In order, they emphasize topdown program design, incremental learning, and program modification and amplification. In contrast, tactics are specific design plans that prescribe methods to reach desired learning outcomes under given circumstances. Based on ACT* (Anderson, 1983) and relevant research, we distinguish between declarative and procedural instruction and present six tactics which can be used both to design courses and to evaluate strategies. Three tactics for declarative instruction involve concrete computer models, programming plans and design diagrams; three tactics for procedural instruction involve worked-out examples, practice of basic cognitive skills and task variation. In our evaluation of groups of instructional strategies, the Reading approach has been found to be superior to the Expert and Spiral approaches.The authors wish to express their gratitude to Sanne Dijkstra, Otto Jelsma and Georg Rakers for their helpful comments on a draft of this article. Correspondence concerning this article should be addressed to Jeroen J. G. van Merrienboer.
This paper presents a detailed description of the ADAPT (Apply Delayed Automatization for Positive Transfer) design model. ADAPT is based upon production system models of learning and provides guidelines for developing instructional systems that offer transfer of leamed skills. The model suggests that transfer of training can be attributed to procedure overlap between the original training task and the transfer task, as well as to analogy between new problem solving situations and acquired cognitive schemata. More specifically, the role of schemata in transfer is thought to increase as the transfer task becomes more different from the original training task. Several instructional tactics are suggested to optimize transfer of training. Declarative tactics pertain to the instructional design for acquiring knowledge which is relevant to performance of the skill; such tactics include demonstrating the skill, verbal instruction, the encouragement to paraphrase particular pieces of information, the application of advance organizers and mnemonic systems, and the presentation of concrete models and examples. Procedural tactics refer to the instructional design for acquiring the skill, that is, to the design of practice; such tactics include the encouragement to imitate the skill, the application of variability of practice and contextual interference, and the presentation of annotated examples. The relevance of ADAPT is evaluated and implications for future research are presented.
Similarity and analogy are fundamental in human cognition. They are crucial for recognition and classification, and have been associated with scientific discovery and creativity. Successful learning is generally less dependent on the memorization of isolated facts and abstract rules than it is on the ability to identify relevant bodies of knowledge already stored as the starting point for new learning. Similarity and analogy play an important role in this process - a role that in recent years has received much attention from cognitive scientists. Any adequate understanding of similarity and analogy requires the integration of theory and data from diverse domains. This interdisciplinary volume explores current developments in research and theory from psychological, computational, and educational perspectives, and considers their implications for learning and instruction. Well-known cognitive scientists examine the psychological processes involved in reasoning by similarity and analogy, the computational problems encountered in simulating analogical processing in problem solving, and the conditions promoting the application of analogical reasoning in everyday situations.
There are two broad processes that people can use when attempting to solve a problem. The first of these is a means-ends strategy in which attempts are made to reduce differences between a given problem state and a goal or subgoal. Moves are generated by the goal or subgoals. The second is a history-cued process in which people use previous moves to generate subsequent moves. It is suggested that a means-ends strategy tends to reduce transfer effects. A history-cued strategy may facilitate rule induction, which in turn may be an important contributing factor to transfer. A series of four experiments using hybrid problems that are soluble either by rule induction or by means-ends analysis supported the above suggestion. Two additional experiments indicated that with respect to the "insoluble problem effect," the use of history-cued strategy was, of itself, insufficient to induce transfer effects. In order for transfer to occur, the structure of the problems and the manner in which they were presented had to be such as to ensure that problem solvers perceived a close relation between problems.
Investigations of the impact of programming instruction on cognitive skills have yielded occasional positive and many negative findings. To interpret the mixed results, we describe two distinct mechanisms of transfer–“low road” transfer, resulting from extensive practice and automatization, and “high road” transfer, resulting from mindful generalization. High road transfer seems implicated where positive impacts of programming have been found; insufficient practice and little provocation of mindful abstraction are characteristic of investigations not demonstrating transfer. Our discussion affirms that programming instruction can improve cognitive skills under the right conditions, but cautions that implementing such conditions on a wide scale may be difficult and that programming instruction must compete with other means of improving cognitive skills.
A screening test was given to three classes of high school students, who were just completing introductory semester-long courses in Pascal. These tests were graded, and subsequently thirty-five students were given detailed clinical interviews. These interviews showed that errors were made with essentially every Pascal construct. Over half the students were classified as having major difficulties—fewer than 10 percent had no difficulties. The errors noted are discussed in detail in this article. A major finding is that the students attribute to the computer the reasoning power of an average person. The article also speculates about how difficult it might be to remediate the errors found, and concludes with an outline of future work.
This article reviews research literature in the area of cognitive psychology which is relevant to training for transfer of skills. Particular attention is given to the role of models, schema or principles in transfer Research throwing light on the way in which examples facilitate schema acquisition is summarized, and the application of schema and principles to procedural learning is outlined Finally, recent studies demonstrating the importance of schema acquisition as a component of exploration‐based approaches to training helps link cognitive theory with traditional discovery learning procedures. This article was prepared as part of a research programme funded by the Australian Council For Employment and Training The ideas expressed, however, are those of the authors alone
Reports an error in "The acquisition of task-specific productions and modification of declarative representations in spatial-precuing tasks" by Robert W. Proctor and T. Gilmour Reeve (Journal of Experimental Psychology: General, 1988[Jun], Vol 117[2], 182-196). The spanner heads were inadvertently omitted from Tables 2 and 3 and Tables 6 and 7. The four tables are reprinted in the erratum. (The following abstract of the original article appeared in record 1988-28565-001.) Three experiments explored the role of the stimulus–response translation stage of human information processing in relating stimuli to assigned responses. The primary findings are that (a) task-specific productions develop with practice that relate stimuli directly to specific fingers, and (b) declarative knowledge is acquired that can lead to the use of modified representations when translation subsequently is required in a transfer session. The results are consistent with the view that the mediating role of the translation stage decreases with practice and is inconsistent with the view that the role does not change or diminish. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Effective problem solving, sound decision making, insightful invention—do such aspects of good thinking depend more on deep expertise in a specialty than on reflective awareness and general strategies? Over the past thirty years, considerable research and controversy have surrounded this issue. An historical sketch of the arguments for the strong specialist position and the strong generalist position suggests that each camp, in its own way, has oversimplified the interaction between general strategic knowledge and specialized domain knowledge. We suggest a synthesis: General and specialized knowledge function in close partnership. We explore the nature of this partnership and consider its implications for educational practice.
Typically, when a programming language is taught, the syntax and the semantics of the language are emphasized. In contrast, we report here on an organization of information for teaching LISP which puts primary emphasis on the structure of and relationships between: a problem, a program, and, an intermediate abstraction, a plan. This organization is based on an analysis of the underlying structure of ostensibly different problems and their program solutions. We present qualitative observations on the use of this organization gleaned from actual classroom teaching. Finally, we attempt to generalize these notions to other problem domains and to other programming languages.
We will describe the results of some initial exploratory studies using an experimental technique which seeks to tap into the specific tacit programming knowledge underlying variables. We begin by highlighting our theory of programming knowledge underlying simple looping programs which we characterize in terms of stereotypic plans. In the experimental studies, we give programmers (novices and nonnovices) a program in which important lines of code have been replaced by blank lines. Their task is to fill in the blank lines with appropriate lines of code. The results of the studies indicated that experienced programmers filled in the blanks correctly and in a plan-like fashion. We conclude by discussing some implications of this approach for teaching programming, building computer-based programming tutors, and developing cognitively based measures of program complexity.
An analysis of the process of analogical thinking predicts that analogies will be noticed on the basis of semantic retrieval cues and that the induction of a general schema from concrete analogs will facilitate analogical transfer. These predictions were tested in experiments in which subjects first read one or more stories illustrating problems and their solutions and then attempted to solve a disparate but analogous transfer problem. The studies in Part I attempted to foster the abstraction of a problem schema from a single story analog by means of summarization instructions, a verbal statement of the underlying principle, or a diagrammatic representation of it. None of these devices achieved a notable degree of sucess. In contrast, the experiments in Part II demonstrated that if two prior analogs were given, subjects often derived a problem schema as an incidental product of describing the similarities of the analogs. The quality of the induced schema was highly predictive of subsequent transfer performance. Furthermore, the verbal statements and diagrams that had failed to facilitate transfer from one analog proved highly beneficial when paired with two. The function of examples in learning was discussed in light of the present study.
This study examined high school students' knowledge about constructs in the BASIC programming language. A screening test was administered to ninety-six students, fifty-six of whom were interviewed. Students were asked to trace simple programs and predict their output. Errors in virtually all BASIC constructs we examined were observed, with many of the misconceptions arising from the application of knowledge and reasoning from informal domains to programming. It is argued that a lack of knowledge of basic features of programming language will prevent students from developing the higher-level cognitive skills that much programming instruction is intended to foster.
Cognitive skills are encoded by a set of productions, which are organized according to a hierarchical goal structure. People solve problems in new domains by applying weak problem-solving procedures to declarative knowledge they have about this domain. From these initial problem solutions, production rules are compiled that are specific to that domain and that use of the knowledge. Numerous experimental results may be predicted from this conception of skill organization and skill acquisition. These include predictions about transfer among skills, differential improvement on problem types, effects of working memory limitations, and applications to instruction. The theory implies that all varieties of skill acquisition, including those typically regarded as inductive, conform to this characterization.
Excerpts available on Google Books (see link below). For more information, go to publisher's website : http://www.routledge.com/books/details/9780805822335/
Four experiments study the errors students make using LISP functions. The first two experiments show that frequency of errors is increased by increasing the complexity of irrelevant aspects of the problem. The experiments also show that the distribution of errors is largely random and that subjects' errors seem to result from slips rather than from misconceptions. Experiment 3 shows that subjects' errors tend to involve loss of parentheses in answers when the resulting errors are well-formed LISP expressions. Experiment 4 asks subjects, who knew no LISP, to judge the reasonableness of the answers to various LISP function calls. Subjects could detect many errors on the basis of general criteria of what a reasonable answer should look like. On the basis of these four experiments, we conclude that errors occur when there is a loss of information in the working memory representation of the problem and when the resulting answer still looks reasonable.
Two experiments studied the acquisition and transfer of text-editing skill. The first experiment, originally reported in Singley and Anderson (1985) but reanalyzed in greater detail here, found nearly total transfer between two similar line editors and partial transfer from the line editors to a screen editor. Analyses of the keystroke data revealed that the majority of the improvement during both learning and transfer was concentrated in the planning components of the skill. The second experiment found little evidence for negative transfer between a pair of screen editors designed for maximal interference using a classic interference paradigm. The few instances of negative transfer observed were better characterized as the positive transfer of nonoptimal methods rather than instances of true procedural interference. These results support an identical elements model of transfer based on a production system representation of cognitive skill. The relative magnitudes of transfer observed were consistent with detailed measures of production system overlap. In addition, localized transfer sites were hypothesized and identified through a series of microanalyses. Finally, specific transfer predictions based on the differential practice of general and specific components were tested and confirmed.
Research on intelligent tutoring seeks to develop systems for automating education and to explore epistemological issues concerning the nature of the knowledge presented by tutors and how that knowledge can be learned. The ACT theory of skill acquisition and its successor (PUPS--PenUltimate Production System) provides production-system models of the acquisition of skills such as LISP programming, geometry theorem-proving, and solving of algebraic equations. The three major sections of the document discuss; (l) the cognitive theory that serves as the basis for the tutoring endeavors, (2) the model-tracing methodology and how it derives from the cognitive theory, and (3) issues that arise in implementing the methodology. Knowledge begins in declarative form and is used by analogical processes to solve specific problems. Domain-specific productions are compiled from the traces of these problem solutions. The model-tracing methodology has been developed as a means of displaying this cognitive theory in intelligent tutoring. Implementation of the methodology involves developing a student model, a pedagogical module, and an interface. Work on tutoring and on skill acquisition have proven to be symbiotic -- each has furthered the other's development. (Author/MNS)
The knowledge required to solve algebra manipulation problems and procedures designed to hasten knowledge acquisition were studied in a series of five experiments. It was hypothesized that, as occurs in other domains, algebra problem-solving skill requires a large number of schemas and that schema acquisition is retarded by conventional problem-solving search techniques. Experiment 1, using Year 9, Year 11, and university mathematics students, found that the more experienced students had a better cognitive representation of algebraic equations than less experienced students as measured by their ability to (a) recall equations, and (b) distinguish between perceptually similar equations on the basis of solution mode. Experiments 2 through 5 studied the use of worked examples as a means of facilitating the acquisition of knowledge needed for effective problem solving. It was found that not only did worked examples, as expected, require considerably less time to process than conventional problems, but that subsequent problems similar to the initial ones also were solved more rapidly. Furthermore, decreased solution time was accompanied by a decrease in the number of mathematical errors. Both of these findings were specific to problems identical in structure to the initial ones. It was concluded that for novice problem solvers, general algebra rules are reflected in only a limited number of schemas. Abstraction of general rules from schemas may occur only with considerable practice and exposure to a wider range of schemas.
This article is a brief introduction to some of the issues that teachers of programming may find helpful. It starts by presenting a fairly idiosyncratic view of teaching programming which makes use of mechanistic analogies and points out some of the pitfalls. The article goes on to examine certain errors based on the misapplication of analogies as well as certain interaction errors. The main emphasis is on the notional machine both at the general level of understanding (and misunderstanding) the relationship of the terminal to the computer as such, as well as at the more specific level of understanding assignment. Notation and mistakes that poorly-designed languages can induce novices to commit are discussed.
Abstract Since the publication of Seymour Papert's ‘Mindstorms’ in 1980 there has been a steady flow of articles reporting on research which has attempted, in some sense, to test the claims made for programming as an educationally beneficial activity. The overall picture that these articles paint is not a clear one, but they leave a residual impression that the claims are, as yet, unsubstantiated. This paper argues that we are not in need of further controlled experimental studies so much as a clarification of our understanding of the novice programmer's learning task. It reviews a number of studies which have attempted to model the cognitive demands of programming through detailed observation rather than through abstract analysis. It emphasizes the distinction between programming as it is experienced in introductory classes and programming as an idealized cognitive activity.
Making a good programming environment for beginning programmers is an enterprise which can exploit the strong connections between machine learning and human learning. Applying what we know about teaching and learning to improve the programming environment can result in a system which allows beginners to more readily acquire programming skills.
Surprisingly, a universally accepted principle of good teaching and good learning has not been taken seriously enough in designing programming environments-learning by example. A good teacher presents examples of how to solve problems, and points out what is important about the examples. The student generalizes from the examples to learn principles and techniques. This paper describes a programming environment called Tinker, in which a beginning programmer presents examples to the machine, distinguishing accidental and essential aspects of the examples. The programmer demonstrates how to handle the specific examples, and the machine formulates a procedure for handling the general case. Because people are much better at thinking about concrete examples than they are at thinking about abstractions, and because examples provide immediate feedback, Tinker is a more congenial environment for a beginner than conventional programming systems.
While only in the past ten years have large numbers of people been engaged in computer programming, a small body of studies on this activity have already been accumulated. These studies are, however, largely atheoretical. The work described here has as its goal the creation of an information processing theory sufficient to describe the findings of these studies. The theory postulates understanding, method finding, and coding processes in writing programs, and presents an explicit model for the coding process.
Computer-naive subjects were taught to use either one or two line editors and then a screen editor. Positive transfer was observed both between the line editors and from the line editors to the screen editor. Transfer expressed itself in terms of reductions in total time, keystrokes, residual errors, and seconds per keystroke. A simple two-component model of transfer is proposed that allows for the differential practice of general and specific components when learning a skill.
Considerable evidence indicates that domain specific knowledge in the form of schemas is the primary factor distinguishing experts from novices in problem-solving skill. Evidence that conventional problem-solving activity is not effective in schema acquisition is also accumulating. It is suggested that a major reason for the ineffectiveness of problem solving as a learning device, is that the cognitive processes required by the two activities overlap insufficiently, and that conventional problem solving in the form of means-ends analysis requires a relatively large amount of cognitive processing capacity which is consequently unavailable for schema acquisition. A computational model and experimental evidence provide support for this contention. Theoretical and practical implications are discussed.
The creation of plan schemas is examined in a naturalistic, longitudinal study of problem solving. Ten novice programmers each wrote eight Pascal programs to solve a series of problems. Their protocols were analyzed to determine how declarative programming knowledge was used to create simple procedural program plans, and how the simple plans were used to create complex plans. Plan creation showed a process of backward development, from the goal to the plan focus, that part of the plan that directly implements the goal. Once defined, it is expanded to create the complete plan, showing backward development of plan parts from the focus. Once the plan is complete, it may be stored as a plan schema and retrieved in subsequent problem solving. The plan will then show forward development as it is implemented in schema order, the order in which plan pieces occur in the finished program. The change from backward development during creation to forward development after retrieval was strongly evident in the statistical analysis of the protocol data. Previous studies of novice programming, which showed only forward development, are explained as special cases of this more general model, cases in which schema knowledge was available to the problem solver.