Educational Simulations
research of literature by Nico Rutten www.linkedin.com/in/NicoRutten
in general
Simulations ...
help the students to identify relations between components of a system and to learn to control such a system.
establish a cognitive framework or structure to accommodate further learning in a related subject area.
provide an opportunity for reinforcing, integrating and extending previous learned material.
offers the opportunity to ..., learn in a relatively realistic problem-solving context, practise task performance without stress, systematically explore both realistic and hypothetical situations, change the time-scale of events, interact with simplified versions of the process or system being simulated.
allow students to ..., postulate abstract concepts in a more concrete manner, convey insight into complicated phenomena and relationships, practice lab techniques prior to the actual laboratory experience
engage student interest
provide the learner with an active role in the learning process
help students observe and understand dynamic processes
enhance decision making skills
provide a realistic cause-and-effect environment in which students can quickly, safely and efficiently investigate to learn.
features that help better learning of difficult concepts of science:
Simulations should be based on real events and data., Too simplified representations may confuse learners,, but exact representations on the other hand may make them over complex.
Use of multiple representations, graphs and an opportunity to observe any graphs forming while the experiment is running (in real time).
Facilities to tailor activity to student ability levels and a narrative for students to follow.
theory
Constructivism
constructivism: the view that students construct their knowledge from individual and/or interpersonal experiences and from reasoning about these experiences
constructivistic approach: a strong emphasis is placed on the learner as an active agent in the process of knowledge acquisition
Constructivist learning environments may frustrate students ..., who have shallow motivations for academic work ..., because there is no correct answer, specific task requirements are not furnished for them, and there is difficulty quantitatively comparing understanding among students., who have more sophisticated motivations to understand the material ..., because of the relatively unstructured nature of the constructivist learning environment.
Cognitive theory
characteristics, attention to non-observable processes and knowledge states, a detailed study of expertise as compared to novice behaviour, revival of research methods such as thinking aloud, and the study of complex, real life tasks additional to highly structured artificial tasks in a lab situation.
instructional principles, conceptual knowledge, sequence and choice of models and problems: offering models in a simulation in a specific order, such that relatively ‘easy’ models are offered first and the ‘target’ models last., by providing a transition from qualitative to quantitative models, or by increasing the number of elements and relations between elements in a model., providing explanations, minimization of error: instead of remediating incomplete knowledge and misconceptions in learners, one can try to prevent these misconceptions from occurring., operational knowledge, learning takes place in a problem solving context, model tracing, immediate feedback, provides the opportunity to point directly at the cause for taking a wrong path in the solution, avoids lengthy explanations that might be necessary if the learner should be allowed to proceed down a wrong path, minimization of working memory load, knowledge acquisition skills, strategies for monitoring comprehension, asking questions about the content, summarizing the content thus far, the teacher/tutor as ..., model: showing the learner the target behaviour, coach: correcting the learner while s/he performs the behaviour, shared responsibility for the task: knowledge acquisition skills seem to be better acquired when working in groups. In groups, learners are ..., asked for explanations, individual thinking is monitored by the audience, and complex tasks are more manageable., teaching and supporting multiple learning strategies: e.g. learners can choose to follow open ended exploration or to hear an explanation
Domain models
instructional recommendations:, providing multiple views of the same model, multiple viewpoints or perspectives, introducing qualitative descriptions of basically numeric relations, offering information on the epistemological qualities of the model, presenting a demonstration mode, in which no learner activity is involved, before continuing with a discovery mode, showing relations in the model in different ways (diagrams, functions), progressive implementation of models, increasing complexity of models by adding elements and relations, increasing level of difficulty, presenting some central, general principles and then proceed by introducing the less central concepts and principles, start with the recall of prerequisite knowledge, working with a sub-model in a complex model, offering domain information in a more direct instructional way
Instructional Design theories
instructional features include:, sequencing of increasingly complex problems to be solved, the availability of a range of help information on request, the presence of an expert troubleshooting module which can step in to provide ..., criticism on learner performance, hints on the problem nature, or suggestions on how to proceed., the option of having the expert module demonstrate optimal performance afterwards, the use of different ways of depicting the simulated system
Conversation theory
distinction between two levels of learning:, learning about “how”, also called “operation learning”, learning about “why”, also known as “comprehension learning”
Full comprehension of a topic means the ability to explain both the how and why.
Bloom
Bloom’s learning goal tripartition:, knowledge-related objectives, related to the acquisition and application of knowledge and understanding, attitude-related objectives, concerned with attitudes and feelings which are brought about as a result of some educational process, motor skills-related objectives, dealing with the development of manipulative or physical skills
Gagné and Briggs
Gagné’s types of learned capabilities:, intellectual skill, which enables the learner to do something that requires cognitive processing – procedural knowledge, cognitive strategy, referring to skill by means of which learners exercise control over their own processes of learning and thinking, verbal information, sometimes called declarative knowledge, to imply that its acquisition makes it possible for the learner to declare/state it, attitude, involving an acquired internal (motivational) state that influences the learner’s choice of personal action, motor skill, which involves refining (muscular) movement in timing and smoothness by the learning that occurs during practice
Gagné and Briggs’ strategy consists of the following phases:, gaining attention, informing the learner of the objective, stimulating recall of prerequisite learning, presenting the stimulus material, overt: the model is presented in one or another way, covert: the learner has to induce model properties from input/output relations, providing learner guidance, eliciting the performance, providing feedback, assessing performance, enhancing retention and transfer
Romiszowski
Romiszowski’s learning goal taxonomy:, Knowledge refers to the information stored in the learner’s mind., factual knowledge, facts: knowing facts, objects, events or people, procedures: knowing what to do in given situations, conceptual knowledge, concepts: specific concepts or groups of concepts, principles: rules or principles which link certain concepts or facts in specific ways, skill, Skill refers to ... which are performed in a competent way in order to achieve a goal., actions (intellectual or physical), reactions (to ideas, things, or people), categories:, cognitive skills, psychomotor skills, reactive skills, reacting to things, situations or people in terms of values, emotions, feelings, interactive skills, interacting with people in order to achieve some goal, such as communication, education, acceptance, persuasion, etc.
Alessi and Trollip
Alessi and Trollip’s classification of simulation types:, physical simulation: allows for the manipulation of physical objects displayed on the screen, giving the student the opportunity to use it or learn about it., procedural simulations: more directly linked to a particular goal, which involves learning about how the simulated machine works, as a means for acquiring the skills and actions needed to operate it., situational simulations: deal with the attitudes and behaviours of people in different situations, rather than with skilled performance., process simulations: learner activity is restricted to initial condition setting, after which the independent development of a process can be observed.
Component Design theory
two models of instruction:, tutorial: information is presented to the student, experiential: the learner may directly interact with the subject matter
Inquiry Teaching theory
instructional actions:, varying cases systematically: presenting specific situations to learners should occur in a systematic order, forming hypotheses, evaluating hypotheses, considering alternative predictions, entrapping students, tracing consequences to a contradiction
Elaboration theory
Proceed from simple ideas from a theory/procedure and then proceed by presenting more complex information.
In order to integrate knowledge the learner is offered …, summarizers: aimed at memorizing knowledge, synthesizers: aimed at deeper knowledge, analogies: relate new ideas to existing ideas
Reigeluth and Schwartz
types of learning goals:, procedural simulation: used to teach the learner to perform a sequence of steps and/or decisions, process simulation: teaches naturally occurring phenomena composed of a specific sequence of events, causal simulations: teaches the cause-effect relationship between two or more changes
stages in the instructional/learning process:, introduction phase, the learner is informed about the scenario of the simulation and about the goals, acquisition phase, knowledge is presented to the learner, demonstration mode in which no learner activity is involved, discovery mode in which there is learner activity, The learner must acquire ..., an understanding of change relationships (process/causal principles), or knowledge of what steps to follow when and how (procedures)., application phase, The acquired knowledge is generalized to a larger set of situations., Conceptual and operational knowledge is transformed from a declarative to a compiled form., assessment phase, The knowledge of the learner is assessed according to some criterion.
Anderson
Anderson’s stages in the acquisition of a particular skill:, declarative stage: skill-independent productions use declarative representations relevant to the skill to guide behaviour, knowledge compilation stage: the system goes from this interpretive application of declarative knowledge to procedures (productions) that apply the knowledge directly., tuning stage: the ‘compiled’ knowledge is refined and consolidated through further practice
Gredler
types of simulations:, experiential simulations, provide students with a psychological reality in which students play roles within that reality ..., by executing their responsibilities, and carry out complex problem-solving in that knowledge domain., types:, data management, diagnostic, crisis management, social process simulations, symbolic simulations, are dynamic in nature and represents the behaviour of a system, or phenomenon, on a set of interacting processes., The students’ role in symbolic simulation is that of principal investigator.
Klahr and Dunbar
theory of scientific discovery as dual search (SDDS), describes the scientific discovery process as a search process in an ..., hypothesis space, containing all possible hypotheses about studied system, experiment space, consisting of all experiments that can be carried out, types of learners:, experimenters: experiments are conducted to identify variables and generate hypotheses, theorists: searching hypothesis space during advanced knowledge acquisition, types of strategies for searching these spaces:, bottom-up (used by experimenters), top-down (used by theorists)
Kolb
The Kolb Learning Cycle, Information can either be obtained through ..., abstract conceptualization: focuses on ..., logical analysis, abstract thinking, systematic planning, or concrete experience: deals with specific encounters, particularly with people, and is fundamentally less systematic., Information is transformed to knowledge by ..., active experimentation: learning is accomplished by ..., performing actions, examining results, or reflective observation: involves ..., looking at the information from several angles, drawing conclusions
Research has shown that ..., 20% of knowledge is retained if only abstract conceptualization is used., Knowledge retention climbs to 50% when reflective observation and abstract conceptualization are used,, 70% with concrete experience, reflective observation and abstract conceptualization, and 90% with all four modes.
computer simulation
computer simulation: a program that constrains a model of a system or a process.
Intelligent Tutoring System (ITS)
a computer simulation which is expanded by implementing teaching functions
types:, expert module, implementation of the rules about how to control the simulated system, diagnosis and student module, an on-line judgement of the learner’s behaviour, tutor module, adaptive tutorial rules which are the basis for linking specific learner behaviour and instructional interventions
In order to create an instructional computer simulation, one must ...
create a simulation model
create a learner interface to the simulation
create an instructional design of the environment
create instructional interventions
integrate the parts of the environment to a complete system
mismatch
On the one hand, most teachers lack expertise and time for creating these environments themselves.
On the other hand, off-the-shelf simulations often do not match the requirements of a specific teacher.
learning support
categories
providing background-knowledge
helping learners to make hypotheses
helping learners to conduct experiments, Learners often show inefficient behaviors when conducting experiments:, manipulating variables that are irrelevant to the hypotheses, being unable to utilize all the information that is relevant to the experiments, tending to design identical experiments, focusing on obtaining the required results rather than understanding the concept model, focusing on entertainment factors rather than on obtaining a deeper understanding of the concept model
helping learners to interpret data
helping learners to regulate the learning process
types
Interpretative support that helps learners with ..., knowledge access and activation, the generation of appropriate hypotheses, and the construction of coherent understandings.
Experimental support that scaffolds learners in ..., the systematic and logical design of scientific experiments, the prediction and observation of outcomes, and the drawing of reasonable conclusions.
Reflective support that ..., increases learners’ self-awareness of the learning processes, and prompts their reflective abstraction and integration of their discoveries.
ways in which guidance can be provided:
providing favourable conditions
stimulating learning processes, Socratic dialogue in order to have the learner revise his/her thoughts, systematic variation of cases, providing hints/suggestions on how to perform exploratory learning, giving goals or assignments with the simulations, fault diagnosis task, prodding techniques for encouraging the statement of hypotheses, take over processes from the learner in order to enable other processes
problem-solving strategies:
goal decomposition, by explicitly suggesting a goal, or by asking leading questions.
reminding the learner of previous solutions
simplification, can help learners who ..., get stuck at some point in a complex problem, or are intimidated by a complex problem.
assignments
Assignments can be introduced as a mechanism to help learners in their goal setting behavior.
types of assignments:, do-it assignments: give the learner the general assignment to explore the model, investigation assignments: ask learner to investigate the relation between two or more given variables, explicitation assignments: have an initial state or sets of initial states for the simulation associated with them, the role of the learner being to run the simulation and to observe the impact on the simulation, specification assignments: the learner has to predict the values of certain variables when the associated simulation stops, optimization assignments: the learner has to vary the simulation’s variables’ values so that the constraints specified by the author are not broken and a target specified by the author is reached.
heuristics
Providing students with heuristics ..., supports performing systematic experiments, encourages them to provide evidence for the conclusions they draw, can facilitate them in ..., generating meaning from data, and making connections among ..., procedures, data, evidence, claims
scaffolding
scaffolding: the process by which assistance is provided that enables learners to succeed in problems that would otherwise be too difficult
The intention is that the support not only assists learners in accomplishing tasks, but also enables them to learn from the experience.
conceptual change
Alternative conceptions have the general characteristics of ...
being poorly articulated
being internally inconsistent
being highly dependent on context
and having tremendous explanatory power in the mind of the student.
Educational conditions that promote conceptual change:
The student must experience dissatisfaction with an existing conception.
The new conception must be intelligible.
The new conception must be plausible.
The new conception must be fruitful.
Children often resolve their misconceptions with the goal of ...
receiving a good grade
preserving self-esteem in an intellectually overwhelming situation
or bringing closure to a learning situation at any cost.
before/after instruction
before
Simulation provided before instruction may function as a conceptual model that allows students to better understand and encode the didactically presented information. Students receiving such a model may be better able to recall the presented didactic information and to reason with the principles taught in transfer situations.
instructional role: setting the stage for future learning; simulations can ..., provide motivation, reveal misconceptions that would inhibit learning, provide an organizing cognitive structure for receiving new material, and serve as concrete examples of complex, abstract concepts.
after
Students who receive simulation after didactic instruction may find it difficult to make sense from the model with which to assimilate the instructional information. That being the case, during didactic instruction, they may not be able to encode a given information into cognitive structure as well as students who had had prior simulations.
instructional role: providing an opportunity to apply or integrate newly acquired knowledge; they ..., are used to support the acquisition of diagnostic skills or processes, and can uncover misconceptions in newly acquired knowledge.
multiple representations
Combining different representations in one interface may have several advantages:
Each representation can show specific aspects of the domain to be learned., Text and pictures are good representations to present the context of a problem., Diagrams are well suited for presenting qualitative information., Graphs, formulas, and numeric representations can be used to show quantitative information.
One representation can constrain the interpretation of another representation., The purpose is not to provide new information, but to support the learners’ reasoning about the less familiar representation.
By translating between representations, learners build abstractions that may lead to a deeper understanding of the domain.
When learning with multiple representations, learners are faced with four tasks:
They have to understand the syntax of each representation.
They have to understand which parts of the domain are represented.
Learners have to relate the representations to each other if the representations (partially) present the same information., relating: linking the surface features of different representations
Learners have to translate between the representations., translating: having to interpret the similarities and differences of corresponding features of two or more representations
Problems learners may encounter when learning with multiple representations:
They have difficulties relating different representations., split-attention problem: When learning with separate representations, learners are required to relate disparate sources of information, which may generate a heavy cognitive load that may leave less resources for actual learning.
Novices have problems in translating between representations.
Ways to make relations between representations explicit for the learner:
physically integrate the representations, Multiple representations, when integrated, appear to be one representation showing different domain aspects.
provide the learner with dynamic linking, Actions performed on one representation are automatically shown in all other representations. problems with dynamic linking:, Dynamic linking may discourage reflection on the nature of the translations, leading to a failure by the learner to construct the required understanding., With multiple dynamically changing representations, learners need to attend to and relate changes that occur simultaneously in different regions of various representations, which may lead to cognitive overload.
scientific discovery learning
scientific discovery: the processes of mindful coordination between hypothesized theories and evidence collected by experiments
The content of a domain is not explicitly stated to learners.
Learners experiment and construct knowledge as ‘scientists’:
They provide the simulation with input.
observe the output
draw their conclusions
and go to the next experiment.
A technique that has proven useful in eliciting inquiry learning processes is by the use of computer simulations.
It leads to knowledge that ...
is more intuitive and deeply rooted in a learners’ knowledge base, Intuitive knowledge is ..., hard to verbalize, immediately available or not available at all, often relies heavily on visualizations, and acquired only after inferring knowledge in rich, dynamic situations.
has a more qualitative character.
approaches
a top-down method or concept-driven way of discovery, knowledge plays a central role
a bottom-up approach or data-driven way of discovery, features of the environment (e.g. a simulation interface) are of central importance
processes
transformative processes: the reasoning and decision-making that guide manipulating a computer simulation and extracting information from it:, orientation, hypothesis generation, identifying variables, selecting variables, and defining the relation that is hypothesized to hold between the selected variables., hypothesis testing, hypothesis train: a set of consecutive hypotheses concerning one set of variables or related variables, and drawing conclusions.
regulative processes: meant to control the discovery learning process on a metacognitive level:, monitoring one’s own behavior, keeping track of progress, and planning in advance what steps to undertake.
influence on effectiveness
prior domain-specific knowledge
generic knowledge of quantitative and qualitative relations between variables
discovery skill: student’s aptitude at performing and interpreting experiments
and metacognition: the generic ability to regulate discovery learning processes
problems
for the process of hypothesis generation:, Choosing hypotheses that seem “safe” and unsuccessfully transforming data into a hypothesis., The distance between the theoretical variables and the variables that are manipulated in the simulation.
for designing experiments: learners who ..., design inconclusive experiments, show inefficient experimentation behavior, follow a confirmation bias, apply an engineering approach instead of a scientific one.
Learners quite often have trouble with the interpretation of data as such.
Students are not very capable of regulating the learning process.
computer-mediated collaboration
transformative processes, learners’ activities in these phases are performed for the sole purpose of yielding knowledge, support:, hypothesis generation, Offering fully specified hypotheses has the potential disadvantage of revealing the key variables and relations to the learner., experiment design, Inconclusive experiments often arise because learners vary too many variables at once., Inefficient experimentation occurs ..., when learners fail to design informative experiments, or when they repeat previous experiments., Designing experiments that are not intended to test hypotheses can be indicative of an engineering approach., engineering approach: one in which learners attempt to create a desirable outcome instead of trying to understand the model., data interpretation, Learners often experience difficulties in drawing conclusions from their data.
regulatory processes, serve to manage and control the inquiry learning process., support:, planning, monitoring, evaluation
deficiencies, Face-to-face dyads attain higher performance scores than students who collaborate online., Face-to-face dyads have more prolonged discussions on learning tasks., Communication in computer-mediated dyads is more directed at ..., coordinating efforts, operating the communication tool, and expressing emotions.
epistomology
epistemic motivation: one’s beliefs toward knowledge and the process of building knowledge
Epistemologically less mature students ...
believe that knowledge is simple and certain
should find less compatible an approach that emphasized self-exploration and self-construction of knowledge
and achieve best in a more prescribed, confirmatory simulation environment.
Students with greater epistemological sophistication ...
do better in an exploratory simulation environment.
metacognition
metacognition: both the knowledge about one’s own cognitive processes and the control over these processes
Metacognition appears to be ...
one of the most important determinants for successful learning in general
and for successful inductive learning with computer simulations in particular.
metacognitive knowledge: knowledge about the interplay between ...
person characteristics
task characteristics
and the available strategies in a learning situation.
metacognitive skills: self-regulatory activities actually being performed by a learner in order to structure the problem solving process. operationalizations:
deep orientation
systematic orderliness
accuracy
evaluation
elaboration
instruction
instructional measures: either instructional strategies or actions
providing direct access to domain information
providing learners with assignments (or questions, exercises, or games)
model progression
instructional principles: common rules of thumb which can be implemented in particular features of the environment
instructional approach: the overall policy involved
learning by observation
coaching: the student is not guided through the domain in a structured way, but is monitored and corrected by a coach
learning by instruction: the student is guided through the domain in a structured way
instructional features: both the more passive features of the environment as well as the active measures taken by the instructional agent
learning goal
dimensions on which a particular learning goal can be classified:
kinds of knowledge to be learned:, conceptual knowledge: knowledge of principles, concepts and facts related to the (class of) system(s) being simulated, operational knowledge: knowledge about sequences of cognitive and/or noncognitive operations (procedures) that can be applied to the (class of) simulated system(s).
formats knowledge can be encoded in:, declarative knowledge: represented in a format ..., that is relatively easy to acquire, that makes the knowledge relatively easy to report upon, that makes the knowledge of potential use in an unlimited number of problem contexts, and that requires interpretation in order to use it in a task., compiled knowledge: represented in a format ..., that is only obtained after using the knowledge in a problem-solving context, that makes the knowledge hard to report upon, that restricts its potential use to a limited number of contexts, and that can be used in a more automatic, effortless way.
scope of the target knowledge:, domain-specific knowledge: specific to the simulation domain at hand, generic knowledge: not specific to the simulation domain at hand, but extends to other domains as well.
cognitive load
distinction:
due to the difficulty of the subject material
induced by the instructional context in which the subject material is embedded
To maximize learning, instructors must minimize cognitive load by ...
limiting the amount of material presented
having a clear organizational structure to the presentation
linking new material to ideas that the audience already knows
and avoiding unfamiliar technical terminology and interesting little digressions.
