2.1. Intuitive Knowledge
Intuitive knowledge, which is considered unreflected knowledge (cf. [18
]), enables the anticipation of a situation’s possible outcomes in comparatively less time than reflecting about the situation [20
]. Concerning this matter, Swaak and de Jong [8
] speak of intuitive knowledge as being quickly available, and they defined this knowledge type as “the quick perception of anticipated situations” (p. 288). Fensham and Marton [21
], as referred to in Lindström, Marton and Ottoson [22
], defined intuition as “formulating or solving a problem through a sudden illumination based on global perception of a phenomenon” and stated that it “originates from widely varied experience of that phenomenon over a long time” (p. 265). It is emphasized that necessary experiences attained over time are stored in the long-term memory as prior knowledge (cf. [10
]) to allow the acquisition of intuitive knowledge.
Research literature shows a wide range of terms associated with intuitive knowledge, such as intuition [10
], tacit knowledge [29
], implicit knowledge [31
], and intuitive understanding [22
]. Among the various definitions of intuition, e.g., references [20
], we refer to the definition by Swaak and de Jong [8
] and implement the concept of intuitive knowledge synonymously with the term intuition.
It is supposed that intuitive knowledge (or intuition) can be seen as a kind of ‘hunch’ [28
] and may influence an individual’s judgment, decisions, and behavior. In this respect, Betsch [10
] defined intuition as “a process of thinking” where the output “can serve as a basis for judgement and decisions” (p. 4). Here, it is assumed that intuitive knowledge offers the initiation of ensuing learning processes [11
]. It is argued that intuition seems to influence human decision-making [34
]. Fischbein [26
] constitutes that intuitions are implicit and operate automatically on a subconscious level (cf. [10
]). Referring to Westcott [35
], Fischbein [26
] cited that intuition “occurs when an individual reaches a conclusion on the basis of less explicit information than is ordinarily required to reach that conclusion” (p. 97). Hence, intuitive knowledge is considered to be unreflected knowledge and allows conclusions to be reached by using less existing and necessary information [35
]. In this regard, intuition differs from deliberate cognitive and metacognitive processes which require attention and working memory capacity [36
]. Therefore, intuitive knowledge originates beyond conscious thought [10
] and rather, may be regarded as naive lay knowledge [37
]. Mostly, learners cannot pay attention to all given information simultaneously, and thus, they are requested to focus on relevant information. This deliberate process requires sequential processing of particular given information. In contrast, intuition is capable of handling an enormous amount of information concurrently by using already existing experiences stored in the long-term memory, and it can play an essential role in problem solving.
The five characteristic criteria for intuitive knowledge according to Swaak and de Jong [7
], are as follows:
Intuitive knowledge can only be acquired by using already existing previous knowledge in perceptually rich dynamic situations. It is assumed that when applying previous knowledge in situations containing a huge amount of information, implicitly induced learning processes lead to intuitive knowledge acquisition.
Intuitive knowledge is hard to verbalize. This means that intuitive knowledge differs from conceptual knowledge, which is regarded as a network of concepts and their relationships to a functional structure generated through reflective learning that can be articulated (cf. [38
]). However, according to Lindström, Marton and Ottoson [22
], intuitive and conceptual understanding should not be considered as separate knowledge types. They believe intuitive and conceptual understanding to be intertwined aspects of a learner’s awareness. Hence, intuitive knowledge can be seen as a quality of conceptual knowledge [39
Perception is crucial when referring to intuitive knowledge. The illustration of situations plays an essential role in the acquisition of intuitive knowledge. In this regard, Fischbein [26
] emphasized the importance of visualization through external representation.
Another characteristic referring to intuitive knowledge is the importance of anticipation. Anticipation refers to the presumption of occurrences, developments, or actions. Intuitions anticipate what will or will not happen, and intuitive evaluation anticipates the possible outcomes of a situation without the ability to explicitly explain them [7
]. Here, intuitive knowledge can be ascribed as ‘know without knowing’ [10
] (p. 4), so that ‘the input to this process is mostly provided by knowledge stored in the long-term memory that has been primarily acquired via associative learning. The input is processed automatically and without conscious awareness. The output is a feeling that can serve as a basis for judgments and decisions’ [10
] (p. 4).
It is assumed that the access to intuitive knowledge in the memory is different to the access to declarative knowledge as factual and conceptual knowledge. The difficulty of verbalizing intuitive knowledge might be one reason for this differential access. Swaak and de Jong [8
] mention that ‘the action-driven and perception-driven elements in learning ‘tune’ the knowledge and give it an intuitive quality’ (p. 287).
It is argued that learners acquire intuitive knowledge while learning with computer simulations [7
]. Thomas and Hooper [43
] mentioned that computer simulations could be considered to be ‘experiencing programs’ with the opportunity for learners to acquire “an intuitive understanding of the learning goal” (p. 499). Swaak and de Jong [7
] proposed that intuitive knowledge can only be acquired after applying already existing previous knowledge in situations perceived as rich and dynamic. From a rich learning environment, such as a computer simulation, learners are capable of extracting a great amount of domain-specific information that is usually displayed as a dynamic, graphic representation of the output. In this regard, intuitive knowledge cannot be generated by learning by a more traditional approach, for example, by only reading textbooks. Here, Fischbein [26
] constitutes that intuitions “can never be produced by mere verbal learning”. Intuitions “can be attained only as an effect of the direct, experiential involvement of the subject in a practical or mental activity” (p. 95). Learners have to participate in the learning process actively [44
The next section describes the special features related to learning with computer simulations.
2.2. Learning with Computer Simulations
Computer simulations are considered a technically highly sophisticated option that offer numerous benefits for the learning and teaching of science [45
]. For example, they can potentially improve learners’ understanding of abstract biological phenomena, such as predator–prey relationships and give learners opportunities to harmlessly and interactively conduct experiments [48
]. Learning from computer simulations in discovery environments sees learners as active participants in their learning processes constructing their individual knowledge bases [6
]. Learners actively participate in their learning processes by conducting computer-simulated experiments. They are required to independently find relationships between given variables in a domain and do not merely passively absorb information. Hence, it is supposed that the learning outcomes using computer simulations are different from those acquired by learning with an approach emphasizing pure knowledge transfer [7
]. Consequently, knowledge acquisition when learning with computer simulations seems different from learning with a more or less explanatory approach, such as learning from texts [7
Swaak, de Jong, and van Joolingen [8
] described three characteristics related to the use of computer simulations as discovery environments. Firstly, a computer simulation can be regarded as a rich environment that offers learners a great amount of information they could extract by themselves. Secondly, opportunities for active experiences are characteristics ascribed to computer simulations. Learners actively engage in the learning process and should not merely absorb information via the computer screen. Learners are asked to conduct experiments using a computer simulation so as to be aware of a domain. Thirdly, another characteristic of computer simulations, in contrast to textbooks, is that they are learning environments with low transparency. Information and relationships between given variables in the computer simulation extracted by the learner are not explicitly presented. Learners are asked to deduce the characteristics underlying the simulation when they explore rules, principles, terms or concepts while conducting experiments [3
]. Learners are to determine the hidden model behind the simulation by conducting experiments [6
However, empirical findings report several cognitive and metacognitive difficulties regarding scientific discovery learning with computer simulations [2
]. De Jong and van Joolingen [2
] differentiate four types of difficulties learners may experience during discovery learning. The current study focuses on two problems learners often encounter when interpreting data and on difficulties pertaining to self-regulation when learning with a scientific computer simulation [13
Specific instructional support for learning using computer simulations can foster successful knowledge acquisition [2
]. In their review article de Jong and van Joolingen [2
] pointed out that specific instructional interventions to support data interpretation and self-regulation are effective for computer simulation-based learning. In a literature review Urhahne and Harms [55
] inferred that specific supporting measures for data interpretation and self-regulation strongly effected students’ knowledge gain. Also, negative effects on achievement could not be determined when learning with computer simulations was supported with instructional support for data interpretation and self-regulation [55
The following text presents approaches to overcoming difficulties that learners encounter with scientific discovery learning using computer simulations, and improving knowledge acquisition with particular instructional interventions.
2.3. Supporting Learning from Computer Simulations
Instructional support is necessary to cope with difficulties when learning with computer simulations, to foster learning outcomes and to enhance more successful and goal-oriented knowledge acquisition [2
]. The latter should be supported in several ways. In their triple scheme for discovery learning with computer simulations, Zhang et al. [57
] differentiate three spheres of instructional support to be given in three different phases of the learning process:
Interpretative support enables learners to access and use prior knowledge and develop appropriate hypotheses;
Experimental support enhances learners’ ability to design verifiable experiments, to predict and to observe simulation results, and to adequately draw conclusions;
Reflective support increases the learners’ ability to raise self-awareness of the learning processes and helps support the combining of abstract and reflective integration of their discoveries.
The four learning difficulty categories associated with computer simulations (hypotheses generation, design of experiments, interpretation of data, and regulative learning processes) suggested by de Jong and van Joolingen [2
] can be integrated (cf. [57
]) into the above mentioned spheres. The scheme proposed by Zhang and colleagues [57
] provided the theoretical framework for the presented study and is used to explain the various options of instructional support.
enhances awareness of the importance of the discovery process. Learners need to activate their prior knowledge to generate appropriate hypotheses and to obtain an appropriate understanding. Interpretative support is provided to enhance problem representation, ease access to prior experiences, facilitate the handling of the computer simulations and is generally given before
learners begin conducting experiments with the computer simulations. Here, domain-specific background information available to the learner is an effective and supportive intervention tool [53
]. Providing learners permanent access to specific information, such as principles, concepts, terms or facts of a domain, during their interaction with the computer program seems beneficial for knowledge acquisition [53
], whereas offering this information earlier on seems less effective [53
]. Another kind of interpretative support is concrete assignments that learners have to work on using computer simulations [13
A basic framework for an assignment can be based on a POE (Predict-Observe-Explain) strategy [61
]. Using this strategy, learners, are first required to predict the outcome of a given task. Learners are then asked to conduct a related experiment and then describe its outcome. Finally, learners have to compare the outcome with their prediction. Hereby, learners can be directed to explore important relationships between variables. Worked-out examples can be a type of interpretative support. In numerous studies, worked-out examples show positive effects on knowledge acquisition (e.g., [62
]), especially for novice learners. Worked-out examples introduce a specific problem, suggest problem solving steps, and give a detailed description of the appropriate solution [68
]. Applying worked-out examples may be considered an effective method to increase learners’ problem solving skills [16
is used to improve scientific inquiry while
learners use the computer simulation. This kind of support helps learners adequately design scientific experiments, appropriately predict outcomes, observe outcomes, and draw conclusions (cf. [71
]). Effective experimental support interventions include the progressive and cumulative introduction to handling the computer simulations, explanation of important simulation parameters within the computer program, the request to predict possible simulation outcomes by the learner, and answering the learner’s requests to describe and to interpret the simulation outcome. Progressive and increasing introduction to a computer simulation gives learners the relevant information to work with the simulation and to work on assignments. In particular, highly informative computer simulations possess a structure of increasing complexity [72
] to assist in avoiding a possible cognitive overload due to the informational richness. In a study of Lewis, Stern, and Linn [73
], the prediction of possible simulation outcomes by the learners had a tendency to lead to a higher knowledge gain than in a control group. The ability to describe and justify one’s own simulation outcome as an experimental support indicates successful knowledge acquisition [74
scaffolds learners’ metacognitive knowledge after
conducting computer simulated experiments. This kind of instructional support fosters the integration of the newly discovered information. Reflective support enables learners to increase their self-awareness of the learning processes and supports abstract and reflective implementation of their discoveries. For example, a reflective assessment on one’s own inquiry can lead to an improved comprehension regarding the meaningful knowledge acquisition. This can be realized, for example, when learners are encouraged to assess and reflect on their own inquiry using a reflective assessment tool within the computer simulation [75
]. In the ThinkerTools-Curriculum [75
], learners are initially required to assess their own inquiry. Subsequently, learners are requested to justify their assessment in written form. This reflective support has been shown to lead to positive effects in knowledge acquisition.
The following describes two categories of difficulty that learners often encounter with processes of scientific discovery learning: data interpretation and self-regulation. Subsequently, for these two categories specific instructional interventions are presented to support scientific discovery learning with computer simulations.
To enhance the effectiveness of data interpretation, empirical findings have shown that explanatory and justified feedback about a simulation outcome supports effective knowledge acquisition, particularly for novices [77
]. In a study by Moreno [77
], learners received explanatory or justified feedback about their approach to the computer program after they had worked on computer-based tasks. A study by Lin and Lehmann [78
] confirmed the assumption that the description and interpretation of simulation outcomes generated by a learner leads to effective knowledge acquisition. After learners completed their computer simulated experiments, they were requested to justify the simulation outcomes.
It has been proposed that self-reflective tasks can effectively support self-regulated learning. A concrete method of improving awareness, such as reflective self-assessment, shows positive effects on knowledge acquisition [75
]. Learners were asked to reflect on their own and their classmates’ inquiries based on provided criteria after having completed computer-simulated experiments [75
]. The method of reflective self-assessment supported learners in the self-reflection phase [79
] and shows positive effects with regard to learners’ knowledge gain.
Less systematic has been the examination of the impact when using specific instructional interventions on the acquisition of intuitive knowledge to call upon learners’ reflective activities, such as the justification of simulation outcomes, and tasks or hints to encourage the reflection on inquiries during the learning processes. However, these instructional interventions have the potential to improve simulation-based learning and intuitive knowledge acquisition. Hence, the current study developed and tested specific instructional interventions for data interpretation and self-regulation. Intuitive knowledge acquisition was measured depending on these instructional interventions. One assumption is that when learning with a scientific computer simulation learners’ intuitive knowledge acquisition could be effectively enhanced by the instructional interventions for data interpretation and self-regulation.
2.4. Assessing Outcomes from Learning with Computer Simulations
Due to the lower transparency of computer simulations [44
], learners have no direct view on the variables and their relationships. Therefore, learning with computer simulations has several effects which cannot be revealed by knowledge tests designed in a more conventional way, e.g., traditional multiple-choice tests. Through a meta-analysis, Thomas and Hooper [43
] found that ‘the effects of simulations are not revealed by tests of knowledge (…)’ (p. 479). Numerous empirical studies that have focused on knowledge acquisition when learning with computer simulations have used tests emphasizing knowledge gain in an explicit matter. These tests have mostly involved requests for domain-specific facts (e.g., terms and definitions) or concepts to discover interrelations between the given variables. In contrast, to assess declarative knowledge with more or less traditional paper-and-pencil methods, intuitive knowledge can be assessed methodologically by asking for predictions of situations [7
]. In this regard, referring to Swaak and de Jong [8
], and based on their definition of intuitive knowledge as ‘the quick perception of anticipated situations’ (p. 288), the particular components of their definition can be described with respect to the assessment of intuitive knowledge, as follows.
Quick: Time taken to answer items is considered to be an indicator as to what degree knowledge can still be regarded as intuitive. Learners are requested to answer an item as quickly as possible. In this regard, the test format to assess intuitive knowledge described in this study differs from more traditional approaches, for example, multiple-choice tests. It is assumed that knowledge gained by quickly answering items has an intuitive quality.
Perception: With respect to the item format, perception seems to be crucial. For this reason, and compared to many other tests assessing declarative knowledge, short texts and pictures with minimal textual information can be used.
Anticipated: An important aspect of intuitive knowledge is anticipation. An item consists of a given initial situation with predetermined variables. The change in a particular variable leads to a possible outcome that can be anticipated.
Situation: Each item contains a question part, consisting of a question or a commencement of a given statement with a change in that situation, and a response part, consisting of possible answers. Consequently, each item comprises an initial situation influenced by a given action or a changed value of a variable and possible (post)situations to be anticipated.
These components, suggested by Swaak and de Jong [7
], were considered and provided a basis for developing the test instrument to assess intuitive knowledge used in the study presented here.