前往顾页

Collaborative Discovery Learning Based on Computer Simulatio

时候:2010-07-28 22:00来源:知行网www.zhixing123.cn 编辑:麦田守望者

 

Abstract
A growing number of studies have focused on scientific discovery learning based on computer simulation. However, most previous studies focus on individual discovery learning without considering the perspective of social interaction. This study implemented an experiment among eighth graders to investigate the effects of peer collaboration and questioning-explanation prompt on simulation-based scientific discovery learning. In the result, peer collaboration had prominent effect on the discovery outcomes, intuitive understandings and variable control skills. The influence of the questioning-explanation prompt was less clear, hence need to be addressed in depth in future studies.
Keywords: Collaborative learning, discovery learning, simulation, peer learning
 
 
INTRODUCTION
 
The most recent developments in learning theory including social constructivism, situated learning and distributed cognition have converged to interpret learning as an active knowledge construction process occurring in social and authentic contexts. From these perspectives, new technologies should be harnessed as powerful mind tools to enhance generative understanding, higher order thinking, as well as intensive social interactions, rather than merely be added in classrooms as media for information transmission. Within the constructivist paradigm, a growing number of studies have focused on scientific discovery learning based on computer simulation (van Joolingen and de Jong, 1997). This study is dedicated to explore the processes and effectiveness of collaborative discovery learning in computer simulation environment.
 
Supporting scientific discovery learning in simulation environment
 
Since computer simulation has the capacity to provide learners with an exploratory learning environment, it has been regarded as a powerful tool for scientific discovery learning. Nevertheless, many studies designed to contrast the effectiveness of simulation-based discovery learning to some mode of traditional learning found little persuasive evidence in its favor (see Banggert-Drowns, Kulik, & Kulik, 1985; Carlsen & Andre, 1992). The question arises, why does simulation-based learning, involving learners in active inquiry, not improve learning outcomes more consistently? One explanation lies in the wide range of difficulties learners may encounter in coping with discovery learning processes. De Jong and van Joolingen (1998) classified the problems that learners may encounter into four categories: (a) difficulties in handling hypotheses, (b) poorly designed experiments, (c) difficulties in data interpretation, and (d) problems regarding the regulation of discovery learning. Despite its potential in stimulating constructive learning activities, it seems that the simulation-based learning environment cannot guarantee effective learning without sufficient support (“scaffolding”) for discovery learning activities. This conclusion has also been supported by Lee’s (1999) meta-analysis showing that hybrid simulation involving instructional elements is more effective than pure simulation for new knowledge acquisition.
Studies have been conducted to help learners with particular strategies from specific aspects. For example, some researchers developed supportive methods to help generate hypothesis in simulation-based discovery learning (Njoo & de Jong, 1993; Quinn & Alessi, 1994; Shute & Glaser, 1990). Others have looked at the issues connected with experimental design (Leutner, 1993), planning (Tabaket al., 1996), explaining the phenomenon and predicting the result (de Jong, 1999), and access to an appropriate knowledge base (Lewis et al., 1993).
Until the recent times, most studies on supporting simulation-based discovery learning adopted the strategies-oriented solution, intended to examine the effect of specific strategies according to learners’ difficulties in particular aspects. In his doctoral dissertation, Zhang (2000) proposed a triple scheme of learning support design for simulation-based scientific discovery learning. It was hypothesized that three interrelated main conditions may determine the effectiveness of scientific discovery learning to a great extent. (1) The meaningfulness of discovery processes: Learners need to activate their prior knowledge and map the knowledge on to the problem being addressed to help representing the problem and generating appropriate hypotheses and understandings. (2) The systematicy and logicality of discovery activities: Effective discovery learning involves proper scientific reasoning, the systematical manipulation of the variables and qualified experiment designs. (3) The reflective generalization over the discovery processes, which means the self-monitoring of the discovery processes and the reflective abstraction and integration of the discovered rules and principles. According to the three hypothesized conditions, three types of learning supports were designed and geared towards the three spheres: (a) interpretative support that helps learners with knowledge access and activation and the generation of appropriate hypotheses and coherent understandings, (b) 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, and (c) reflective support that increases learners’ self-awareness of the thinking and understanding processes and prompts their reflective abstraction and integration of their discoveries. Three experiments were implemented to investigate the internal conditions by examining the effects of the three types of learning support. Overall, the results support the main hypotheses, showing that generative meaning-making, systematic and logical design of experiments, and the reflective generalization constitute three essential internal conditions in scientific discovery learning, hence learning supports geared to the three spheres should be taken into account when designing simulation-based discovery learning environment.
 
The role of collaboration in discovery learning and reasoning
 
In most studies, scientific discovery learning is regarded as an individual scientific reasoning process that involves the generation of hypotheses and testing them against the collected evidence (Klahr and Dunbar, 1988). The perspective of social interaction has however been disregarded to some extent. Collaborative discovery is a common and growing practice in science, but has not yet been extensively researched (Okada & Simon, 1997). In a study, Li & Chen (2000) compared two modes of cooperatively using computers in primary geometry learning: Cooperative Tool mode in which learners in small groups used Windows PBrush to explore the area of rectangles, and Cooperative CAI mode where learners used tutorial CAI program cooperatively to learn the same knowledge. In the results, Cooperative Tool mode brought on significant outcomes in reasoning ability, geometric knowledge, and learning motivation than Cooperative CAI. The outcomes seem to imply that collaboration combined with open technology tools have more potentials to promote explorative learning and higher-order thinking.
Among the few studies concerning collaborative reasoning or discovery, Gorman et al. (Gorman et al., 1984; Gorman, 1986) studied confirmation bias in group scientific discovery using a rule-discovery task - the "2-4-6 task". Subjects were instructed to follow either a confirmatory strategy, a disconfirmatory strategy, or a combination of the two. In the result, groups outperformed individuals, and subjects in the disconfirmatory condition performed best, followed by those in the combined condition and the confirmatory condition respectively. Freedman (1992) studied the effect of group versus individual problem solving and the effect of entertaining multiple hypotheses versus a single hypothesis in the 2-4-6 task. The groups outperformed individuals, and in the multiple hypotheses condition groups used more diagnostic tests than individuals, which had improved their performance.
In an important study of collaborative discovery learning, Teasley (1995) investigated the role of verbal behavior in fourth grade students' peer collaborations in accomplishing the spaceship task. The subjects were assigned to one of the four conditions: Talk Alones solved the problem alone while talking aloud; No-Talk Alones solved the problem alone without talking aloud; Talk Dyads solved the problem with a partner while talking to each other; No-Talk Dyads solved the problem with a partner without talking to each other. Talk Dyads performed best, followed by Talk Alones and No-Talk Alones, with No-Talk Dyads performing worst.
Pilkington & ParkerJones (1996) described a formative evaluation study in which medical students were recorded interacting with a simulation model of calcium balance. The role of dialogue interaction, between students and between experimenter and student, was investigated as a potential source of learning support. Although all students demonstrated learning gains, these were greater amongst single students interacting with the experimenter than amongst students co-operating on tasks in pairs. Dialogue analysis revealed that experimenter-student interaction contained many more inquiries causing students to explain their reasoning, justify their conclusions, or state implications from data. This prompting of reflective activity seemed to result in a deeper understanding of the physiological model.
Okada & Simon (1997) explored the processes and effectiveness of Singles' and Pairs' discovery activities using protocol analysis method. Subjects who were undergraduate students were asked to discover molecular genetics laws in a computer micro-world. Compared with Singles, Pairs were more successful in discovering the correct rules, and participated more actively in explanatory activities (i.e. entertaining hypotheses and considering alternative ideas and justifications). Explanatory activities were promotive for discovery only when the subjects also conducted crucial experiments. Explanatory activities were facilitated when paired subjects made requests of each other for explanations.
 
Discourse patterns for mediating collaborative learning
 
Collaborative learning context provides opportunities for learners to interact with one another in verbal and non-verbal, paralinguistic ways. It is the levels and patterns of interactions that determine the effectiveness of collaborative learning. In the studies of group learning, Webb (1989) consistently found that giving elaborated explanations to others in the group is a strong predictor of achievement.
In a series of studies of peer learning, King (1999) developed a procedure called "Guided Peer Questioning", which is a question-asking and answering procedure designed to promote interaction and learning in small groups. This procedure is guided by structured question "starters" such as, "Why is …important?", "How is …similar to …", "What does … remind you of ? And why?". These open-ended question starters are provided to members of the learning group, who then use the starters to guide them in generating their own specific questions related to the material being studied. A series of studies (King, 1992, 1994; King & Rosenshine, 1993) revealed that when students were taught to ask each other these thought-provoking questions during learning, their questions prompted high-level interaction and learning. Similarly, Coleman (1998) examined the effect of scaffolded explanation-based approach in collaborative discussion in classroom settings. During discussion, learners in experimental groups were given prompt cards to guide their questioning activity toward each other. This approach was found to be of great help in promoting learners' understanding of scientific knowledge. However, the learning tasks used in the above studies were mainly knowledge acquisition in lecture or text-based context. Could such kind of reciprocal questioning and explanation approach facilitate collaborative discovery learning based on computer simulation? No study so far has been found to address this question.
 
The Purpose of the Present Study
 
While collaborative discovery learning with simulation is an important topic, relatively few experimental studies have been conducted around it. Among the few relevant studies, what researchers are interested is not how students deal with discovery learning collaboratively, but mainly human's scientific discovery as a topic in psychology of science (Okada & Simon, 1997; Gorman, 1986; Freedman, 1992). Very few studies have been conducted to explore collaborative discovery learning among young learners in schools. Could collaboration function as a type of learning support to facilitate learners' discovery processes based on computer simulation? How will reciprocal questioning and explanation activities promote discovery learning? This study aims at answering these questions through an experiment. Peer learning approach will be adopted among collaborative subjects. Explanatory prompt cards will be used to guide learners' questioning and explanation activities. Both the effects and the processes will be evaluated to reveal the influence of the treatments.
 
METHOD
 
The Simulation Learning Environment
 
The domain chosen for the simulation was floating and sinking, where the subjects were required to explore the upthrust on objects floating or sinking in water. This topic is one of the core issues in secondary science learning and encompasses the representative variable structure of scientific discovery tasks. In this study, learners’ task was to discover which one or more of three given factors (shape, mass and volume of object) were related to the size of the upthrust on an object. For floating objects, the upthrust equals to its weight and the two forces are a pair of balanced forces, which make an object stay still in water. So the mass of object is the only influential factor to the size of upthrust under this circumstance. For objects submerged in water, as can be predicted based on Archimedes's law, the size of the upthrust depends only on the volume of the object, because the volume of object determines the volume of the excluded water, which influences the size of the upthrust in turn. Learners often hold misconceptions about this phenomenon, e.g. assuming that the size of the upthrust depends on the shape or volume of the floating object.
The simulation started with floating objects determined by the default parameters (volumes, masses) in the program. No deliberate instruction was offered to guide the subjects to take both floating and sinking circumstances into account when conducting experiments with the software.
 

 

 
 
 
 
 


 

Figure 1. Interface of the simulation software
The simulation adopted paired-instance design that requires learners to construct a pair of experiments at a time, so that they could contrast the outcomes of two instances conveniently. For example, in order to examine the effect of the volume of object, a learner can drag two objects of the same shape (e.g. ball) to the top of the left and right container, set the values on the left and right to keep their masses the same, and vary the volumes to be different. Then he/she can click the “RUN” button to see whether the upthrusts will be different or not in result. For all the students, a data sheet was provided on screen to record and display the value of the input and output variables in each pair of experiments. In addition, a permanent button “Main Steps” was prepared to remind learners of the main steps in an experiment. The simulation program was written in such a way that it registered learners’ manipulations during the learning processes and wrote a log-file for each subject.
 
Research design
 
In order to investigate the effects of peer collaboration and questioning-explanation prompt on simulation-based discovery learning, a 2 (pair/single) 5 2 (with/without explanatory prompt) 5 2 (floating/sinking) mixed design was adopted to compare the learning effectiveness of subjects under four conditions: Prompt Pairs, No-Prompt Pairs, Prompt Singles, and No-Prompt Singles. The effectiveness in discovering the underpinning rules for floating and sinking objects was contrasted by including floating/sinking as a within-subjects factor. Considering the influence of learners' pre-existing physics performance, the achievements on the most recent final physics exam was collected and used as a covariate when conducting ANOVAs. Log-files were used to analyze how learners designed simulated experiments and made the discovery.
 
Subjects
 
Subjects were 44 eighth grade students from a junior high school in Beijing. The students ranged in age from 13 to 15, with an average age of 13.88. Twenty-one were boys and twenty-three were girls. The subjects were randomly distributed into four groups: Prompt Pairs (n=12), No-Prompt Pairs (n=12), Prompt Singles (n=10), and No-Prompt Singles (n=10). The overall evaluation of the subjects provided by the physics teacher indicated no significant difference among the four groups in respects of their physics knowledge and experience with computers.
 
Assessments
 
The effectiveness of the discovery learning was assessed from the following perspectives:
Discovered rules. The subjects were required to summarize their discovery when they finished the experiment. A full score of 3 was given if a subject had clearly stated the exact rules for floating and sinking objects. A score of 2 indicated that the subject had made a general description (i.e. "Mass and volume is relevant to the size of upthrust.") without considering the differences between floating and sinking objects. Score 1 was given when only one of the related factors (volume or mass) was mentioned. Score 0 meant that a subject had drawn a thoroughly wrong or irrelevant conclusion.
Intuitive understanding. Nine multiple-choice questions were designed to measure learners’ intuitive understanding, which is regarded as an important goal in scientific discovery learning (see de Jong et al., 1999). Each item depicted a pair of objects with different combinations of shapes, masses, and volumes. Learners were asked to predict how the upthrusts would compare in size. Four of the items were related to upthrusts on floating objects and five were geared to upthrusts on sinking ones.
Variable control skills. Variable control (“Change one thing at a time”) is an important principle in scientific experiment. Unfortunately, learners are often found to vary many variables in one experiment (Glaser et al., 1992; Lin & Lehman, 1999). This test included six items to evaluate learners’ variable control strategies in scientific experiment by providing the cases of experiment designs (either well-controlled or with flaws) for learners to make critical evaluations and justifications. Similar methods have been frequently used in the researches of variable control skills (see Ross, 1988).
Evaluation of learners' experiments. Surrounding the principle of variable control, we evaluated learners' experiments conducted during the learning process, which had been recorded in their log-files. In specific, we evaluated how learners had made focused examinations of the three variables under floating and sinking conditions. We identified a pair of experiments as having undergone a “focused examination” of certain variable (shape, mass or volume) if that variable was the only variable that was varied in that pair of experiments. For each variable, a full score of 2 was given when it had been examined by at least two pairs of experiments at different levels of the controlled variables (an example is shown in Appendix). Score 1 indicated that the variable had been examined by only one pair of experiments or by more than one pair of experiments but at constant levels of controlled variables. Sequentially, score 0 meant that no experiment had been focused on this variable at all. Average scores across the three variables were used to evaluate learners' focused examinations under floating and sinking conditions respectively. This index reflects the distribution of well-controlled experiments across the features and conditions, hence indicates learners' logical and systematic searches in the experiment space during discovery processes.
 
Procedure
 
The study took place in a computer laboratory equipped with 50 networked Pentium II computers. The subjects were required to accomplish the following tasks in pairs or individually:
1.      Warm-up. Subjects worked with a tutorial version of the simulation program. Three experimenter was present to answer their questions regarding the program. This stage lasted approximately 10 minutes.
2.      Problem presentation. The subjects were asked to explore which one or more of the factors among shape, mass and volume influenced the upthrust on an object either floating or sinking in water. A brief description of the problem was available on the top-right corner of the screen throughout the discovery process.
3.      Exploration. All subjects were reminded that their task was to discover the rules on the basis of sufficient evidence through simulated experiments. Subjects in pairs were required to accomplish the task collaboratively and were told that they were to be evaluated as a group. For the Prompt Pairs and Prompt Singles, explanatory prompt cards were put in front of their monitors, reminding them the questions they can ask their partners or ask themselves during the exploration processes. These questions included: (1) what's the objective of this experiment? (2) How should the experiment be designed to achieve this objective? (3) What does this experiment mean? (4) How can the experiment support this conclusion? (5) I think the result doesn't support this conclusion, because… (6) I think this is not the exactly right conclusion, because… The subjects were required to summarize their discovery at the end of this stage.
4.      Posttest. The posttests of intuitive understanding and variable control skills were administered immediately after the completion of the exploration in written format. A total of 30 minutes was allotted for this session.
 
RESULTS
 
Evaluation of the rules discovered by different groups
 
Table 1. The discovery outcomes of different groups

Groups
M
SD
Prompt Pairs
1.500
.905
No-Prompt Pairs
1.667
.492
Prompt Singles
1.200
1.033
No-Prompt Singles
1.400
.843
Total
1.455
.820

 
Table 1 shows the evaluation results of the discovered rules of different groups. There were relatively few students who had clearly stated the exact rules for floating and sinking objects. The univariate ANOVA using collaboration and explanatory prompt as fixed factors and learners' recent physics scores as covariate indicated that collaboration had significant main effect on the discovery outcome (F(1, 36)=5.786, p=.021). Pairs have discovered the underpinning principles more successfully. A clear interaction was observed between collaboration and the covariate (F(1, 36)=5.410, p=.026). In order to make clear the pattern of the interaction, we aggregated the subjects into two groups according to the median of their physics scores. Simple effect analysis revealed that collaboration had significant positive effect among lower achievement learners (F(1, 41)=6.07, p=.018), whilst had no significant influence to higher achievement learners. The overall attainment of the prompted learners displayed to be lower than learners without prompts. However no significant main effect or interaction was observed for explanatory prompt (p>.10) in the statistic analysis.
 
The effects of collaboration and explanatory prompt on intuitive understanding test
 
Table 2 indicates the means and standard deviations of learners' intuitive understanding of the upthrusts on floating and sinking objects.
Table 2. The means and standard deviations of intuitive understanding test

 
Groups
Floating
Sinking
M
SD
M
SD
Prompt Pairs
1.000
.603
2.000
1.045
No-Prompt Pairs
2.333
1.303
1.250
1.288
Prompt Singles
1.100
.738
1.500
.850
No-Prompt Singles
1.100
.994
2.100
.738
Total
1.409
1.085
1.705
1.047

 
The repeated measures MANOVA was implemented using collaboration and explanatory prompt as between-subjects factors, floating/sinking as within-subjects factor, and learners' physics scores as covariate. There was a marginally significant main effect for collaboration (F(1,36)=3.047, p=.089), indicating that Pairs outperformed Singles on intuitive understanding test. A marginally significant interaction was also observed between collaboration and learners' physics achievements (F(1, 36)=3.303, p=.077). Simple effect analysis revealed that collaboration had significant positive effect to higher achievement learners (F(1, 41)=8.000, p=.007), whilst had no significant effect to lower achievement ones (p>.10). No notable main effect or interaction was observed for explanatory prompt (p>.10).
 
Analysis of learners' variable control skills
 
 Table 3. The means and standard deviations of learners' variable control skills
Groups
M
SD
Prompt Pairs
4.75
1.42
No-Prompt Pairs
4.83
1.03
Prompt Singles
3.30
2.06
No-Prompt Singles
4.40
1.58
Total
4.36
1.60
 
Table 3 shows the scores of variable control skills by different groups of learners. The ANOVA using collaboration and explanatory prompt as fixed factors revealed that there was a significant main effect for collaboration (F(1, 40)=4.098, p=.050). Pairs demonstrated higher performance on variable control skills. Again, no notable main effect or interaction was observed for explanatory prompt (p>.10).
 
Process analysis


 
Table 4. Learners' focused examination of upthrust during discovery processes

 
 
Groups
Focused investigation of floating objects
Focused investigation of sinking objects
M
SD
M
SD
Prompt Pairs
.67
.24
.06
.15
No-Prompt Pairs
.58
.32
.17
.19
Prompt Singles
.83
.33
.00
.00
No-Prompt Singles
.78
.19
.00
.00
Total
.71
.27
.06
.13

 
We evaluated learners' experiments conducted during discovery processes using the procedure described in the method section. The evaluation results concerning learners' focused examinations of the related factors under floating and sinking conditions were reported in Table 4. The repeated measures MANOVA using floating/sinking as within-subjects factor and collaboration and prompts as between-subjects factors indicated that there was a significant main effect for the within-subjects factor, showing that learners had made more focused investigations concerning the upthrust on floating objects (Sphericity Assumed F (1, 36)=64.342, p=.000). A marginally significant interaction was observed between collaboration and floating/sinking condition (Sphericity Assumed F (1, 36)=3.293, p=.095) (Figure 2). Compared to single learners, Pairs had made more focused investigations on the upthrust of sinking objects, relatively fewer examinations for floating objects. Actually, no Singles had ever made any focused experiment for sinking objects at all.
Figure 2. Focused examinations of floating/sinking objects made by pair and single learners
 
DISCUSSION
 
This study examined the effects of peer collaboration and questioning-explanation prompt on scientific discovery learning among secondary students. We found that: (1) Peer collaboration had notable effect on the overall discovery outcomes and intuitive understanding generated from the discoveries; (2) There were clear interactions between collaboration and learners' physics performance on the overall discovery outcomes and intuitive understanding test but with different trends; (3) Collaboration had positive effects on learners' variable control skills and their focused design of experiments; (4) No significant main effect or interaction was revealed concerning explanatory prompt.
 
The effect of peer collaboration on simulation-based discovery learning
 
The positive effects of peer collaboration on scientific discovery learning were clearly observed on the overall discovery outcomes, the intuitive understanding test, as well as the variable control skill test. Pairs had discovered the rules more successfully, and generated more insightful intuitive understandings on the relationships between the investigated factors and the size of upthrust under floating or sinking conditions. Hence they could more successfully predict the comparison of the upthrusts on each pair of objects depicted in the intuitive understanding test. These results support the consistent findings that collaborative learners outperformed single learners in accomplishing discovery learning or reasoning tasks (Okada & Simon, 1997; Teasley, 1995). It was revealed by Okada & Simon (1997)'s protocol analysis that an important reason why Pairs performed better than Singles is because Pairs participated in explanatory activities such as entertaining hypotheses, talking about alternative ideas, and considering justification more often than Singles. It is not because Pairs simply had twice as great a chance as Singles of getting the right hypothesis, even without active interaction.
Clear interactions were found between collaboration and learners' physics performance on the overall discovery outcomes and intuitive understanding test, however with different patterns. On the evaluation of the discovered rules, collaboration had significant positive effect to lower achievement learners, whilst had no significant influence to higher achievement ones. On the intuitive understanding test that involved more complex cognitive tasks, collaboration demonstrated significant positive effect to higher achievement learners, but had no significant effect to lower achievement ones. The peer collaboration might be helpful to both low and high achievement learners. However, its effectiveness among low and high achievement learners may manifest in different levels and aspects of learning.
As far as the discovery process is concerned, in this study, collaboration manifested a positive effect on learners' variable control skills in scientific experiments. Also the index of focused examinations of the related features used in this study reflected the distribution of learners' focused experiments across the features and conditions, hence indicated learners' systematic searches in the experiment space during discovery learning processes. The interaction between collaboration and experiment condition (floating/sinking) manifested that Pairs had made more focused experiments for upthrusts on sinking objects. Since the default values of the volumes and masses of the objects in the simulation software determine a density (mass/volume) lower than that of water, the simulation starts with objects floating in water. No deliberate instruction was offered to guide the subjects to change the comparisons of masses and volumes to make objects submerge in water, taking both floating and sinking conditions into account. In the result, the subjects conducted significantly more focused examinations for floating objects than sinking ones. Pairs made more focused experiments for upthrusts on sinking objects, whilst Singles made no focused experiment for sinking objects at all. The result implies that peer collaboration can help learners make more systematic searches in the experiment space and design more crucial and informative experiments during scientific discovery learning. This disagrees with Okada & Simon (1997)'s finding that there was no significant difference between Pairs and Singles on the informativeness and systematicy of experiments conducted during discovery processes.
 
The influence of the questioning-explanation prompts
 
The explanatory prompts used to elicit reciprocal questioning and explanation activities manifested no significant effect in this study. This is inconsistent with Okada & Simon (1997)'s conclusion that explanatory activities facilitated by peers' requests had important influence to the effect of discovery, Coleman (1998)'s report about the effect of scaffolded explanation-based approach, as well as King's studies showing that "Guided Peer Questioning" could help learners in acquiring and understanding knowledge (King, 1992, 1994; King & Rosenshine, 1993). In the present study, the treatments concerning questioning-explanatory prompt included the use of prompt cards reminding the questions learners could request each other or ask themselves, and the special instruction to Prompt Pairs and Prompt Singles asking them to give attentions to the questions on the prompt cards. However, there was no particular procedure to train and coach these subjects in asking the questions and giving corresponding responses, as had been done in King et al. 's researches. These treatments might not have guaranteed the subjects, who were young learners from a middle school, to follow the explanatory prompt and perform intensive questioning-explanation activities. This might be one of most important reasons why explanatory prompt didn't display any expected significant effect in this study. For Chinese students who are not used to the ways of reciprocal activities in regular classroom settings, they might need particularly more guiding and coaching in questioning-explanation activities. More powerful and clear guidance should be adopted in the future studies to investigate the effect of reciprocal questioning-explanation activities as well as its interaction with the mode of collaboration in scientific discovery learning.
 
CONCLUSION
 
Nowadays, a growing number of studies have focused on scientific discovery learning based on computer simulation. However, most previous studies in this field focus on individual discovery learning without considering the perspective of social interaction. This study converged scientific discovery learning with collaborative learning and made a preliminary experiment in the context of secondary science learning. Peer collaboration was found to have prominent effect on the processes and outcomes of discovery learning. It can help learners make more systematic searches in the experiment space and design more crucial and informative experiments during scientific discovery learning. The influence of the questioning-explanation prompt was less clear, hence need to be addressed in depth in future studies.
 
REFERENCES
 
Banggert-Drowns, R., Kulik, J., & Kulik, C. (1985). Effectiveness of computer-based education in secondary schools. Journal of Computer-Based Instruction, 12 (1), 59-68.
Carlsen, D. D. & Andre, T. (1992). Use of a microcomputer simulation and conceptual change text to overcome students’ preconceptions about electric circuits. Journal of Computer-Based Instruction, 19, 105-109.
Coleman, E. B. (1998). Using explanatory knowledge during collaborative problem solving in science. The Journal of the Learning Sciences, 7 (3&4), 387-427.
de Jong, T. and van Joolingen, W. R. (1998) Scientific discovery learning with computer simulations of conceptual domains. Review of Educational Research, 68, 179-201.
de Jong, T., Martin, E., Zamarro, J., Esquembre, F., Swaak, J., and van Joolingen, W. R. (1999) The integration of computer simulation and learning support: An example from physics domain of collisions. Journal of Research in Science Teaching, 36, 597-615.
Freedman, E. G. (1992). Scientific induction: Individual versus group processes and multiple hypotheses. In Proceedings of the Fourteenth Annual Conference of the Cognitive Science Society. Hillsdale, NJ: Erlbaum.
Glaser, R., Schauble, L., Raghavan, K., and Zeitz, C. (1992) Scientific reasoning across different domains. In E. de Corte, M. C. Linn, H. Mandle, and L. Verschaffel (1992) Computer –Based Learning Environments and Problem Solving (pp. 345-373). Berlin: Springer- Verlag
Gorman, M. E. (1986). How the possibility of error affects falsification on a task that models scientific problem solving. British Journal of Psychology, 77, 85-96.
Gorman, M. E., Gorman. M. E., Latta, R. M., & Cunningham, G. (1984). How disconfirmatory, confirmatory and combined strategies affect group problem solving, British Journal of Psychology, 75, 65-79.
King, A. & Rosenshine, B (1993). Effects of guided cooperative questioning on children's knowledge construction. Journal of Experimental Education, 61, 127-148.
King, A. (1992). Facilitating elaborative learning trough guided student-generated questioning. Educational Psychologist, 27, 111-126.
King, A. (1994). Guiding knowledge construction in the classroom: Effects of teaching children how to question and how to explain. American Educational Research Journal, 30, 338-368.
King, A. (1999). Discourse patterns for mediating peer learning. In: O'Donnell, A. & King, A. (Eds.), Cognitive perspectives on peer learning. Lawrence Erlbaum Associates, Inc.
Klahr, D. and Dunbar, K. (1988) Dual space search during scientific reasoning. Cognitive Science, 12, 1-48.
Leutner, D. (1993) Guided discovery learning with computer-based simulation games: Effects of adaptive and non-adaptive instructional supports. Learning and Instruction, 3, 113-132.
Lewis, E. L., Stern, J. L., and Linn, M. C. (1993) The effect of computer simulations on introductory thermodynamics understanding, Educational Techology, 33, 45-58.
Li, M., & Chen, Q. (2000). Comparison of different computer application modes in instruction of primary school geometry. Chinese Journal of Psychology, 32(1), 75-81.
Lin, X. & Lehman, J. (1999). Supporting learning of variable control in a computer-based biology environment: Effects of prompting college students to reflect on their own thinking. Journal of Research in Science Teaching, 36 (7), 837-858.
Njoo, M. & de Jong, T. (1993) Exploratory learning with a computer simulation for control theory: learning processes and instructional support. Journal of Research in Science Teaching, 30, 821-844.
Okada, T. & Simon, H. A. (1997). Collaborative discovery in a scientific domain. Cognitive Science, 21, 109-146.
Pilkington, R & ParkerJones, C. (1996). Interacting with computer-based simulation: The role of dialogue. Computers & Education, 27 (1), 1-14.
Quinn, J. & Alessi, S. (1994) The effects of simulation complexity and hypothesis generation strategy on learning. Journal of Research on Computing in Education, 27, 75-91.
Ross, J. A. (1988). Controlling variables: A meta-analysis of training studies. Review of Educational Research, 58(4), 405-437.
Shute, V. J. & Glaser, R. (1990) A large scale evaluation of an intelligent discovery world: Smithtown. Interactive Learning Environments, 1, 51-77.
Tabak, I., Smith, B. K., Sandoval, W. A., & Reister, B. J. (1996) Combining general and domain-specific strategic supports for biological inquiry. In C. Frasson, G. Gauthier, & A. Lesgold (1996) Intelligent Tutoring Systems (pp. 288-297) Berlin: Springer-Verlag.
Teasley, S. D. (1995). The role of talk in children's peer collaborations. Developmental Psychology, 31, 207-220.
Webb, N. M. (1989). Peer interaction and learning in small groups. International Journal of Educational Research, 13, 21-39.
Zhang, J. (2000). Scientific discovery learning based-on computer simulation: The inner conditions and the facilitative learning support. Doctoral dissertation submitted to Beijing Normal University.
Zhang, J., Chen, Q., & Reid, J. (Aug. 2000). Simulation-based Scientific Discovery Learning: A Research on the Effect of Experimental Support and Learners’ Reasoning Ability. In: D. Benzie & D. Passey (Eds.) Proceedings of Conference on Educational Use of Information and Communication Technology (WCC2000). Beijing: Publishing House of Electronic Industry, 2000.
 
Appendix: Focused examination of volume: An example that was given full score

No.
Shape
Mass (g)
Volume (cm3)
Upthrust (N)
1 left 
1 right
Ball
Ball
18.00
18.00
30.00
20.00
0.18
0.18
3 left 
3 right
Box
Box
5.00
5.00
10.00
400.00
0.05
0.05

------分开线----------------------------
标签(Tag):Discovery
------分开线----------------------------
保举内容
猜你感兴趣