The Knowledge Integration Perspective on Learning (Linn, Eylon, & Davis, 2004)

Linn, M.C., Eylon, B., & Davis, E.A. (2004). The Knowledge Integration Perspective on Learning. In M. C. Linn, E. A. Davis, & P. Bell (Eds.), Internet environments for science education (pp. 29–46). Mahwah, NJ: Lawrence Erlbaum Associates.

The authors introduce the knowledge integration perspective on learning, a theoretical framework based on findings from the fields of science learning and cognitive processing.

“Investigations of the systemic, complex nature of science education require a perspective on learning matched to the rich, dynamic classroom experience” (p. 29)

Knowledge integration (p. 30)

  • Defined as “the process of adding, distinguishing, organizing, and evaluating accounts of phenomena, situations, and abstractions.”
  • For other perspectives: Dewey, 1900; Piaget, 1971; Schwab, 1978; Vygotsky, 1962.
  • Other accounts of research and theorizing about this process: Bereiter, 1994; Bransford, Brown, et al., 1999; Bransford, Sherwood, et al., 1990; Carey, 1985; diSessa, 2000; Hawkins, 1991; Krajcik, Marx, et al., 2000; Pea & Gomez, 1992, 1993; Reiser, Copen, et al., 1994; White & Frederiksen, 1998.

“Learners generally possess a fragmented, fragile, or incoherent repertoire of ideas about specific scientific topics (Chi & Slotta, 1993; diSessa, 1988, 1993, 2000; Eylon & Linn, 1988; Slotta, Chi, et al., 1995)… the repertoire expands or contracts depending on the context of investigation–often students respond to questions by narrowing the contexts to which their ideas apply… Learners establish the context of application for their ideas and may modify the context to achieve coherence rather than seeking to make their ideas coherent across contexts” (p. 31). Learners hold a repertoire of incoherent ideas and tend to add new ideas, rather than generate connections to existing ones.

“Knowledge integration involves both developing a repertoire of ideas and sorting out the web of connections among ideas” (p. 30) — i.e., knowledge webs.

Repertoire of ideas refers to the “varied, potentially conflicting views of the same phenomena held by students” (p. 30).

Knowledge webs may form based on surface features or on scientific principles (Chi, Feltovich, et al., 1981; Eylon & Linn, 1988; Reif & Larkin, 1991). They may also come from proximity in learning, or from a critical analysis of similarities and differences (Bagno & Eylon, 1997). They may be based on empirical evidence or conjecture.

“Effective science instruction enables students to make their web of ideas more robust and cohesive” (p. 31). This may be done by helping students develop criteria for adding new connections, make connections across contexts or settings, create knowledge webs that make it easy to incorporate new ideas (p. 31), and distinguish contexts of explanation (microscopic and macroscopic depictions or causal and descriptive models, for example).

Science lens on knowledge integration: The science learning field has emphasized the influence of interpretive, cultural, and deliberate natures of learning on knowledge integration.

Interpretive nature of learning (p.33-38):

-includes study of the spontaneous ideas of learners, the development of scientific understanding, and the acquisition of expertise

1. Students bring diverse ideas to science class that emerge from a process of generating interpretations of experience. See Shonkoff & Phillips, 2000; also, Vygotsky (1962) and Piaget ( 1929/1951) identified the spontaneous ideas students develop about the natural world and speculated about their origins.

2. Recognizing and adding ideas to the repertoire of views does not ensure that new ideas connect to existing ideas–students often contextualize their ideas such that existing and new ideas can exist side by side.

3. Students rarely spontaneously sort out the ideas in their knowledge web and often lack criteria for evaluating new ideas.

Spontaneous ideas of learners

-For research on misconceptions, preconceptions, alternative ideas, and phenomenological primitives formed when processing natural phenomena, see: diSessa, 1988; Driver, Guesne, et aL, 1985; Eylon & Linn, 1988; Gordin et al., 1994; Pfundt & Duit, 1991.

-DiSessa (2000): phenomenological primitives are ideas about the natural world that emerge among most individuals as they observe natural phenomena. Students base these views on observation and interpretation of a relatively limited set of linguistic and experiential information sources.

-“D. Clark (2000) charted the connections learners make among their ideas over time and showed that some students primarily add ideas, whereas others both add and sort out their ideas. Carey (1985), in studying the trajectories of young children as they seek to distinguish living and nonliving things, drew on the distinction between normal and revolutionary science put forth by Thomas Kuhn (1970). Carey concluded that students’ reasoning generally proceeded along the lines of normal science in which students added ideas to elaborate or refine their views rather than along the lines of revolutionary science in which students made paradigm shifts based either on new information or on some developmental accomplishment. Carey did identify individual students who responded to an unusual event by making a revolutionary reorganization of their ideas. In general, learners seem more inclined to add ideas about scientific phenomena than to sort out those that arc inconsistent with others (Carey, 1985; Case, 1985; Inhelder & Piaget, 1970; Siegler, 1978; Singer ct al. , 2000; Smith et al., 1994).”

Development of Scientific Understanding

-Piaget (lnhelder & Piaget, 1970) postulated processes of accommodation and assimilation to explain how learners interpret their experiences. Learners accommodate to some ideas–adding, for example, the idea that the earth is round. Learners assimilate other ideas, adjusting views to incorporate the new information. For example, learners may assimilate information about the earth being round to their perception of the earth as flat by determining that the earth must be round like a pancake (Nussbaum, 1985; Vosniadou & Brewer, 1992) .

-Many developmental theorists argue that reasoning ability develops with age. They point to evidence of similar patterns of reasoning among learners of the same age to support this view (e.g ., Clement, 1988; diSessa, 2000; Driver, Guesne, et al., 1985; Piaget, 1969).

-Biological development of logical reasoning: Piaget (1929/1951) described students as progressing from concrete operations to formal operations. Piaget (Inhelder & Piaget, 1958/1972) argued for adjusting expectations for student scientific understanding based on biological limits on reasoning abilities, suggesting that full logical reasoning ability emerged around age 15. Several recent accounts of students’ scientific reasoning are in accordance (Slotta, Chi, et al., 1995; Vosniadou & Brewer, 1992).

-Some call for instruction in critical thinking. Others argue that critical thinking instruction can only succeed when enacted in the disciplinary context (e.g., Hawkins & Pea, 1987; Linn, Clement, et al., 1989).

-Researchers have argued that reasoning capacity increases with age, and they have distinguished the forms of reasoning possible with greater capacity (e.g ., Case, 1985; Pascual-Leone et al., 1978; Scardamalia, 1977; Siegler, 1978). As students mature, they can handle more information efficiently and therefore solve more complex scientific dilemmas (assumes that learners use similar strategies for solving these problems at different ages and rely primarily on increased processing capacity to perform in a more sophisticated fashion). Important corollary: instruction designed to reduce processing demands can enable students to use more powerful reasoning strategies.

-Although students can use logical reasoning strategics starting at a young age, they often neglect opportunities to do so. Students often add new ideas to their repertoire that contradict existing ideas without reconciling the apparent discrepancies (e.g., Smith et al., 1994). Piaget used discrepant events to enhance learning (Piaget & Inhelder, 1974; Relf & Heller, 1982). However, students may consider the new examples as augmenting rather than contradicting existing ideas and end up with a less cohesive repertoire than they had when they started the course.

Acquisition of expertise

-Experts spend thousands of hours mastering a field and become qualitatively different from novices (e.g., Chase & Simon, 1973; Reif & Larkin, 1991). Individuals add information and organize it to solve more and more complex problems (Anderson, 1982; Eylon & Reif, 1984; Gagne, 1965; Glaser, 1976). Over time, experience with a complex domain such as mechanics enables experts to represent and organize their knowledge around principles and foci that improve reasoning. (Expert physicists organize info around abstract ideas; novices rely on formulas). These expert methods may depend on extensive experience and other knowledge that novices lack — making it difficult to untangle the necessary and sufficient conditions of learning.

-Logical reasoning and disciplinary knowledge go hand in hand in science learning (D. Clark, 2000; Linn & Eylon, 1996; Linn & Hsi, 2000). For example, D. Clark (2000) demonstrated that students started middle school with a broad range of disconnected ideas about thermal equilibrium and only after extensive opportunities to revisit their ideas in a concentrated curriculum were they able to sort out these ideas and develop more robust, cohesive, and normative views. Students often used sophisticated forms of reasoning and creative interpretations to defend their non-normative ideas (sometimes due to empirical evidence).

-Research has shown the difficulty that students face when asked to transfer a concept from one context to another (Bransford, 1979; Gick & Holyoak, 1980; Holyoak, 1985). Students struggle to recognize ideas in new contexts and often fail to connect evidence from experiments conducted at different grain sizes (like microscopic and macroscopic) or from a dynamic rather than a more static perspective (e.g., Linn & Hsi, 2000; Scardamalia & Bereiter, 1991a; Slotta, Chi, et al., 1995).

Cultural nature of learning (pp. 38-42):

-Learning takes place in a complex, influential sociocultural environment. Individuals learn norms for behavior, dress, argumentation, and aspiration from the practices in their cultural setting. Students infer views about scientists, the nature of science, and science learning as well as specific ideas about the natural world by observing their parents, peers, teachers, and others interpreting scientific information (Bell & Linn, 2002; Cole, 1996; Collins, Brown, & Holum, 1991; Davis, 2003b; Dewey, 1900; Lave & Wenger, 1991; Linn, Songer, et al., 1996; Means & Coleman, 2000; Means, Middleton, et al., 1996; Songer & Linn, 1991; Vygotsky, 1962).

-Students may fail to integrate ideas across cultural contexts because generating connections is difficult. For example, textbooks may represent scientific ideas in formulas, animations, visualizations, or experimental results that might not easily connect to each other or to more personally relevant situations (Chi, Feltovich, et al. 1981; Lewis, 1991; Reif & Larkin, 1991) (p. 41).

-Students generate their ideas about science based on their own experiences, yet science learning involves managing a complex set of information from diverse sources, in distinct formats, of varied validity, at different levels of analysis, and at different time scales. Because students have difficulty intcrpreting information from one context in a new setting, students may not test ideas from, for example, media or peers against the ideas they gained in science class” (p. 41).

Deliberate nature of learning (pp. 42-45):

-Students who monitor their progress, reflect on their ideas, and alter their practices based on these ideas learn more than others and develop more durable understanding. Interest in science in general contributes to successful learning, but specific personally relevant problems seem even more effective in promoting knowledge integration. Inquiry learning in which students carry out authentic projects has the potential to elicit deliberate attention to learning (p. 44).

-The deliberate nature of learning concerns the process of translating intentions, goals, values, and expectations into action.

-Learners make decisions about their science learning, including decisions about whether to include a particular idea in the mix they consider for a given problem, whether to explore ramifications of an issue, whether to study hard for a course, whether to take the next more advanced course, and whether to consider a science career. To make these decisions, learners can monitor their progress, rcflect on their own learning and select learning practices.

Monitoring progress

-Students can monitor their own learning by comparing their progress to some set of norms or standards. Often science classes set norms that deter knowledge integration (i.e., requiring memorization of facts: see Linn & Hsi, 2000).

-Students make deliberate decisions about whether or not to monitor their own science learning behavior. They can deliberately limit or expand their opportunities for complex and cohesive scientific knowledge integration.

-Most students need support or guidance to learn to monitor and guide their own science learning (e.g., Scardamalia & Bereiter, 1991a, 1992a) (p. 42).

Reflecting on learning

See A. L. Brown & Campione, 1994; Chi, 2000; Davis & Linn, 2000; Scardamalia & Bereiter, 1992b; Songer, 1996. Learners who practice reflection are more successful (and earn a more durable understanding) than those who are not (Chi, 2000; Songer & Linn, 1991; Linn & Eylon, 2000).

Selecting learning practices

More students engage in knowledge integration about topics that have personal relevance for learners. Students make more connections when science topics connect to their personal interests and concerns (Krajcik, Blumenfeld, Marx, et al., 1994; Linn & Hsi, 2000; Scardamalia & Bereiter, 1991a) (p. 43).

Cognitive process lens on learning

Psychological research on memory and skill acquisition (e.g., Gardner, 1985; Sternberg, 1977. 1985) highlights three cognitivc processes that occur naturally and contribute to learning.

  1. Recognizing new ideas.

    1. Learners spontaneously identify new ideas, often without considering whether the new idea resembles or augments an existing idea. Learners may also dismiss new ideas as identical to currently held views or irrelevant to the context instead of adding them to the repertoire. Often learners cannot distinguish new ideas from ideas that were introduced previously, helping to explain why individuals may develop a repertoire of views (Baddeley & Longman, 1978).
    2. Work on the context effect in memory helps explain why learners isolate their new ideas or connect ideas based on superficial features (R. C. Schank, Fano, et al., 1993/1994).
    3. Research on the role of priming and stereotype threat in social psychology sheds light on how the interpretive nature of learning works in practice. Learners may be primed by context to select ideas not considered in other contexts (p. 44).
  2. Generating connections among ideas. Learners vary in their tendency to generate connections and in the types of connections that they develop. Many learners rarely transfer information from one context to another and therefore rarely generate connections across contexts or across instances of the same idea (Gick & Holyoak, 1980; Sternberg, 1977, 1985). As a result, learners might treat a new context as a new idea or fail to revisit ideas when they reoccur (p. 44-5).
  3. Self monitoring is the process of evaluating ideas in the repertoire and the connections among ideas in the knowledge web and promoting the most promising ideas. Ideally, students would seek a cohesive knowledge web that distinguishes normative ideas, conjectures, and views at diverse levels of analysis and that spans all the contexts where the topic might arise. Learners who monitor their progress end up revisiting their ideas more often and develop more durable understanding (Bjork, 1999). Research on the deliberate nature of the learner connects to psychological research on self-monitoring (e.g., Bjork, 1999; Chi, 2000; diSessa, 2000; Linn & Hsi, 2000) (p. 45).

“All learners do these things to varying degrees and with unique consequences…instructional designers face the challenge of harnessing thesc cognitive processes to improve scientific ideas and to promote a lifelong quest for more cohesive and powerful accounts of natural phenomena” (p. 45).

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: