Encyclopedia of Science Education

Living Edition
| Editors: Richard Gunstone

Information Processing and the Learning of Science

  • Richard WhiteEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-94-007-6165-0_113-1

Keywords

Science Education Turing Machine Serial Processing Meaningful Form Information Processing Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

In the second half of the twentieth century, cognitivism replaced behaviorism as the dominant paradigm in psychology. Where behaviorism avoided speculation about unobservable processes such as thinking and reflection and concepts such as understanding, mind, and will, cognitive psychology postulated models of mental processes that are involved in learning and problem solving. These models of information processing derived from digital computers, which became common at this time. The brain was pictured as a machine, like the computer, which accepted input via the senses, processed it into meaningful form, stored it, and was able to retrieve it on demand.

The classic model was of serial processing, as in the first Turing machines, where each incoming bit of information is dealt with by a central processing unit following a predetermined algorithm with the outcome placed in its turn in a memory slot, in fixed sequence. The model had shortcomings: serial processing is too slow to explain the brain’s rapidity in managing complex performances; it is brittle, in that damage to any one step destroys the whole process, while the brain is robust, able to fill in gaps in input and to recognize and remedy minor errors; it is inflexible, following a rigid sequence of operations and so cannot cope with novel or messy input, something that the brain does quite well. In an attempt to reduce these shortcomings, parallel processing which handles multiple tasks simultaneously, a feature of later computers, became part of the model. Nevertheless, the model remains a model, of a machine, imperfectly matching the subtlety and complexity of the brain. As has been pointed out frequently, computers cope easily with tasks that the brain finds challenging, while the brain readily solves problems that defeat even the most powerful computers. Therefore, application of information processing to the subtleties and complexities of teaching any subject, and perhaps especially of science in its own complex depiction of the natural world, will not provide a comprehensive guide to classroom action. That does not make the model totally useless, however, just as the inabilities of the wave and particle models to explain all the phenomena of light do not make them useless in physics. There are aspects of the model that can inform teaching.

In its basic form, information processing depicts learning as a sequence of operations: selection, translation, storage, and retrieval White (1998).

Selection. As do all living organisms, humans react to stimuli coming from their surroundings. They are not, however, aware of every single stimulus. They attend to things that they judge important. While some of this judgment is instinctive, as in reaction to fast-moving objects near the head, much depends on previous learning. People observe stimuli that seem important to them at the time and ignore the rest. In this they differ from a computer, which accepts all of the input it receives. Thus, learning is not a linear process, but an interactive or recursive one.

Teachers manage students’ selection by pointing out the features that matter of an object or event, such as the parts of a flower or a chemical reaction. Their success depends on whether the students accept the teacher’s assertion that these features are worth attending to, and that acceptance in turn depends on a complex mix of the student’s needs and purposes and on existence or absence of competing stimuli (such as behavior of other students). In other words, the students’ motivations for learning, liking for science and for individual topics within it, and relation with the teacher affect what they attend to. So does their physical state: as teachers soon realize, tiredness or illness inhibit alertness and consequently selection.

Translation. According to the information processing model, learners translate observed stimuli into a meaningful form, initially held in a short-term memory buffer of limited capacity. In early childhood children acquire many skills of translation, rapidly increasing their capacity to make sense of their environment. Translation remains important in later learning, where students might need guidance to understand what they are seeing or are told. For instance, differentiating a landscape into features such as horsts, peneplains, and faults has to be learned. An important consequence is that the experienced and knowledgeable observer sees a smaller number of units than the tyro, who can be overwhelmed by the detail. If too much information comes in, the short-term buffer is overloaded, and much is lost. Teachers therefore need to teach students how to see, how to form the information into a smaller number of larger meaningful units. This needs care, for students might come to only ever see what the teacher directs them to see, and so might miss curious elements.

Information that has been processed into meaning can follow two paths. In one, it is erased by newer incoming stimuli and immediately forgotten. In the other, it remains in working memory until it is either processed further into long-term memory or is lost. Thus, people can recall events for some days, but unless the event is linked with existing knowledge for some reason, after a few days or even shorter the memory is lost.

Storage. Some of what is translated is stored as knowledge in long-term memory. Theorists have proposed various divisions of this knowledge, such as semantic versus episodic, propositional versus procedural, declarative versus non-declarative, words and images, propositions, algorithms, images, episodes, and strings. These divisions may be important in science education, since different forms of knowledge may be learned differently and therefore need to be taught in different ways; because individual learners may have idiosyncratic preferences and vary in the ease with which they acquire each form; and because different mixes of forms will result in different qualities of understanding.

Storage in long-term memory is a function of individual preference of the learner and of actions of the teacher: what the learner is interested in and perceives as meeting a need and what the teacher emphasizes as important or makes interesting. Storage can be as unorganized, unconnected elements, or as a highly interlinked network. Level of understanding is a consequence of extent of linking. Effective teaching is likely to involve making clear connections between individual pieces of knowledge within a topic and across topics. Thus, a student who perceives commonalities between gravitational, electric, and magnetic fields will have better understanding than one for whom these are unrelated topics.

Retrieval. Models of information processing do not prohibit reorganization of knowledge subsequent to its acquisition, through reflection that creates perception of new links, but essentially they present a static notion of knowledge: it is acquired and remains in long-term memory more or less in its original form, whereas human memory is more dynamic. Where in a computer what is retrieved from a memory cell is what is stored there, in humans retrieval involves reconstruction, in which factors such as context affect the recall. In one context, one might recall knowledge of motion as Aristotelian and in another as Galilean/Newtonian or a tomato as a fruit in a biology class but as a vegetable in the kitchen. The influence of context on reconstruction is largely responsible for the alternative conceptions that became the focus of much research in science education in the last third of the twentieth century. Students who learned a scientist’s explanation for a phenomenon often maintained a non-science explanation as well, offering one explanation in one context and the other in another.

Metalearning. The notion of metalearning came later than models of information processing but fits readily with them. Metalearning refers to knowledge of processes of learning, awareness of their application, and ability and willingness to control them. Thus, metalearning can refer to conscious and deliberate selection, translation, storage, and retrieval (Brown 1987; Georghiades 2004).

While metalearning is fundamental to quality of learning in all subjects, it is particularly important in science education, which presents students with explanations of the complex natural world that may be at odds with beliefs that they have acquired through folk lore or through unguided experiences. Resolution of such differences is unlikely without conscious management of learning processes. Training students in metalearning is an advanced teaching skill. For accounts of practical programs, see Baird and Northfield (1992) and Adey and Shayer (1994).

Cross-References

References

  1. Adey P, Shayer M (1994) Really raising standards. Routledge, LondonGoogle Scholar
  2. Baird JR, Northfield JR (1992) Learning from the PEEL experience. Monash University, MelbourneGoogle Scholar
  3. Brown A (1987) Metacognition, executive control, self-regulation and other more mysterious mechanisms. In: Weinert FE, Kluwe RH (eds) Metacognition, motivation, and understanding. Erlbaum, HillsdaleGoogle Scholar
  4. Georghiades P (2004) From the general to the situated: three decades of metacognition. Int J Sci Educ 26(3):365–383CrossRefGoogle Scholar
  5. White RT (1988) Learning science. Blackwell, OxfordGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Monash UniversityCaulfield EastAustralia