Allan Collins

Situated Cognition and the Culture of Learning

Many teaching practices implicitly assume that conceptual knowledge can be abstracted from the situations in which it is learned and used. This article argues that this assumption inevitably limits the effectiveness of such practices. Drawing on recent research into cognition as it is manifest in everyday activity, the authors argue that knowledge is situated, being in part a product of the activity, context, and culture in which it is developed and used. They discuss how this view of knowledge affects our understanding of learning, and they note that conventional schooling too often ignores the influence of school culture on what is learned in school. As an alternative to conventional practices, they propose cognitive apprenticeship (Collins, Brown, & Newman, in press), which honors the situated nature of knowledge. They examine two examples of mathematics instruction that exhibit certain key features of this approach to teaching.

A spreading-activation theory of semantic processing.

This paper presents a spreading-acti vation theory of human semantic processing, which can be applied to a wide range of recent experimental results. The theory is based on Quillians theory of semantic memory search and semantic preparation, or priming. In conjunction with this, several of the miscondeptions concerning Qullians theory are discussed. A number of additional assumptions are proposed for his theory in order to apply it to recent experiments. The present paper shows how the extended theory can account for results of several production experiments by Loftus, Juola and Atkinsons multiple-category experiment, Conrads sentence-verification experiments, and several categorization experiments on the effect of semantic relatedness and typicality by Holyoak and Glass, Rips, Shoben, and Smith, and Rosch. The paper also provides a critique of the Smith, Shoben, and Rips model for categorization judgments. Some years ago, Quillian1 (1962, 1967) proposed a spreading-acti vation theory of human semantic processing that he tried to implement in computer simulations of memory search (Quillian, 1966) and comprehension (Quillian, 1969). The theory viewed memory search as activation spreading from two or more concept nodes in a semantic network until an intersection was found. The effects of preparation (or priming) in semantic memory were also explained in terms of spreading activation from the node of the primed concept. Rather than a theory to explain data, it was a theory designed to show how to build human semantic structure and processing into a computer.

Toward a Design Science of Education

We have had many technologies introduced in classrooms all over the world, but these innovations have provided remarkably little systematic knowledge or accumulated wisdom to guide the development of future innovations. Bolt, Beranek & Newman (BBN) Inc. is part of the new Center for Technology in Education located at Bank Street College in New York. A major goal of the Center is to synthesize research on technological innovation, to develop a methodology for carrying out design experiments, to study different ways of using technology in classrooms and schools, and to begin to construct a systematic science of how to design educational environments so that new technologies can be introduced successfully.

A Systems Approach to Educational Testing

Our concern in this paper is with the validity of educational tests when they are employed as critical measures of educational outcomes within a dynamic system. The problem of validity arises if an educational system adapts itself to the characteristics of the outcome measures. We introduce the concept of systemically valid tests as ones that induce curricular and instructional changes in education systems (and learning strategy changes in students) that foster the development of the cognitive traits that the tests are designed to measure. We analyze some general characteristics that contribute to or detract from a testing systems systemic validity, such as the use of direct rather than indirect assessment. We then apply these characteristics in developing a set of design principles for creating testing systems that are systemically valid. Finally, we provide an illustration of the proposed principles, by applying them to the design of a student assessment system. This design example addresses not only specifications for the tests, but also the means of teaching the process of assessment to users of the system.

The Logic of Plausible Reasoning: A Core Theory

Abstract The paper presents a core theory of human plausible reasoning based on analysis of peoples answers to everyday questions about the world. The theory consists of three parts: 1. 1. a formal representation of plausible inference patterns; such as deductions, inductions, and analogies, that are frequently employed in answering everyday questions; 2. 2. a set of parameters, such as conditional likelihood, typicality, and similarity, that affect the certainty of peoples answers to such questions; and 3. 3. a system relating the different plausible inference patterns and the different certainty parameters. This is one of the first attempts to construct a formal theory that addresses both the semantic and parametric aspects of the kind of everyday reasoning that pervades. all of human discourse.

The computer as a tool for learning through reflection

A unique aspect of computers is that they not only represent process but also naturally keep track of the actions used to carry out a given task, so that the process with its trace can become an object of study in its own right. One effect of this can be seen vividly in the sciences, where computers and computational languages have improved our ability to develop and test process theories of complex natural phenomena. Before powerful computers became readily available as scientific tools, process models were expressed in mathematical languages, such as differential equations— languages primarily effective in capturing a static “snapshot” of a process. Computation provided formal languages that are more flexible than mathematics but just as precise. In part because computation is itself dynamic, it provides an ideal medium for representing and testing richer, more varied, and more detailed theories of process. The use of this medium for process modeling has radically changed the nature of many current theories in both the physical and social sciences. Particularly in the arena of the cognitive sciences, computational techniques have proved to be powerful tools for both experimental and theoretical investigations of the mind.

The second educational revolution: rethinking education in the age of technology

This paper drew upon a recent book (Rethinking Education in the Age of Technology) to summarize a number of prospects and challenges arising from the appropriation of digital technology into learning and educational practice. Tensions between traditional models of schooling and the affordances of digital media were noted, while the promise of these technologies for shaping a new system of education was reviewed. It was argued that new technology brings radical opportunities but also significant challenges. The urgency of seeking a coherent model for the future of education in a technological age was stressed.

REASONING FROM INCOMPLETE KNOWLEDGE

Publisher Summary It does not trouble people much that their heads are full of incomplete, inconsistent, and uncertain information. With little trepidation they go about drawing rather doubtful conclusions from their tangled mass of knowledge, for the most part unaware of the tenuousness of their reasoning. The very tenuousness of the enterprise is bound up with the power it gives people to deal with a language and a world full of ambiguity and uncertainty. This chapter describes this kind of human reasoning in terms of how a computer can be made to reason in the same illogical way. It presents SCHOLAR, a computer program whose knowledge about the world is stored in a semantic network structured like human memory. SCHOLARs aim is to teach people by carrying on a tutorial dialog with them. The distinction between explicit and implicit knowledge also exists in SCHOLAR.

Facilitating retrieval from semantic memory: The effect of repeating part of an inference☆

Abstract In Collins and Quillian (1969) we found evidence that people decide whether simple sentences are true or false by using inferences. For instance, a sentence like ‘A canary can fly’ apparently was confirmed by inference from the two facts that a canary is a bird and that birds can fly. If so, then this has a possible implication for reaction time (RT) to such sentences presented in succession. Prior exposure to one sentence should reduce RT to a second sentence whenever the same fact is involved in confirming both sentences. For example, prior exposure to ‘A canary is a bird’ should reduce RT to ‘A canary can fly’ more than to ‘A canary can sing’, since we assume that no inference is used to confirm the latter sentence. In total eight RT difference predictions were made for various kinds of sentence pairs, and all eight of these predictions held. Two possible models could explain these results.

Does category size affect categorization time

Landauer and Freedman (1968) found that it takes longer to categorize object names (e.g., collie or tulip) into larger categories (e.g., animal) than into smaller categories (e.g., dog). They attributed this effect to category size. But we suspected the effect was caused by the fact that their different-size categories were always nested. Two experiments, one a partial replication of their first experiment, were conducted to disentangle these two possible explanations. Several factors were found to affect categorization time: nesting, whether or not subcategories were utilized by S s in the categorization task, and semantic relatedness or confusability. There was no evidence that larger categories, in and of themselves, required longer categorization times than smaller categories.