Jean M. Mandler

Remembrance of things parsed: Story structure and recall☆

Abstract An analysis of the underlying structure of simple stories is presented. It is claimed that this type of representation of stories is used to form schemata which guide encoding and retrieval. A type of tree structure containing basic units and their connections was found to be adequate to describe the structure of both single and multi-episode stories. The representation is outlined in the form of a grammar, consisting of rewrite rules defining the units and their relationships. Some transformational rules mapping underlying and surface structures are discussed. The adequacy of the analysis is first tested against Bartletts protocols of “The War of the Ghosts.” Then a developmental study of recall is presented. It is concluded that both children and adults are sensitive to the structure of stories, although some differences were found. Finally, it is suggested that the schemata used to guide encoding and recall are related but not identical and that retrieval is dependent on the schemata operative at the time of recall.

How to Build a Baby: II. Conceptual Primitives.

A mechanism of perceptual analysis by which infants derive meaning from perceptual activity is described. Infants use this mechanism to redescribe perceptual information into image-schematic format. Image-schemas create conceptual structure from the spatial structure of objects and their movements, resulting in notions such as animacy, inanimacy, agency, and containment. These earliest meanings are nonpropositional, analogical representations grounded in the perceptual world of the infant. In contrast with most perceptual processing, which is not analyzed in this fashion, redescription into image-schematic format simplifies perceptual information and makes it potentially accessible for purposes of concept formation and thought. In addition to enabling preverbal thought, image-schemas provide a foundation for language acquisition by creating an interface between the continuous processes of perception and the discrete nature of language.

How to build a baby: On the development of an accessible representational system ☆

Abstract The notion of a sensorimotor stage in infancy is called into question. First, some of the recent experimental literature on cognitive development in infancy is examined to determine the kinds of representational capacity that these data require. It is concluded that most of the recent work on perceptual development and the object concept in infancy is compatible with the notion of a sensorimotor stage but that other work showing imitation, motor recognition, the acquisition of manual signs, and recall of absent objects is not, requiring, instead, a conceptual form of representation. Such a system is apparent early in development. It is suggested that there is a viable alternative to Piagets theory that conceptual representation consists of a transformation of sensorimotor schemas into a new, more advanced code. It is proposed that an accessible conceptual system develops simultaneously and in parallel with the sensorimotor system, with neither system being derivative from the other. It is further proposed that the mechanism by which infants encode information into an accessible system consists of a process of perceptual analysis.

Concept Formation in Infancy.

Four experiments investigated conceptual categorization in 7- to 11-month-old infants. Experiments 1 and 2 showed that 9-and 11-month-olds differentiated the global domains of animals and vehicles. Within the animal domain no subcategorization was found: the infants did not differentiate dogs from fish or from rabbits. Within the vehicle domain infants differentiated cars from both airplanes and motorcycles. Experiment 3 showed similar, although weaker, categorization for 7-month-olds. Experiment 4 showed that categorization of animals and vehicles was unaffected by degree of between-category similarity. Birds and airplanes were treated as different even though the exemplars from both categories had similar shapes, including outstretched wings, and were of the same texture. These data, showing global differentiation of animals and vehicles, with lack of differentiation of “basic-level” categories within the animal domain, contrast with data from studies designed to assess perceptual categorization. Even younger infants differentiate various animal subcategories perceptually. However, the results presented here suggest that infants may not respond to such perceptual differences as being conceptually relevant.

Perceptual and Conceptual Processes in Infancy

It is suggested that we must distinguish 2 types of object categorization in infancy. One is perceptual categorization, which is an automatic part of perceptual processing that computes the perceptual similarity of one object to another. It creates perceptual schemas of what objects look like. The other is conceptual categorization, which is based on what objects do. It consists of the redescription of perceptual information into conceptual form, particularly the paths that objects take and the interactions among them. This process creates the notion of kinds, such as animals, plants, vehicles, and furniture. The similarity in this kind of categorization is of roles in events, not the physical appearance of the objects. Several differences between the 2 types of categories are discussed, of which the most important is the different functions they serve. Perceptual categories are used for object identification; conceptual categories control inductive inference. Experimental results are described showing that because early conceptual categories tend to be global in scope, the inductive generalizations based on them are global in scope as well.

Separating the sheep from the goats: Differentiating global categories ☆

Abstract The nature of the conceptual categories that children have developed in the second year was studied in a series of experiments using an object-manipulation task. In the first two experiments, it was shown that by 18 months children have developed global conceptual categories of animals and vehicles without yet clearly differentiating basic-level categories within these domains. The basic-level categories were tested by using a series of contrasts: a low degree of contrast was provided by presenting the children with dogs versus horses and with cars versus trucks. A moderate degree of contrast consisted of dogs versus rabbits and cars versus motorcycles. A high degree of contrast consisted of dogs versus fish (or birds) and cars versus airplanes. A domain-level contrast of animals versus vehicles was included as well. From 18 to 30 months the children tended to respond categorically only on the global domain-level contrast and on the high-contrast basic-level distinctions. Not until 30 months did the children consistently differentiate the low and moderate basic-level contrasts. Experiment 3 replicated the finding of global animal and vehicle categories, using the widest possible range of exemplars. Experiment 4 extended the study of global categorization to the domains of plants, furniture, kitchen utensils, tools, and musical instruments. Global categorization was found for plants, furniture, and kitchen utensils, but not for tools and musical instruments. Experiment 5 found little evidence for basic-level categorization of plants, and only suggestive evidence for basic-level categorization in the domains of furniture and utensils. The data demonstrate the presence of a number of global conceptual categories from an early age, and suggest that at least in some domains (animals, vehicles, and plants) such categories develop before true basic-level distinctions are made.

A tale of two structures: Underlying and surface forms in stories

Abstract A number of recent models of the structure of stories from the oral tradition have been couched in the form of ‘story grammars’, which describe the types of information that listeners expect to encounter in a story and the organization they tend to impose on that information. In the present paper, we argue that such analyses of story structure may complement analyses of sentence structure as a source of evidence concerning ways in which psychological processing needs constrain the form of complex serial productions. In addition, we present two major extensions of our earlier model of story structure (Mandler and Johnson 1977). First, the use of the base rules to characterize stories which consist of more than one episode is developed, and selection restrictions on the application of these rules are presented. Second, a set of transformational rules is proposed to account for meaning-preserving variations in the surface form of stories, and constraints on the nature of such rules are considered. Finally, we discuss criteria by which the observational and descriptive adequacy of models of story structure may be evaluated. We conclude that more traditional sources of evidence for evaluating grammars, such as intuitive judgments about well-formedness and constituent structure, must be supplemented with evidence based on the relation of the proposed structures to the details of psychological processing.

The cradle of categorization: Is the basic level basic?

Abstract A common assumption in the developmental literature is that the earliest kind of conceptual categories to be formed are basic-level categories. A corollary assumption is that superordinate categories are formed after, and out of, previously acquired basic-level categories. Two experiments using an object-manipulation task explored these assumptions by studying response to a variety of categories in children aged from 12 to 20 months. The first experiment examined responses to basic-level categories (dogs vs. cars),superordinate categories (animals vs. vehicles), and contextual categories (kitchen things vs. bathroom things). At all ages tested, the children performed best on the basic-level categories but, even at 12 months of age, some children were responsive to the superordinate and contextual categories. By 20 months of age, approximately half of the children showed such sensitivity. The second experiment showed that 16- and 20-month-olds differentiated basic-level categories only when the categorical contrasts were taken from different superordinate classes (e.g., dogs vs. cars) and not when the categories were drawn from the same superordinate class (e.g., dogs vs. horses). The data suggest that basic-levels categories are not the first kind of conceptual categories to be formed. Instead, it appears that children may form more global categories, with basic-level differentiation occurring later.

On the psychological validity of story structure

Three expriments tested the psychological validity of the constituent units and sequencing rules of the Mandler and Johnson story grammar. If peoples knowledge about stories reflects such a grammar, then it should have noticeable effects on their processing. The first experiment tested the effects of constituents structure on comprehension and recall by measuring reading and recall times of sentences within and across constituent boundaries. First sentences of constituents were slower to read and recall than second sentences. Various tests and a second experiment showed that these effects were due to story structure rather than to lexical overlap or semantic relatedness per se. The third experience tested the sequencing rules of the grammar by systematically moving constituents away from their normal positions, while at the same time providing them with surface markers to indicate the intended sequence of events. In all cases movements slowed reading time both at the place where the expected constituent was missing and at the place where it acutally occurred. Movements also results in more recall errors. The data support the position that people have incorporated knowledge about the canonical structure of stories which they use slurring processing.

On the coding of spatial information.

Two experiments studied recall of objects and their locations in an intentional-incidental learning paradigm. When studying spatial information, the usual incidental condition is not truly incidental, because subjects often deliberately use locations to help organize objects for recall. Therefore, a true incidental task was devised in which neither objects nor locations were expected to be recalled and for which explicit encoding of locations was irrelevant. There was only a small loss in recall of objects or their locations in a true incidental condition. It was concluded that a great deal of location information is automatically coded into long-term memory storage in the sense that active processing is not required. The data were contrasted with incidental processing of other attributes, such as color. Although adults performed better than children, there were no age-related interactions, indicating similarity of functioning at all ages studied.