Comparative Perspectives on the Development of Memory

R. V. Kail, Jr., N. E. Spear

• Publisher: Psychology Press
• Year: 2013
• Print ISBN: 9780898593174
• eBook ISBN: 9781134923656
• Topic: Psychology
• Subtopic: Cognitive Psychology & Cognition
• Index: Psychology

Published in the year 1984, Comparative Perspectives on the Development of Memory is a valuable contribution to the field of Developmental Psychology.

I HISTORICAL OVERVIEWS

1 The Development of Human Memory: An Historical Overview

Robert Kail

Purdue University

Mark S. Strauss

University of Pittsburgh

The purpose of this chapter is to provide a broad overview of current theoretical and empirical work on memory in infants and children. Detailed reviews of this work abound (Brown, 1975, 1979, 1982; Brown, Bransford, Ferrara, & Campione, in press; Hagen, Jongeward, & Kail, 1975; Kail & Hagen, 1977, 1982; Ornstein, 1978; Reese, 1976). Hence, we see little value in describing the database yet again. Instead, our approach will be historical, tracing the evolution of the research literature. We begin with the 1940s, when the prevailing framework for the study of children's learning was derived from studies of animal learning, and describe the literature up to the present, when artificial intelligence, linguistics, and Piagetian theory are the dominant theoretical forces in the analysis of the development of learning and memory.

The literature dealing with infants and toddlers (defined roughly as from birth to 2 years of age) has evolved separately from comparable research dealing with preschoolers and school-age children. This dissociation of research by the age of the individual stems, in part, from the fact that pressing theoretical questions have differed for the very young versus the somewhat older child. It is also borne of the limited response repertoire of the young child who cannot produce and has limited understanding of human speech. We consider the course of these research literatures separately, then point to some of their common themes.

RESEARCH WITH PRESCHOOL AND SCHOOL-AGE CHILDREN

Little work on children's learning during the first four decades of the twentieth century has had direct impact on current theorizing. The number of relevant studies during the period is small, and most were of a descriptive or normative rather than theoretical bent. (For an exception, see early work described by Cornell, this volume). However, beginning with a study by Kuenne (1946) described by many observers (e.g., Stevenson, 1972) as the beginning of the “modern era” of research on children's learning, one can trace a continued progression of research and theory up to the present time.

Kuenne's (1946) research concerned young children's performance on transposition problems of the following type. Children first learned to select the smaller of two otherwise identical squares to receive a toy. Two transfer tasks then followed in which one square was approximately twice the size of the other. On a near test, the smaller of the original squares was paired with an even smaller square. On a far test, the two squares were both novel and were considerably smaller than those presented during original learning.

Animals such as rats and monkeys would consistently select the smaller of the two squares on a near test (i.e., they would transpose), but would chose squares indiscriminately on a far test (Spence, 1937). Spence's account of these findings was that learning the original discrimination establishes excitatory and inhibitory tendencies towards the rewarded and nonrewarded stimuli. These excitatory and inhibitory potentials generalize to other stimuli. Summing the excitatory and inhibitory potentials for a stimulus yields the net or effective excitatory tendency. On near tests, the animal selects the smaller stimulus because of its greater excitatory potential. On far tests, generalization of excitatory and inhibitory potential is negligible because of the distance from the original training stimuli; lacking a systematic basis for distinguishing stimuli, the animal chooses randomly.

Kuenne (1946) hypothesized that young preverbal children would solve transposition problems in the manner described by Spence. Hence, they should transpose on near tests but not far tests. Kuenne suggested that older children's learning would be directed not by the net excitatory potentials. Older children would, instead, verbalize the relation between the stimuli and respond on the basis of that verbalized relation (e.g., “pick the smaller one”). These verbalizations should transfer as readily to much larger and much smaller squares as to similar sized ones. Hence, older children should solve near and far tests with equal skill. In fact, no 3-year-olds consistently selected the smaller square on far tests, while 27%, 50%, and 100% of the 4-, 5-, and 6-year-olds, respectively, did so. Thus, Kuenne's data were consistent with her hypothesis that learning was more likely to be controlled by verbal responses as children developed.

This hypothesis of verbally mediated learning in older but not younger children motivated much of the research on children's learning for the ensuing 20 years. This was a vital period of research during which investigators carefully probed all facets of the mediational theory of children's learning. Among the key theoretical questions addressed were the following: (1) Are there simply two developmental phases, as described above, or is there an additional intermediate phase in which appropriate verbal responses are available but do not direct learning (Maccoby, 1964)? (2) Do later, mediated forms of learning displace the earlier nonmediated variety, or are both mediated and unmediated learning present in older children and adolescents (White, 1965)? (3) Is the mediator necessarily verbal, or is it conceptual or attentional in nature (Kendler & Kendler, 1962; Zeaman & House, 1963)?

Throughout this period, neobehaviorism, or as it is sometimes called theoretical behaviorism, provided the overarching theoretical framework. This framework, while not monolithic (Kendler & Kendler, 1975) does entail a set of claims about the nature of the organism, its development, and the theoretical tools needed to explain the organism's development (Bolles, 1975; Kendler & Kendler, 1975; White, 1970). Theoretical behaviorism is derived from the mechanistic worldview in which the organism is viewed as inherently at rest. Like a machine, this passive organism shows activity only in response to external forces. It was assumed that the concepts of stimulus, response, and associations among stimuli and responses provided sufficient theoretical machinery to explain the behavioral development of this fundamentally passive organism.

Beginning in the 1960s, the limitations of S-R concepts became apparent to many investigators. White (1976), for example, argued that

The great plague of learning experiments in their heyday was the plethora of odd effects which could not be handled in the conventional S-R terminology and which therefore had to be handled in discussion sections of papers. People kept discovering untoward set effects, expectancy effects, orienting effects; they kept inventing new terminology for odd, non-S-R transfer mechanisms. Nothing so much forced the rediscovery of the active organism as the recalcitrance of learning experiments designed on the premise of a passive organism. (pp. 105–106)

A study by Flavell, Beach, and Chinsky (1966) illustrates the phenomenon described by White (1976). These investigators used a serial recall task to determine if there is a stage in learning during which children can name stimuli (and hence have available an appropriate verbal mediator), but they do not do so spontaneously in the course of a learning task. Children of ages 5, 7, and 10 years were shown seven pictures. Then the experimenter pointed to a subset of two to five pictures for the children to remember on that trial. Children's spontaneous verbalizations were observed by an experimenter trained as a lip reader, a feasible procedure because pictures had been chosen so that their names would be quite discriminate from one another. Only 10% of the 5-year-olds verbalized the names of the pictures during the memory task, while 60% of the 7-year-olds and 85% of the 10-year-olds did so. Subsequently, children were asked to name all of the pictures, which proved to be a trivial task for all. Thus, a large number of 5-year-olds and some 7-year-olds seemed to be in the hypothesized intermediate stage in which appropriate verbal mediators were available but not used.

The intriguing question, of course, is why the younger children did not use the available and beneficial mediators. Flavell et al. discussed two possible explanations. In the first, representing their attempt to stay within the confines of theoretical behaviorism, they proposed that only gradually do children learn to use language beyond its original communicative context. The memorization task used by Flavell et al. represented one of those contexts to which language had not yet been extended by the younger children.

The second explanation goes far beyond the bounds of S-R constructs. Flavell et al. argued that

Verbal coding and rehearsal on a task such as ours could be construed as reflecting or embodying certain intellectual competencies which have nothing intrinsically verbal about them. An S who codes and rehearses is, first of all, responding to the task in an intellectually active fashion … coding and rehearsal represents a systematic plan for coping effectively with the task requirements; it is, as such, a kind of problem-solving “algorithm” of the S's own devising. And finally, it represents a time-binding, goal-directed effort on his part … Viewed in this way, our kindergarten Ss may have failed to talk to themselves for reasons having nothing whatever to do with their level of linguistic development. That is, they may simply have been too young to engage in the kinds of intellectual activities which assume the guise in this particular task, of verbal coding and rehearsal. (pp. 297–298)

Obviously this explanation represents a complete abandonment of neobehaviorism. The organism is now actively interpreting the task instead of acting only in response to external forces. Goals, plans, and strategies are the key theoretical constructs, replacing stimuli, responses, and associations.

Use of Mnemonic Strategies

Flavell et al.'s second explanation spawned an entire decade of research, beginning in the late 1960s, concerning developmental change in children's use of strategies for learning and remembering (Brown et al., in press. Kail & Hagen, 1982). This work revealed that by 8 or 9 years of age, children would use numerous strategies as aids in their efforts to remember. They would rehearse the names of stimuli and organize stimuli in terms of various semantic properties (Hagen & Kail, 1973; Neimark, Slotnick & Urich, 1971; Ornstein, Naus. & Liberty, 1975). Children younger than 7 or 8 years would rarely use these or any of a number of other potentially helpful memory strategies. These children often seemed to be conspicuously inactive in response to instructions to remember, a marked contrast to the elaborate memorization schemes of older children.

The foregoing essentially derogatory characterization of the young child was, for several reasons, unsatisfactory. First, it was primarily a description of mnemonic shortcomings rather than a positive portrayal of competencies. Because most developmental psychologists assume that later competencies emerge (at least in part) from earlier ones, a “description by deficit” is of little help in suggesting how children develop greater expertise. Second, the unflattering description of preschoolers' memory skills was a discordant note in a period replete with demonstrations of competence in preschool children (Gelman, 1978). Third, the inept characterization of young children did not jibe well with parents' reports that their young children often remember quite accurately.

The cumulative impact of these criticisms was to focus research in the late 1970s on the memory skills of preschoolers and young children. One line of research was designed to identify those factors governing the development of effective memory in these children. Much of this work has been done by Marion Perlmutter and her colleagues (Myers & Perlmutter, 1978; Perlmutter, 1980) and is described in Chapter 11 of this volume. A second line of research has addressed the reasons for young children's failure to use mnemonic strategies. It was proposed (e.g., Flavell & Wellman, 1977) that the emergence of strategic behavior may be partly attributable to the growing child's acquisition of the many mnemonic facts that implicitly govern the behavior of a more mature individual. For example, consider the concept ...

Table of contents

• Cover
• Half Title
• Full Title
• Copyright
• Contents
• Preface
• PART I: HISTORICAL OVERVIEWS
• PART II: ONTOGENY OF MEMORY AND ITS PREREQUISITES
• PART III: COMMENTARIES AND PROSPECTS
• Author Index
• Subject Index