Cognitive Structures and Development in Nonhuman Primates

Francesco Antinucci

• Publisher: Psychology Press
• Year: 2019
• Print ISBN: 9780805805444
• eBook ISBN: 9781317785385
• Topic: Biological Sciences
• Subtopic: Emotions in Psychology
• Index: Biological Sciences

The contributors to this volume present research concerning the cognitive structures and development of nonhuman primates from a cognitive psychological perspective. The authors and researchers come to this project from the study of humans and apply their knowledge to research on nonhumans. For professional, researchers, and students in cognitive, developmental, and experimental psychology.

I INTRODUCTION

1 The Comparative Approach to the Study of Cognition

Francesco Antinucci
Istituto di Psicologia, C.N.R., Rome, Italy

The studies presented in this book grew out of a line of research attempting to pursue the most classical aims and themes of comparative psychology as it was originally established and formulated, at the turn of the century, in the works of Romanes, Hobhouse, Kohler, etc. Unfortunately, with very few exceptions, they are not (and have not been for a long time) the aims and themes of what is still today recognized, taught and practiced as comparative psychology. Hence, the need of this introduction that would have been superfluous had it still been true that '‘[Comparative Psychology] ... is concerned with the genesis of the human mind as such. It seeks to determine the stages of development which lead from the first beginning of psychic life to the emergence of the human reason” (Hobhouse, 1968, p. 117). As recently remarked by Tolman (1987, p. 288), “The vision of comparative psychology expressed by Romanes, Morgan, and Hobhouse was one of a distinctly human psychology, charged with identifying the unique qualities of human psyche (or mental functioning), and explaining them in terms of their evolution”. If its object is the evolutionary study of the human mind, then, in purely darwinian terms, “comparative” refers to the classical method that allows, as in morphology or anatomy, the determination and reconstruction of the evolutionary path through the systematic comparison of corresponding structures in increasingly removed, related species. In the specific case, the “minds” of other species, beginning with man’s closest relatives. Kohler writes, in the introductory justification of his study of the intellectual capacities of apes:

Two sets of interests lead us to test the intelligence of higher apes. We are aware that it is a question of beings which in many cases are nearer to man than to the other ape species;. . . .We wished to ascertain the degree of relationship between anthropoid apes and man in a field which seems to us particularly important, but on which we have as yet little information. (1976, p. 1)

As we shall see in a short while, no less important is the second “set of interests” motivating his choice.

Of the three key concepts defining, as we have just seen, the comparative psychology enterprise (evolutionary study of the human mind) not one survived the behavioristic revolution: not “evolutionary through the comparative method”, not “human” and, especially, not “mind”. Through an historical process that has recently been amply documented (see, among many others, Dewsbury, 1984; Epstein, 1987, O’Donnell, 1985), comparative psychology became the “synchronic” study of the behavior of one or two species of animals. By 1927, another one of the “founding fathers”, attending him too, as Kohler and for the same reasons, to the study of the intellectual capacities of man’s closest relatives, R. Yerkes, felt obliged to write in the introduction to one of his classical monographs:

As a fact, the term mind (and in the present title mental) is used to protest against what he [the author] deems unwarranted and unprofitable assumptions of the cult of behaviorism and as indication that he considers mental, no less than behavioral phenomena, material of biological science. (Yerkes, 1927b, p. 1)

The title of Yerkes’ monograph was “The mind of a gorilla” and the subtitle of its second part, in the introduction to which the statement quoted above appears, “Mental development”.

This is not the place to rerun the complex historical and epistemological process that led from the just preoccupations with inferences of mind functioning based on analogical extrapolation of introspective results (“Starting from what I know of the operations of my own individual mind, and the activities which in my own organism they prompt, I proceed by analogy to infer from the observable activities of other organisms what are the mental operation that underlie them”, Romanes, 1883, pp. 1-2) to the curious means-end distortion thereby overt behavior became the direct end of the investigation, rather than the indirect mean to access mental functioning (see Wasserman, 1984).

One point should, however, be mentioned because it is directly relevant to the resurrection of the original enterprise, and to the approach presented here. Quite independently from its physicalist methodological and theoretical restrictions (which, in fact, were to be progressively relaxed in the 60 years of its history), behaviorism espoused a peculiar metatheoretical assumption (and a recurrent one in the history of several disciplines, see Antinucci, 1982, for a wider discussion), viz., that the whole of cognitive structure and functioning could be exhaustively accounted for in terms of one simple atomic mechanism, the association of good empiricist memory. Obvious macroscopic differences among levels of structure and functioning, as well as among species capacities are, in principle, conceived of as being of quantitative nature: more easily established, more extensive, richer and longer, associations and associations of associations account for increasing complexity (see what Roitblat, 1987, pp.54-55, appropriately terms the “ubiquity” and “equipotentiality” assumptions of behaviorism). Given this assumption, what is needed most of all is a thorough understanding of all the parameters regulating the basic mechanism (reinforcement, extinction, generalization, differentiation, etc.): complexity would then result from some simple summation function. Hence, the rat or the pigeon became ideal object of study, not, however, as terms of an extended evolutionary comparison with man, according to the dictate of the original program, but as valuable, because essentially isomorphic, small-scale models of larger organizations. In fact, the complexity-as-quantity assumption thoroughly trivializes the comparative function and, at the same time and for the same reasons, rescues from the realm of biological absurdities the possibility of directly relating even two species belonging to two different taxa, like pigeon and man: “Pigeon, rat, monkey, which is which? It doesn’t matter” (Skinner, 1956, p.230).

Thus, “mind” disappeared for epistemological reasons, “human” became interchangable with animal for theoretical ones and “evolutionary” was emptied of meaning. No wonder that when the study of human cognition was put back on its feet, in the sixties, the comparative approach, turned into the study of animal learning behavior, had very little to say. The two fields drifted more and more apart, and while the first was developing increasingly more sophisticated models of the mind “software”, the second gravitated more and more toward the physiological “hardware”, as testified, for example, by the distribution of articles appearing in the Journal of Comparative and Physiological Psychology.

The repeal of the atomistic assumption is at the core of the “cognitive revolution” (see Chomsky, 1959; Miller, Galanter, & Pribram, 1960). With it there came about a return (obviously, on richer and more sophisticated bases) to structural conceptions. Cognitive capacities are best viewed as integrated structures, that is to say, they cannot be reduced to the simple summation of independent atomic constituents. This is not to say that they are some kind of metaphysical indecomposable whole, of organicistic or vitalistic memory, but simply that their specific properties and functioning depend crucially on the reciprocal relations and delicately orchestrated interactions of their constituent elements, rather than on their elementary nature. To take an obvious, but clear example, human language is not a list of words, and sentences that instantiate its functioning are not sums of juxtaposed words.

It is this network of interdependences and its crucial role in functioning that is referred to by the word “structure”. Because of their very organization, structures tend to show a number of distinct properties: scale differences translate into qualitative differences, i. e., show novel properties, levels of functioning are irreducible to each other, and, especially, causal determinants tend to have non-linear effects (“trigger” causation).

Owing to these properties, model-building within this framework is a much more complicated task. Interdependence of parts, non-scaling and trigger causation cooperate in rendering extremely indirect the relation between external events observed in functioning and underlying processes generating them. This creates two hard, related problems to the theorist: arbitrariness of domain delimitation, and model underdetermination. How are we to delimit underlying unitary mechanisms in largely integrated systems on a non-arbitrary basis? External phenomena are, as we have just seen, no criterial guide because no one-to-one correspondence can be assumed between them and underlying mechanism. What appears, for example, as language behavior can be determined by a host of separate interacting mechanisms, rather than by a unitary, self-sufficent language capacity; and conversely, each of these hypothetical mechanisms (like, for example, a single processor for serial units) might play crucial roles in generating a variety of apparently unrelated behaviors (see, for example, Bever, 1974).

Or, to take an example closer to the topics that will be dealt with in this book, should we pool tool-use behaviors together as corresponding to a proper domain of cognition (as done by many students), or the capacity that makes it possible to use a stick to rake an out-of-reach object has nothing whatsoever to do with the one that makes it possible to use a hammer-like stone to break the hard shell of a fruit?

Much for the same reasons, a structural theorist must face the related problem that data vastly underdetermine theory, that is to say, that quite a number of possible models are compatible with the same set of data.

Holding a structural view, Kohler was well aware of the arbitrariness problem (what is intelligence? how is it delimited from other capacities?), and it was in connection to this problem that he envisaged a fundamental role of comparison. This, in fact, is his second declared “set of interests” in pursuing the study of intelligence in apes:

The second aim is theoretical—writes Kohler in the same introduction quoted above—Even assuming that the anthropoid ape behaves intelligently in the sense in which the word is applied to man, there is yet from the very start no doubt that he remains in this respect far behind man, becoming perplexed and making mistakes in relatively simple situations; but it is precisely for this reason that we may, under the simplest condition, gain knowledge of the nature of intelligent acts. (Kohler, 1976, p. 1, italics added)

Patterns of similarity and difference among closely related species might offer powerful cues as to the constituency of the underlying mechanisms that can generate them. On the one side, they show natural groupings that contribute to answer the question of what goes with what and, hence, offer a non-arbitrary basis to delimit boundaries of domains.

On the other side, structures responsible for species differences must be so modeled as to allow their being interrelated by plausible evolutionary transformations (where “plausible” means that take into account parameters like the evolutionary distance between the species compared, directional trends across a number of related species, functional viability at all intermediate stages, etc.). In view of the fact that one of the defining characteristics of structures is their being internally highly interrelated, this is no trivial contraint: tight interrelation means that even a slight local modification is likely to reverberate diffusely and snowball into widespread reorganizations. Hence, choice among possible models might be further reduced by the requirement that they must be capable of supporting a given transformation. Within such a perspective, a double role for the comparative study is thus provided. Structures and the transformations relating them might enter in a beneficial mutual relation: comparative modeling of the first leads to reconstructing the evolutionary transformation, which in turn feeds back in constraining and hence refining the hypothesized models. Far from being vicious, this circle can generate successive convergences that substantially reduce model underdeterminacy. It is perhaps no accident that the initial rebirth of the original intent of comparative psychology took place with the study of one of the most highly structured, theory dependent, cognitive domain, language capacity, in man’s closest relatives, where the working of this process could be amply exploited (see Gardner & Gardner, 1971; Premack, 1976; Rumbaugh, 1977; Terrace, Petitto, Sanders & Bever, 1979).

Exactly the same set of reasons motivates the addition of another fundamental dimension to the study of cognitive capacities: that of their ontogenetic development. As shown in embryogenesis, at all levels, ontogenesis of structures proceeds by progressive differentiation, on the one side, and increasing integration, on the other side. Corresponding structures can then be compared (much as in evolutionary comparison) at different stages of this process. The interplay between structure and developmental transformation can then generate the same positive effects on constraining models as in the case of phyletic transformation. On the one hand, structures can be investigated before or at a partial level of integration. By seeing developmentally related groups of phenomena, we can again understand more easily what goes with what, what is more likely to depend on a unitary underlying mechanism. On the other hand, and most importantly, the way adult structures are organized might depend to a large extent by the very way they develop (see, for example, Chomsky’s argument [1975] that a theory of language acquisition is ipso facto an explanatory theory of language capacity). The theory of cognitive capacities and functioning developed by Piaget, which will be illustrated in the next chapter, and the evidence supporting it provide ample demonstration of this claim.

Finally, evolutionary and ontogenetic transformations might be related in a much more fundamental way than in the similar roles they play in theory construction by constraining models of capacities. It has been repeatedly argued in the past (de Beer, 1930; Garstang, 1922) that ontogeny is the main locus of evolutionary change. With recent critiques to the “modern synthesis” paradigm of evolutionary theory (Eldredge & Gould, 1972; Stanley, 1979), this position is currently the object of renewed theoretical interest (Gould, 1977; Wake & Larson, 1987; Gottlieb, 1987). Recent techniques in dating evolutionary relatedness (“molecular” clocks), dominant stability in stratigraphic records (“stasis”), as well as other converging pieces of evidence, put more and more in question the slow and gradualistic view of evolutionary transformation (by piecemeal accumulation of small changes) originally maintained by Darwin and strongly reaffirmed in the modern synthesis. The tempo of macroevolution seems to be much more rapid: to take a concrete example, evolutionary distance between man and chimpanzee as measured by similarity in the genetic make-up appears to be much smaller than previously hypothesized and to translate into a few million year divergence, not many more than those separating modern man from its first australopithecine ancestor (see Nute & Mills, 1986). Yet the macroscopic differences of these two species in many structures are wide. This paradox means that small changes at the genetic level must somehow translate into large differences. The combined effects of structural interrelation and developmental transformation might provide just such a mechanism. In fact, small changes taking place early in ontogeny, that is, in “parent” structures that will later differentiate, will be projected onto all “descendant” structures and hence widely amplified. The final effect, as seen in the adult, is, therefore, much more dependent on the time of occurrence of such modification (the earlier, the greater), than on its magnitude (see chap. 7, for a possible example of this kind). Haeckel’s (1866) traditional and influential argument that “Phytogeny is the mechanical cause of ontogeny”, giving rise to the celebrated doctrine of ontogenetic recapitulation, is thus reversed into its opposite.

For these reasons, consideration of ontogenetic development, joined to a structural standing, will be the major focus of this book (as evidenced in its title), in the belief that, even more than simple comparison, the comparative study of the cognitive ontogenies of man’s related species might provide an essential key to the explanation of human cognitive capacities.

2 The Theoretical Framework

Francesco Antinucci
Istituto di Psicologia, C.N.R., Rome, Italy

The approach to the study of cognition that will be followed in this book is grounded in the epistemological and theoretical work of Jean Piaget. Beside the vastness and empirical value of his specific findings in a half century of uninterrupted search for the origin and structure of human cognition, the general principles governing cognitive organization Piaget developed and substantiated along this long work represent an ideal tool to pursue the study of cognition in terms of the biological perspective outlined above.

In the introduction to his studies on the origins of intelligence, under the title “The biological problem of intelligence”, Piaget (1974, p. 3-5) writes:

Intelligence is an adaptation. . . Certain biologists define adaptation simply as preservation and survival, that is to say, the equilibrium between the organism and the environment. But the concept loses all interest because it becomes confused with life itself. . . There is adaptation when the organism is transformed by the environment and when this variation results in an increase in the interchanges between the environment and itself which are favorable to its preservation. . . To say that intelligence is a particular instance of biological adaptation is thus to suppose that it is essentially an organization and that its function is to structure the universe just as the organism structures its immediate environment.

Functional continuity and structural discontinuity are the cornerstones of this view.

Adaptation realized by cognition, in the sense of the transformative process defined above, is the product of functional processes that are invariant across “organisms” (whether these are conceived as different species, or ontogenetic phases of an individual). Their action, however, generates and successively transforms structures that, as organized and organizing totalities, are discontinuous with each other, i.e., that are not merely quantitative amplification of each other. Continuity of function provides the indispensable common basis on which comparison can be effectively operated, while the possibility of discontinuity in the structures produced allows for specificity at each level. In this way, the comparison is not trivialized as in behaviorism, where functional continuity is instead accompanied by structural continuity (see the “atomistic” assumption...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Part I Introduction
• Part II Longitudinal Study of Cognitive Ontogeny: Birth To Stage 4
• Part III Cognitive Domains
• Part IV Structure and Development of Logical Cognition
• Part V Conclusion
• References
• Author Index
• Subject Index