Biological and Cultural Bases of Human Inference

Riccardo Viale, Daniel Andler, Lawrence A. Hirschfeld

• Publisher: Psychology Press
• Year: 2013
• eBook ISBN: 9781134812776
• Edition: 1
• Topic: Psychology
• Subtopic: Cognitive Psychology & Cognition
• Index: Psychology

Biological and Cultural Bases of Human Inference addresses the interface between social science and cognitive science. In this volume, Viale and colleagues explore which human social cognitive powers evolve naturally and which are influenced by culture. Updating the debate between innatism and culturalism regarding human cognitive abilities, this book represents a much-needed articulation of these diverse bases of cognition.Chapters throughout the book provide social science and philosophical reflections, in addition to the perspective of evolutionary theory and the central assumptions of cognitive science. The overall approach of the text is based on three complementary levels: adult performance, cognitive development, and cultural history and prehistory. Scholars from several disciplines contribute to this volume, including researchers in cognitive, developmental, social and evolutionary psychology, neuropsychology, cognitive anthropology, epistemology, and philosophy of mind. This contemporary, important collection appeals to researchers in the fields of cognitive, social, developmental, and evolutionary psychology and will prove valuable to researchers in the decision sciences.

1

Cognitive Development, Culture, and Inductive Judgment

Riccardo Viale

Rosselli Foundation

Daniel Osherson

Princeton University

TWO VIEWS OF INTELLECTUAL DEVELOPMENT

As agents of abductive inference, children have an enviable reputation. It rests on the remarkable transition from benighted neonate to savvy and enterprising 5-year-old. During this period, children acquire a grammatical system, discover the boundaries of tolerated behavior, and begin to discern the biological and physical kinds encountered in their environment. It is no wonder that developmental psychologists often compare children's achievements to that of professional scientists embarked on their own voyage of discovery.¹ If the analogy can be sustained, it suggests that the study of cognitive development might offer clues to the successful implementation of computerized scientific activity by simulating the child's successful strategy.

Conceiving the child as scientist, however, requires more than recording her impressive achievements. Intellectual development must also be shown to derive from the kind of theory construction attributed to professional scientists. Assimilation of development to rational inquiry is precisely the view of an influential school of developmental psychology, recently defended in Gopnik and Meltzoff (1997). Abstracting away from differences in background knowledge and startup funds, children (like scientists) are claimed to develop a succession of theories that evolve adaptively under the impact of data. At each step, the child is supposed to have an organized corpus of belief that amounts to a scientific doctrine. The transition between theories is supposed to depend on their respective, predictive success, as well as on simplicity criteria familiar from the his-tory of adult science. This attractive picture of cognitive growth is meant to apply to knowledge of the physical and social worlds, as well as to word meaning, mental events in other minds, and so forth.

The opposing view is Chomsky's triggering model. According to this conception, it suffices for children to come into contact with a few key data for an articulated and stable theory to leap to mind. The claim is that the child's theory enjoys not the slightest justification of a sort that makes sense to professional scientists. The process is typically described in terms of parameter setting, as is well known.² Let us try to isolate the different commitments of the two perspectives on development—namely, the “little scientist” view versus the triggering model of the Chomskyans. On the former account, we expect each advance in cognitive development to issue in the kinds of theories announced by scientists, for example, involving hypothetical variables related to each other in some deductive way. Such is the synchronic commitment of the little scientist view. But notice that the triggering model is equally compatible with the child having scientifically respectable theories throughout development. It need only be imagined that each parameter flip brings to the fore a prestored body of mature, scientific belief.

The two views of development are more easily distinguished dia-chronically, in terms of the kinds of experience that nudge the child from one theory to the next. The triggering model conceives this process in the same way that visual stimuli trigger imprinting in baby ducks. In contrast, the little scientist view assimilates the child's use of experience to theory selection on the part of a rational agent.

Clearly, the little scientist perspective is more joyous than that of the triggering model. Whereas parameter setting imposes tight limits on the kinds of theories that ordinary folk can construct about their world, little scientists can help themselves to everything accessible to modern science. Preferences notwithstanding, we must try to evaluate which of the rival perspectives is closer to the truth—or closer with respect to some particular domain of knowledge.

THE CHILD AS SCIENTIST

The thesis to be evaluated in this chapter is stated in its most persuasive form by Gopnik and Meltzoff (1997). They wrote: “We want to argue that the cognitive processes that underlie science are similar to, or indeed even identical with, the cognitive processes that underlie much of cognitive development” (p. 32). These authors attempted to clarify this claim in several ways. For our part, it seems that the most straightforward reading is the following:

(1) The little scientist thesis: The theories that children construct of their physical and social environments result from the application of scientific method to the data that are available to them.

In this sense, children are portrayed as little scientists.³ As a first step toward clarifying (1), we note that scientific method is intended here in the normative sense, not the descriptive one. In other words, it is claimed that children draw inferences the way scientists should, not the benighted way that scientists sometimes do. Of course, everyone makes mistakes, so (1) is not falsified by the occasional methodological error. Rather, what is being claimed is that, by and large, children's inductive inferences conform to proper methodology.

There are systematic inductive tendencies, however, that seem to contradict the thesis at issue. These tendencies concern the use of diverse evidence in inductive inference. We now describe the relevant psychological data and then see what can be done to make them appear less offensive to the picture of children as little scientists.

EVIDENTIAL DIVERSITY

Why did Newton's theory end up commanding so much assent? One reason is that in the presence of various background assumptions, the theory accurately predicts heterogeneous phenomena, such as the trajectories of balls thrown into the air, the behavior of gyroscopes, and the orbits of celestial bodies. At the end of the 17th century, these phenomena appeared very diverse, which forced respect for the theory despite reservations about the reliance on occult, nonmechanical entities like gravity. The potency of diverse evidence for conforming theories was subsequently recognized by philosophers of science, who often see it as a way to subject hypotheses to severe test. Now for a more modest example. Consider the hypothesis (2).

(2) Vitamin K deficiency causes neuron death in the auditory nerves of all mammals.

Even in the absence of biochemical knowledge about Vitamin K, most people find that (2) is better supported by the evidence reported in (3)a, than by that in (3)b.

(3)	(a)	Vitamin K deficiency causes neuron death in the auditory
		nerves of both tigers and elephants.
	(b)	Vitamin K deficiency causes neuron death in the auditory
		nerves of both tigers and lions.

Indeed, preference for the tiger / elephant data was expressed by 74% of a group of 80 American college students asked to choose “which of (3)a, (3)b provides a better reason to believe the hypothesis (2)” (see Osherson, Smith, Wilkie, Lopez, & Shafir, 1990). Likewise, 80% of a sample of 58 Korean college students share the diversity intuition, as do 70% of a group of 80 Italian college students.⁴ Evidential diversity makes sense of such intuitions because tigers and elephants resemble each other less than tigers and lions.

Similar results have been obtained with Korean subjects judging arguments based on social categories (see Choi, Nisbett, & Smith, 1997) and likewise in Italy (Baroni & Diamantini, 2001). For example, the latter authors asked 97 Italian college students in Milan to make a forced choice between the following two arguments, selecting the one whose premises gives better reason to believe its conclusion.

Unemployment is increasing among lawyers and notary publics.

__________________________________________________

Unemployment is increasing among all self-employed professionals.

Unemployment is increasing among lawyers and accountants.

__________________________________________________

Unemployment is increasing among all self-employed professionals.

_____________

Among Italians, lawyers and notary publics are more similar to each other than are lawyers and accountants. For this reason, subjects were expected to perceive greater strength in the second argument compared with the first. This was the case for 84% of them. Similarly, 85% of the subjects opted for the second argument in the following pair because Italians judge bakeries and pastry shops to be more similar to each other than are bakeries and fruit vendors. Both arguments bear on a new French holiday designed to honor small businesses.

The holiday honors bakeries and pastry shops.

__________________________________________________

The holiday honors all small businesses.

The holiday honors bakers and fruit vendors.

__________________________________________________

The holiday honors all small businesses.

Of 80 secretaries working in Milan, 80% likewise judged the second argument (with more diverse premises) to give better reason to believe the conclusion than the first. So far, so good for the little scientist view because it looks as if a respectable principle of scientific methodology is present in a diverse sample of adult nonspecialists.

Trouble arises, however, in a developmental study carried out at the University of Michigan (Lopez, Gelman, Gutheil, & Smith, 1992). Two attempts were made to document an appreciation for evidence diversit...

Table of contents

• Cover
• Half Title
• Full Title
• Copyright
• Contents
• Preface
• Editors and Contributors
• Introduction: Local or Universal Principles of Reasoning?
• 1 Cognitive Development, Culture, and Inductive Judgment?
• 2 Culture and Point of View
• 3 Cultural Variation in Reasoning
• 4 Thinking About Biology: Modular Constraints on Categorization and Reasoning in the Everyday Life of Americans, Maya, and Scientists
• 5 Who Needs a Theory of Mind?
• 6 Framing and the Theory-Simulation Controversy: Predicting People's Decisions
• 7 An Evolutionary Perspective on Testimony and Argumentation
• 8 Normativity and Epistemic Intuitions
• 9 Probabilistic Reasoning and Natural Language
• Author Index
• Subject Index