Development of cognition

Psy 416: Reasoning and Problem Solving
Development of cognition

Erwin Segal

(Read the Chapter in Children's Thinking by Siegler on reserve. This summarizes many issues in the development of thinking skills. )

What are the capabilities of children at different age levels? What can children do? What do they start with? What are the processes by which they achieve their competence? What evidence can be found to support the distinction between the acquisition of competence as a function of learning versus the acquisition of competence primarily as a function of maturation? What kinds of things do they need to learn in order to reason and solve problems?

Piaget's theory: The most popular view. The development of cognition is essentially an adaptation to the physical and social environment. One must combine accommodation to the environment and assimilating aspects of the environment to one's needs. More detail

Stages of development--(i) sensory-motor, (ii) preoperational thought,
(iii) concrete operations, (iv) formal operations;
The fact that there is decalage is very important as a significant delimiter on the stages. One can find examples of the early stages at all ages with new materials or domains. Vygotsky: Zone of Proximal Development emphasizes cultural differences and the role of teaching, scaffolding supplied by expert, social precedes individual, complex skills have to be shaped.
more detail

Spelke: Perhaps perceptual organization is innate to some extent: Identify objects, continuity of objects in contour and through space, movement of objects; acquired: the ability to identify continuous contours and smooth motion, perceptual properties as a criterion of unity. Some perceptual phenomena that seem obvious are not innate.

Bruner: Children develop the skill to incorporate more information in a single view. Siegler and others think of this as automatization. This allows the use of resources for other processes. In Bruner's research children with experience can "notice" differences they might not notice otherwise.

Chi: With the right kind of experience and training, children can do very sophisticated categorization.

One analysis: In order to solve some problems children may have to learn to ignore some obvious information, perceptual size is ignored in favor of historical continuity (conservation of volume); habitual location is ignored in favor of most recent location (object permanence), etc. Some conservation tasks can be thought of as gaining perceptual and cognitive structuring (learning what goes together, and how different components are related) followed by the application of particular algorithms to the newly organized system (conservation of number, balance beam).

1. Piagetian Task: Conservation of Number. Algorithm requires understanding the task. (Try to think of how you could get a computer to solve the task.) Each group of objects is a unit. How do you know this? Interesting: Children seem to have an innate concept of a pairing method to identify equal cardinality, but they need to learn how to order and pair the members. When a child is asked to count a set of objects she is likely to err both on the sequence of numbers and the ability to know which are counted and which has yet to be counted.

2. Mixed juice problems as on pp. 306-7. Noelting (1980), Case (1978)
mix juice and water. Keep adding dimensions needed for solving successive problems. Ages 3.5 to about 10. No clear stages as they move from low level to mathematical solutions. Different dimensions. Number, blending principle, quantity

Case (p 305) has a theory that is a development of expertise perspective. As you get better with the elementary components you can turn your attention to other aspects of the problem.

3. Siegler’s approach is the same. (Balance beam) More dimensions of the problem can be noticed. Much older, ages 7 to 17. Here they had to be able to do basic mechanics--principle of torque. Number, distance, side, relation between number and force of gravity. Simple machines and mechanical advantage.

Summary of some developmental data. Perceptual learning, learn intrinsic properties of objects and object types, learn role of language to use later, etc. These seem to be component processes that may underlie many specific domains. Piaget implies that his stages introduce us to universal skills that are applied to all domains. Although there must be some general skills most of us learn that are highly useful, many, if not most are probably more domain specific. There are good data that suggest that we never actually reach Piaget's formal operations in most domains, and we have to go through perceptual learning, and concrete operations when we enter new domains.

We gradually develop competence in different domains. One real problem is finding the component processes that are necessary to do any of the tasks. This can include perceptual integration; object conceptualization, pragmatic uses, causal analyses, physical dynamics, abstract relations, understanding number, appropriate knowledge of the elements, learning the relations between the component parts and how to maneuver between them. One of the very interesting ways to see how many component processes are involved is to try to write an algorithm to do the task. Generate a set of efficient procedures to get there! Most of the researchers, presuppose that the subjects are learning the logic of the situation, but they have to learn some of the meaningful relations. Children have to learn some of the component skills before they can do the next level. It is not obvious what the component skills are.

Ahn, W. & Bailinson, J. (1996). Causal attribution as a search for underlying mechanisms: An explanation of the conjunctive fallacy and the discounting principle. Cognitive Psychology,

Ahn, W. Kalish, C. W. Medin, D. L. & Gelman, S. A. (1995). The role of covariation versus mechanism information in causal attribution. Cognition, 54, 299-352. Ahn presents data and theory to argue for using internal causal mechanisms rather than either covariation or formal derivations as the basis for conclusions. Contrasts with the Cheng study.

Cummins, D. D. (1995). Naïve theories and causal deduction. Memory & Cognition, 24, 646-658. Shows how cause or effect uniqueness accounts for all of the significant variation in implication reasoning, A discussion of "permission" reasoning by children and how identifying permission makes a difference over and above simple relationships containing the same information.

Luria, A. R. (1976). Cognitive Development: Its Cultural and Social Foundations. Harvard University Press. Demonstrations that illiterate people raised in different cultures do not organize information the same way as westernized literate people do. Nor do they reason about hypotheticals in the same way. They found it quite difficult to separate their reasoning from their knowledge.

Siegler, R. S. (1991) Children's Thinking, 2 Ed. Prentice Hall. A general text in cognitive development with a major component directed at topics covered in this course as they apply to their acquisition in children.

Spelke, E. S. (1990). Principles of Object Perception. Cognitive Science, 14, 29-56. An investigation of some cognitive schemata that seem to be innate and some that seem to be acquired.

Back to Syllabus