DEVELOPMENTAL DIFFERENCES IN SCENE RECOGNITION AT SPATIAL DECISION POINTS

Author: Sally Ellen Doherty
Degree: Ph.D. in Education
School: University of California Santa Barbara
Date: 1984
Advisor: James W. Pellegrino

Abstract

The primary purpose of this study was to determine if children who live in a neighborhood have differential knowledge about choice points and non-choice points on frequently traveled routes. The secondary purpose was to determine if knowledge for neighborhood locations varies as a function of age.

Forty-four children ranging in age from seven to fifteen years, all residents of a suburban neighborhood, were tested on a scene recognition task containing slide views taken at decision points and non-decision points from their own neighborhood and distractor neighborhoods. Decision points were operationalized as any type of intersection or location where streets join. Non-decision points were operationalized as locations along streets between intersections. The study also explored differences due to type of scene presented: sidewalk views versus plots with information about a single house and yard.

Results indicated that neighborhood scenes at intersections are recognized more accurately and faster than scenes along routes. This finding supports the prediction that in a developing cognitive representation, more is known at spatial decision points where alternative actions must be considered for navigational purposes. Recognition performance improves with age, and there is a significant interaction of route location and age which supports a three stage model of developing spatial cognition. Young children show little differentiation of knowledge by route location; whereas nine to twelve year old children demonstrate route representations with clear differentiation. Performance of the oldest children suggests an emerging well-coordinated, survey-type knowledge representation in which differentiation by route location decreases. However, metrically-based representations do not simply supercede more primitive route representations; both types of information are available for decision-making.

Children have better knowledge for scenes depicting plots than views. This finding implies that in a relatively homogenous suburban environment, plots-at-intersections become landmarks for navigational decision-making and lends support to the position that specific landmarks are learned within the context of learning routes. Recognition decisions for views also take longer than plots, and response times increase markedly for views in distractor neighborhoods whereas decisions for plots are unaffected. These findings are interpreted as evidence of a plotframe or schemata which facilitates recognition decisions for plots but not views.