INFANT "POINTING" BEHAVIOR IN THE FIRST THREE MONTHS OF LIFE

THOMAS EDWARD HANNAN, Purdue University

Abstract

Observations of infants under 3 months of age have indicated the presence of hand movements that bear at least a superficial similarity to the social gesture of pointing. However, empirical accounts of the emergence of social pointing have suggested that the traditional index finger extended version of pointing does not become differentiated from grasping until sometime after 7 months. Two questions arise: (1) are these early appearing forms of "pointing" developmentally continuous with later forms of pointing, and (2) if they are, why should "pointing" emerge so many months before function seems apparent? A review of available literature highlighted two major problems with the methodologies employed to examine early "pointing". First, there did not seem to be adequate data to support a contention that "pointing" could be reliably detected from the flow of infant behavior. Secondly, because previous accounts have been descriptive, there is no empirical data available to indicate that these "points" are anything more than randomly emitted movements. The purpose of this study was to provide a rationale for further investigation into possible continuities between early and later forms of pointing. A longitudinal, case-study design was employed to explore two general questions: (1) can "points" be reliably detected and coded from videotapes, and (2) are "points"s random, or do they demonstrate regularities of cooccurrence, sequence or patterning with other categories of infant behavior and the flow of dyadic interaction? Two infants and their mothers were videotaped during a free-play situation, from the 1st to the 12th week of the infant's life. These videotaped sessions were transcribed into units of action and time by teams of coders employing a catalog of infant and maternal categories of movement. The catalog listed mutually exclusive movements in each of 4 modes or body regions of the infant (eyes, brows, mouth and hands), and 4 modes of the mother (eyes, brows, mouth and body distance). Reliability analysis of the hand categories was based on .50 of the total observed time. The data analysis was composed of measures of proportion of time spent in each category of behavior correlated with the values for right- and left-handed "pointing", tests of cooccurrence (the degree of temporal overlap), time- and event-lag sequential analyses and transitional or state change probabilities. Statistical tests of significance were based on observed binomial probabilities tested against either a binomial distribution, or a Z distribution, after a correction for continuity was applied. "Pointing" was found to be reliably detected, with estimates generally exceeding .90. The probability of "pointing" appeared to be equal for the interaction and alone periods, suggesting that it is not yet strictly a social gesture. However, "pointing" was not found to be a random movement. "Pointing" occurred significantly more on the right hand than on the left. Right-handed "points" were correlated positively with gaze and relaxed face, but negatively with crying. Left-handed "points", however, were positively correlated with look away and pout, but negatively related to smile. Both hands were positively related to the percent of time the mother leaned forward, and the left hand was negatively related to lean back. Right-handed "points" cooccurred with left-handed fingers curled, while left-handed "points" cooccurred with right-handed finger cycling. Both hands cooccurred positively with nonexpressive mouth and negatively with smile; there was a tendency for cooccurrence with look at. Examination of the sequences of "pointing" with other behaviors suggested that "pointing" is 'sandwiched' between changes of the eyes, brows and mouth. It was concluded that there is evidence for investigating a continuity of early and later forms of pointing.

Degree

Ph.D.

Subject Area

Developmental psychology

COinS