William Kessen

Color vision and hue categorization in young human infants.

Two studies examined the organization of color perception in 4-month-old human infants. In Study 1, infants looked at selected spectral stimuli repeatedly until their visual attention waned. The stimuli represented instances of basic adult hue categories - blue, green, yellow, and red. Following habituation, infants were shown a series of wavelengths which were the same as or different from the stimuli first seen. Analyses of infant attention during this dishabituation phase of the study indicated that infants categorize wavelengths by perceptual similarity; that is, they see hues in the spectrum much as adults do. In Study 2, a group of infants who looked at the alteration of two wavelengths from the same hue category habituated as did the group of infants who looked at the repitition of a single wavelength from that category, but a group of infants who looked at two wavelengths from different categories habituated at a slower rate. Data from the two studies suggest a high degree of organization of the color world prior to language acquisition.

Visual scanning of triangles by the human newborn.

Abstract Ten human newborns were shown a homogeneous black visual field and ten newborns were shown a large black triangle on a white field. Ocular orientation to within approximately ±5° of visual angle was measured by scoring infrared photographs of corneal reflections. The infants showed much less disperison of scanning in the presence of the triangle than in the presence of the homogeneous field. Moreover, ocular orientations were directed toward a vertex of the presented triangle. The results were related to Hebbs theory of perceptual development, to analyzer theories of discrimination, and to studies of complexity and preference in the human newborn.

Preference for Shapes of Intermediate Variability in the Newborn Human

Newborn humans presented with pairs of shapes, each shape differing in number of turns (angles), prefer shapes with 10 turns to shapes with 5 turns or 20 turns, as inferred from photographic recordings of eye fixations.

The imitation of pitch in infants

Twenty-three infants, between 3 and 6 months of age, were seen three times in a study of pitch-matching. In the third session, the babies were presented with two or three sung tones (D, F, and A above middle C). All of the babies responded by vocalizing on the presented pitch significantly often. The ability of infants in the first six months of life to imitate pitched tones poses questions important both for theories of imitation and for early music education.

Age and uncertainty: Developmental variation in preference for variability

Abstract Two hundred sixty elementary school children were run in two studies of the relation between preference and variability of stimulus materials. In Study I, S s made preference judgments in a paired-comparison design of 12 random shapes containing from 3 to 40 independent turns. As predicted, children of all ages showed a point of inflection of preference at figures of ten turns. Significant age differences were found in the preference expressed for figures of high variability (many turns). In Study II, children from two school grades and young adults made preference judgments for sequences of letters and words that varied in their unpredictability. As predicted, a significant age difference was found in point of inflection of preference for these materials. Results were interpreted as indicating control of the response to uncertainty by (a) a limit on human processing capacity, (b) the amount of variability in the environment, and (c) the rules available to the person for coding variability.

The visual response of the human newborn to linear contour

Abstract The visual response of 16 newborn infants to a black-white edge and to a blank homogeneous field was recorded by corneal-reflection photography. Each infant was presented the edge once in a vertical orientation, to the left or right of center, and once in a horizontal orientation, above or below center. Vertical edges clearly attracted the newborns gaze but horizontal edges appeared to have no effect whatsoever. Various interpretations are offered for the greater attractiveness of vertical edges and some preliminary data on scanning patterns are discussed.

Prolonged investigation of a plane geometric triangle by the human newborn

Abstract Five human newborns were presented with a plane geometric triangle repeatedly on the same day and across days. Eye fixations and eye movements to within approximately ±3–4° were recorded by means of corneal photography. There were marked individual differences in the form of visual scanning on the figure. Some infants consistently oriented towards only a single feature while others alternated between single and multiple feature selection. Postnatal age or exposure alone did not appear to guarantee either response.

Response of the Human Infant to Level of Complexity of Intermittent Visual Movement.

Abstract Three levels of visual stimulus complexity were presented to three independent groups of 12 2- to 4-month-old infants. The stimulus was a light which changed position 32 times in a 24-second experimental (E) trial. Three complexity levels were created by varying the predictability of the direction of light movement. Six 24-second E trials alternated with seven 24-second control (C) trials. S was shown the same movement pattern on the first three E trials. On the last three E trials only four Ss in each of the three complexity groups continued to see the same pattern. The remaining Ss in each group were shifted to one of the two levels of stimulus complexity not yet experienced. The dependent variables under observation were limb movement and sucking frequency. In the preshift phase of the experiment, limb movement was suppressed by the most simple and by the most complex levels of stimulation but was facilitated by the intermediate level. The response of hands and feet, although concordant over levels of some variables (e.g., preshift complexity) did not always correspond. There was a suggestion in the later phase of the experiment that a derived score representing a pattern of limb suppression was sensitive to changes in complexity level. Sucking behavior was fairly consistently suppressed by stimulation throughout the experiment. No systematic habituation of responsiveness was found.

Maturational changes in measures of habituation

Abstract The age-related changes and individual stability of measures of visual habituation and changes with age were assessed in a sample of 38 infants seen longitudinally at 3, 4, 5, and 6 months. Three findings are presented. First, there was a gradual decrease in the duration of certain quantitative habituation measures between 3 and 6 months. Five measures—accumulated looking time, length of longest look and look at criterion, length of the baseline looking time, and the rate or slope of decline in looking—changed with age. Second, for individual infants, four of these habituation measures were reliable only between 3 and 4 months and not afterwards. Third, despite only moderate and early reliability in quantitative measures, a qualitative description of the habituation function—the pattern of decline—was more stable. Half of the infants in this sample maintained the same habituation pattern over 2 or 3 months. These findings are discussed in terms of developmental discontinuities in the middle months of the first year.

Structure, variability, and development.

Abstract Seven studies (N = 1559) were conducted in exploration of the relation of age and stimulus structure to several propositions about human cognitive uncertainty ( Munsinger and Kessen, 1964; Munsinger and Kessen, 1966a ). Children and adults responded to figures differing in variability (information) and in symmetry (redundancy) by estimating variability, by categorizing the stimulus materials, and by the expression of preference. In general, (1) estimation and categorization are more accurate with increasing age, (2) practice in estimation and categorization is more useful for older children than for younger children, (3) older children and adults estimate low variability figures more accurately than high variability figures, (4) practice in estimation and categorization with asymmetrical figures transfers well to asymmetrical figures but not to symmetrical figures, (5) the relation between accuracy of categorization and variability of stimuli is a U-shaped function while the relation between accuracy of estimation and variability of stimuli is a decreasing linear function, (6) young children are insensitive to symmetrical redundancy of stimuli, and (7) human response to symmetrical redundancy is not a simple function of information reduction.