Developmental Changes in Infants\' Visual Response to Temporal Frequency

August 24, 2017 | Autor: David Lewkowicz | Categoria: Perception, Perceptual Development, Child Development, Infancy
Share Embed


Descrição do Produto

Copyright 1985 by the American Psychological Association, Inc. 0012-I649/85/S00.75

Developmental Psychology 1985, Vol. 21, No. 5,858-865

Developmental Changes in Infants' Visual Response to Temporal Frequency David J. Lewkowicz Illinois Institute for Developmental Disabilities and University of Illinois, Chicago To investigate the development of visual responsiveness to variations in temporal frequency three studies were carried out at 4 and at 6 months of age. In the first two studies infants viewed pairs of identical check patterns flashing at 2, 4, and 8 Hz, and their visual preferences were observed. In the first study, where rate covaried with duty cycle (i.e., overall intensity), both age groups exhibited differential fixation as a function of frequency. To find out whether intensity variations were responsible for these results, in the second study intensity variations were eliminated (by equating duty cycle at 50%), and only frequency was varied. This manipulation resulted in the elimination of the previously observed differential response in the 4-month-olds but had no effect on the 6-month-olds' differential response. As a final check on the role of intensity, in the third study overall intensity was varied while temporal frequency was kept constant. Results showed that 4-month-old infants continued to respond differentially whereas 6-month-olds did not. In sum, these data show that at 4 months of age infants attend to the overall intensity of stimulation and ignore its temporal characteristics, whereas at 6 months of age they no longer attend to the overall intensity and instead attend to temporal frequency per se.

Most of our knowledge concerning the development of visual functions and their underlying mechanisms comes from studies of infants' response to static stimuli. However, much of the information available to the infant is dynamic in nature. In fact, temporal change is so pervasive that many of our cognitive, linguistic, perceptual, and affective functions are dependent on our ability to respond to it. Although temporal change may be described by reference to just two simple stimulus attributes, namely, succession and duration, manipulation ofjust these two attributes can yield any degree of temporal complexity (Fraisse, 1963). At the simplest level, the attribute of succession can give rise to periodicity, which may be defined as the repeated occurrence of an event with a constant interevent interval. Because the ability to detect periodicity is a basic property of sensory function, it is important to examine both the development of I owe a special debt of gratitude to Deborah Holmes and Jill Nagy Reich for their generosity with space and equipment. I thank Bernard Karmel and Rathe Karrer for helpful comments on a previous veraion of the manuscript. Requests for reprints should be addressed to David J. Lewkowicz, Illinois Institute for Developmental Disabilities, 1640 W. Roosevelt Rd., Chicago, Illinois 60608.

infants' capacity to respond to periodic variations as well as the mechanisms underlying such capacities. Only a handful of behavioral studies (Gardner & Karmel, 1981; 1984; Gardner & Turkewitz, 1982; Karmel, Lester, McCarvill, Brown, & Hofmann, 1977, Nystrom, Hansson, & MarkJund, 1975) have examined infants' visual preferences for temporal frequencies that are below their critical fusion frequency (Regal, 1981). In general, the resultsfromthese studies indicate that temporal frequency is a potent and highly engaging stimulus attribute for young infants and that attentional responses to different frequencies are highly differentiated. However, no studies to date have attempted to determine what dimensions of temporally modulated stimulation are responsible for the differential responsiveness observed and whether there are developmental changes in the relative importance of different temporal dimensions. In addition, no studies of responsiveness to temporally modulated stimulation have been done with infants older than 3 months of age. In order to understand the mechanisms that underlie responsiveness to temporal frequency, it mustfirstbe noted that there are two different ways in which temporal frequency may be

858

VISUAL RESPONSE TO TEMPORAL FREQUENCY

modulated. One consists of varying frequency while keeping intensity (i.e., the overall or total amount of stimulation presented over a unit of time) constant across the different frequencies. This may be accomplished by having an equal "on/off" stimulus ratio at each frequency, respectively (e.g., 250 ms on and 250 ms off for a 2-Hz stimulus; 125 ms on and 125 ms off for a 4-Hz stimulus). The problem with this method, however, is that although it permits the equation of intensity across different frequencies, the duration of an individual stimulus is different at each frequency. The second method involves modulating the repetition rate of a stimulus whose duration is constant across different temporal frequencies (e.g., 50 rns on and 450 ms off for a 2-Hz stimulus; 50 ms on and 200 ms off for a 4-Hz stimulus; 50 ms on and 75 ms off for an 8-Hz stimulus). The problem with this method, of course, is that intensity will increase with increases in frequency. Obviously, neither method in itself would permit one to draw unambiguous conclusions regarding the specific dimensions that underlie responsiveness.

859

arately in 4- and 6-month-old infants. These two age groups were chosen partly because no other studies have examined infants' response to temporal frequency past the third month of life and partly because this age span is characterized by a great deal of change in sensory/ perceptual functioning (Cohen, DeLoache, & Strauss, 1979). General Method Apparatus and Stimuli During testing the infant sat inside a three-sided chamber designed so that the two lateral walls obstructed his or her view of the laboratory. The wall facing the infant was covered with black posterboard and had two openings each covered with a light diffiiser (milk-white plexiglass). Each diffuser was covered with an identical transparency of a black-and-white random check pattern. This check pattern was a replica of the random % inch pattern used by Karmel (1969). Each opening measured 15 cm X 15 cm, and the inner edges of the openings were 23.5 cm apart. At a viewing distance of 43 cm each checkerboard subtended 19°18' of visual angle, and each check within the checkerboard subtended 2°9' of visual angle (check size 1.27 cm). Each check pattern was lighted from behind the diffuser by two 14watt white fluorescent bulbs whose operation was silent and which were housed inside a 50 cm X 18 cm X 23 cm box. To permit an essentially instantaneous onset, these bulbs were kept "warm" by a 9 VDC current during the "off" periods. To light the bulbs a 300 VDC current was applied to them. The luminance of the visual stimuli was 21.92 cd/m 2 as measured on the transparent part of the checkerboard pattern. The contrast between the black and white checks was 1.0. Visual fixations were observed through a .64 cm peephole located in the center of the wall facing the infant, and the duration of each fixation was recorded on an Apple 11+ computer (because of the presence of discriminable corneal reflections and because the observations were done on-line, it was not possible to make the observer blind with respect to the stimuli). A set of five colored light emitting diodes arranged in a cross configuration was located in the center between the two visual stimuli and was used to attract the infants1 attention during the interstimulus intervals. A custom-built pulse generator was used to control the temporal parameters of the stimuli. Two independent channels of the generator controlled the "on" and "off" periods of each of the two visual stimuli. Gating of the stimuli was accomplished by having the square-wave output of each channel of the generator drive a silent relay, which in turn controlled the onset of the stimuli. Different aspects of the temporal distribution of stimulation were modulated in each study. Regardless of the temporal attribute being varied, there were three levels of that attribute in each study.

It is possible, however, to disentangle the contribution that each dimension makes to responsiveness through a set of convergent operations. Thus, if a differential response to frequency is obtained when intensity covaries with it but not when intensity is equated, then this would suggest that intensity is the dimension being attended to. This possibility could be further borne out by the finding of differential responsiveness to intensity variations when they are presented at a constant frequency. In contrast, if a differential response to frequency is obtained regardless of whether intensity covaries with it or not, and if no differential responsiveness is obtained when only intensity is manipulated, then this would suggest that temporal frequency is the dimension being attended to. The current set of studies was designed to investigate the mechanisms underlying responsiveness to temporal variations in frequency by using the set of convergent operations outlined above. Consequently, visual preferences for (a) frequency variations accompanied by intensity variations (Studies 1 and 4), (b) frequency variations equated in Procedure terms of intensity (Studies 2 and 5), and (c) Testing took place in a dimly illuminated room. The intensity variations equated in terms of fre- ambient sound pressure level in the room, as measured at quency (Studies 3 and 6) were examined sep- the infant's ear, was 46 dB (re .0002 dynes/cm2, A scale).

860

DAVID J. LEWKOWICZ

The 4-month-old infants were placed in a semireclining position in a commercially available infant seat. The 6-month-old infants were held on the parent's lap during the test. Although the parent was able to see the stimuli, he or she was blind with respect to the purpose of the study. In addition, it was not possible for the observer to see the parent's face, thus eliminating the possibility that the observer relied on the parent's visual behavior to make judgments of the infant's behavior. All infants were seated approximately 43 cm from the stimuli. The stimulus pair was presented only when the infant fixated the colored lights in the center between the stimuli. As soon as the infant's gaze was judged to be in the center, the trial was initiated by turning the central fixation display off and turning the stimuli on. Each group was administered a series of six 15-s trials consisting of the presentation of all possible pairs of the three levels of a given attribute, counterbalanced for side. To counterbalance for order effects, each infant was administered these six trials according to one of six possible orders. Across these six orders each pair appeared an equal number of times at each ordinal position, and a given stimulus was not followed by itself on the same side.

Four-Month-Old Infants: Study 1 Method Subjects. Separate groups of 12 infants each were tested in Studies 1 through 3.1 In all, there were 20 boys and 16 girls, who ranged in age from 16 weeks and 6 days to 19 weeks and 5 days (mean age = 18 weeks and 2 days). They were all full term at the time of birth and healthy at the time of testing. Seventeen additional 4-month-olds were tested but were excluded from data analysis due to equipment failure (3), fussing and crying (13), or experimenter error (1). Apparatus and stimuli. Three temporal frequencies of 2, 4, and 8 Hz were used in this study. These particular frequences were chosen because they are well below infants' critical fusion frequency, which is estimated to be 51.5 Hz by 3 months of age (Regal, 1981), and because they are the same frequencies that were used in the previously cited preference studies. It should be noted that each successive frequency represents a doubling of the lower frequency and that on the Iog2 scale they are equidistant from one another. In this study the duration of the "on" portion of the cycle was 50 ms for all three frequencies, whereas the duration of the "off" period varied. The duration of the "off" periods and a schematic representation of the temporal distribution of the stimuli used in this study may be seen in Figure 1. In addition, the duty cycle of each stimulus is also shown. Duty cycle is a convenient metric for expressing the intensity of stimulation. Its main advantage is that it is independent of the unit of time over which stimulation is being presented and is defined as the proportion of time the stimulus is "on" relative to the total amount of time required for completion of one cycle (a cycle consists of a single "on" and "off" phase of the stimulus).

Results and Discussion Analyses of the total duration of looking indicated a monotonic increase as a function of

temporal frequency (see Figure 2A). A oneway repeated measures analysis of variance (ANOVA) indicated that the main effect of frequency was significant, F(2, 22) = 17.94, p < .001. An analysis for orthogonal components of trend indicated that the relationship between visual attention and frequency was best described by a linear trend, F{\t 11) = 35.74, p
Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.