INTERNATIONAL JOURNAL OF PSYCHOPHYSIOLOGY International Journal of Psychophysiology 16 (1994) 209-218
Visual electrophysiology to achromatic and chromatic in premature and full term infants Gillian A. Rudduck,
Graham
F.A. Harding
stimuli
*
Department of Vi.sionSciences, Aston Unicersity, Aston Triangle, Birmingham 84 7ET, UK (Accepted 17 January 1994)
Abstract We have previously reported on the development of black/white pattern reversal VEPs in premature babies of more than 30 weeks post-menstrual age (PMA). Unlike the flash VEP, the pattern reversal VEP shows a similar morphology to that of the full term infant and the major positive component (Pl) decreased in latency with increasing PMA. The Nl and N2 components were more likely to be present with increasing maturity. In our present study we are examining the development of the transient chromatic pattern VEP. In order to produce a purely chromatic stimulus it is necessary to rcmovc luminance cuts. Based on forced choice preferential looking we developed a method of determining the isoluminant point for infants. Preference was tested for a flickering sinusoidal red and green grating over the uniform field. As sensitivity for chromatic flicker is much poorer than for luminance flicker, sensitivity is expected to be least when the residual luminance variation in the stimulus is at a minimum. The red/green luminance ratio at which this occurs represents the isoluminant point. From this method we found the subjective isoluminant point for infants of 2-3 months of age to be very close to the objective measure of isoluminance. Using this information, pattern reversal VEPs to 20 chromatic red/green and achromatic checks were studied and it would appear that pattern reversal VEPs cannot be obtained to isoluminant stimuli before 7 weeks chronological age. Key words: Infant;
Premature;
Development;
VEP; Achromatic;
1. Introduction
Chromatic;
The ture
of the visual evoked potential in premature and neonate infants, the flash and achromatic pattern reversal VEP were recorded from infants of 32-35 weeks postmenstrual age between their first and 18th week from birth [l]. In our
previous
studies
* Corresponding author. Tel.: (021) 359 3611; Fax: (021) 333 4220.
Isoluminancc
development
infants
is well
of the known.
flash
VEP
In premature
in premababies
less than 35 weeks post-menstrual age the flash VEP shows a mainly negative wave known as Nl. With increasing post-menstrual age (PMA) a more obvious positivity is seen which gradually reduces in latency. The pattern reversal VEP however does not show the early negativity but always consists of a positive component which shows a gradual reduction in latency from greater than 300 to 200 ms at around term and then continues to reduce in latency (Fig. 1). In most of the
0167.8760/94/$07.00 0 1994 Elsevier Science B.V. All rights reserved SSDZ 0167.8760(94)00019-B
210
G.A. Rudduck, G.F.A. Hurding /Internationul
Journal
of Psychophysiolog) 16 (1494) 204-218
45.5 WEEKS PMA.
Pl Fig. 1. The most consistent feature of the pattern reversal response was found to be the positive wave (Pl) with a mean latency of 345 ms at 30.5 weeks PMA, the latency of this wave shortens as the infant. grows older and the wave becomes more sharply defined.
G.A. Rudduck, G.F.A. Harding /International
premature babies even beyond full term, that is 40 weeks PMA, the positive component continues to reduce in latency with a slight increase in amplitude. This component becomes the PlOO component described in studies of full term infants and children. The pattern-reversal Pl component is apparent from 32 weeks at around 300 ms. With increasing PMA the latency reduces to around 250 ms at term and continues to reduce until at 3 months past full term the latency is around 150 ms [2]. Few studies have been carried out into the development of the chromatic response of infants. Dobson [3] measured the implicit time and amplitude of flash VEPs to monochromatic light and plotted implicit time as a function of the log energy of the stimulus. From this a spectral sensitivity curve for 2-month-old infants was constructed. Moskowitz-Cook [4] undertook a similar investigation for infants between the ages of 3 and 22 weeks, using a pattern reversal stimuli with an alternation rate of 4 Hz (8 reversals per second). Checkerboards composed of 45’ black and coloured checks were used to elicit monopolar VEPs. Ten wavelengths were investigated over a range of radiance levels and clear responses were obtainable from all stimuli in infants above 7 weeks of age. Allen and co-workers [5] used sweep VEPs with sinusoidal red/green gratings to study luminance and chromatic contrast thresholds in infants. Clear responses to isoluminant stimuli could be recorded in all infants investigated, including a 2-week-old infant, with clear performance minima at isoluminance. Morrone and colleagues [6] also used the steady-state technique to study the characteristics of the chromatic VEP in infants. VEPs were recorded as a function of the proportion of red in the stimulus [7] and a performance minimum could be seen at around photometric isoluminance. No response to isoluminant stimuli could be recorded in any infant below 8 weeks of age. In any study to assess the development of colour vision, it is necessary to use a stimulus in which luminance cues have been eliminated, thus revealing a purely chromatic response. Isoluminance, the point at which two components of different wavelengths are perceived to be of equal
Journal
of Psychophysiology 16 (1994) 209-218
21 I
brightness, is a subjective response and has been assessed in adults in a variety of ways, however the major drawback of all these techniques with respect to infants is that they require some subjective decision making. Maurer and co-workers 181 adapted Anstis and Cavanagh [91 Minimum Motion Technique to assess isoluminance in human infants by eliciting optokinetic nystagmus (OKN) to the apparent motion. In order to produce chromatic stimuli for VEP studies of the visual system we have utilised another form of visual function assessment. Forced choice preferential looking, introduced by Teller [lo] has been widely and successfully used in the assessment of visual acuity and other aspects of visual function; e.g. colour [11,12], stereopsis [13,14], contrast sensitivity [15-171. Thompson and Drasdo [l&19] adapted the forced choice testing protocol and it is this method that we have utilised to determine the isoluminant point of infants from 1-3month-olds using a flickering stimulus [20].
2. Isoluminant
study
2.1. Materials and methods Human infants of 1, 2 and 3 months of age were recruited via their general medical practitioner or health visitor. All the babies had been assessed as normal at health centres in Birmingham. A total of 26 individual infants took part; 5 infants attended for two visits, separated by 4 weeks, 1 3-month-old attending twice within 2 weeks, the other 19 attending only once (n = 31). All the babies were full term by maternal report (gestational age 2 38 weeks and birth weight 2 2500 g> with no family history of colour deficiency. Five 1-month-olds (mean age 35 days, range 22-41 days), 13 2-month-olds (mean age 58 days, range 51-68 days) and 13 3-month-olds (mean age 80 days, range 78-98 days) were investigated. The study was approved by both the University and Health District Ethical Committees. Informed consent was obtained from the mothers after the technique had been fully explained. The isoluminant points of two adult con-
trols (G.R. and R.D.) were assessed using the same spatial frequency. Stimuli were produced by a Venus stimulator (Neuroscientific) controlled by application software. The frame rate is 119 Hz with independent control of red, green and blue guns. The output of the blue gun was removed. The display was presented on a standard CRT colour monitor (CIE x and y co-ordinates, 0.591,0.373 for red and 0.297,0.596 for green). The stimulus was split in half vertically and a pattern appeared randomly in the left or right half screen. The pattern was composed of sinusoidal red and green horizontal gratings presented 180” out of phase with each other. The individual red and green constituent gratings were made to counterphase 180” out of phase in a sinusoidal manner at a temporal frequency of 16 Hz. At high temporal frequencies the detection of the temporal modulation, i.e. flicker, is dominated by any residual luminance information in the stimulus. Flicker will thus reduce to a minimum when there is no detectable luminance difference between the two gratings. The mean luminance of the pattern was maintained at 10 cdm-’ but the individual luminances of the gratings could be varied relative to one another to produce different ratios of red to green. At a red/green ratio of 1.0 both the gratings were of equal objective luminance, measured photometrically. The other half of the screen contained a uniform field having the same mixture of red and green and the same luminance as the gratings producing a neutral brown-yellow appearance. For the infant assessment, the observer and the stimulus generator were hidden by a black screen with two apertures, one for the screen and one for observing the infants’ rcactions. The edges of the stimulus and the central division of the pattern were masked to produce two panels visual angle 20” X 9” separated by 0.6” when viewed at 50 cm. The spatial frequency of the gratings in the stimulus were chosen to be of a size appropriate for infant visual acuity at birth [21,22]. Morrone and co-workers [6] found very low spatial frequencies were required to elicit VEP responses to chromatic stimuli in young infants ( < 8 weeks). The isoluminant point for individuals and groups
of infants were thus estimated with stimuli of both 0.1 and 0.4 cpd. Incorrect refraction of 1 to 2.OOD has been demonstrated to have an effect on the point of isoluminance at spatial frequencies of 4 cpd and blurring of 4 to 8.OOD affects isoluminance at lower spatial frequencies [8]. It was thus desirable to have a working distance that would produce minimum blur for the subjects, as it was impractical to refract and correct each individual. Newborns are capable of adjusting their accommodation and do so most accurately for close targets (5 75 cm) [23]. Employing a short viewing distance, i.e. 50 cm from the target and low spatial frequencies (i.e., 0.1 cpd), the effect of blur on isoluminance induced by inaccurate accommodation and refractive error was taken to be negligible. The test was carried out in a dark room with each infant sitting on the holder’s (usually the mother) lap. Infants were allowed to bottle feed or suck a dummy during the test. The observer could not see the stimulus or the holder’s face and the holder was instructed to keep the baby facing the centre of the screen. To attract attention, the observer tapped on the back of the screen prior to grating presentation. The observer selected a spatial frequency (initially 0.1 cpd) and a chromatic modulation contrast (amplitude of modulation). Trials commenced from a red green luminance ratio randomly selected from one of two ratio ranges (0.6-0.72) and (1.32- 1.58). Two staircases were carried out at each chromatic contrast level, with one starting point from each range, with no relationship between the two starting ratios at each level. Starting points were randomly varied between chromatic contrast levels and infants. If the observer correctly identified the position of the flickering grating by interpretation of the infants looking behaviour on two successive trials the red green luminance ratio was adjusted to bc nearer to 1.0 by a standard log unit at the next presentation. This was + 0.015 log units (1 step) for the first step in any sequence and then iO.03 (2 steps) for the subsequent changes. If an incorrect response from the observer occurred within the first two trials the procedure was interrupted and a new starting ratio chosen which was further
G.A. Rudduck, G.F.A. HardinR/International
away from 1.0. Whenever an incorrect response was made within a sequence the observer reversed the sequence by 3 steps and the sequence recommenced. This 3 step change was required (due to the program sequence) to allow retest of the steps around the reversal point. Repeated trials were conducted at red green ratios around the point at which the staircase reversed until a minimum of six trials had been carried out at the ratios around the point of reversal. A written record of the observers score at each ratio was made and from this the > 70% correct level was calculated. The end point of the staircase was taken as the red green ratio at which the observer’s score fell below 70%. Up to six contrast levels distributed with logarithmic equality between 94 and 63% were assessed (100% was the maximum chromatic contrast available). At the highest contrast levels the observer correctly estimated the position of the grating for all luminance ratios presented on repeated trials to the 2- and 3-month-old infants. This was noted and hence for six contrast levels assessed only five sets of no preference points for an individual were obtained. The mid-points of the pairs of staircase end points at each amplitude of chromatic modulation (chromatic contrast level) where found to have a very small variation for any one infant and the mean of these mid-points was accepted as the final isoluminant point for an individual.
Journal of Psychophysiology 16 (1994) 209-218
213
The mean isoluminant red/green ratio for a spatial frequency of 0.1 cpd using this technique was found to occur at 1.00 + 0.03 for the sample of 12 3-month-olds, 1.03 & 0.03 for the 12 2month-olds and at 1.01 f 0.02 for a sample of 5 l-month-old infants. For a spatial frequency of 0.4 cpd this ratio was 1.03 & 0.03 for a sample of 8 3-month-old infants and 1.03 k 0.03 for a group of 5 2-month-olds. No l-month-old was successfully assessed at this spatial frequency. The distribution of isoluminant points with age can be seen in Fig. 2. The isoluminant points for the adult subjects were G.R. 0.98 and R.D. 0.98 for 0.1 cpd and G.R. 0.97 and R.D. 0.96 for 0.4 cpd. The infant who attended twice within 2 weeks allowed us to assess the repeatability of the technique. All end point estimations varied by less than two steps between visits and final isoluminant points were 1.00 and 0.99 for this infant. 2.3. Discussion The method we have presented here strates that a technique of preferential
demonlooking
Gi
2.2. Results Of the 31 infants assessed at separate sessions, a total of 29 sets of completed data for 0.1 cpd were collected at separate visits, with only 2 infants becoming too sleepy for the recording to be continued. In addition, where infants were cooperative and attentive we repeated the procedure for 0.4 cpd. Complete sets of data were collected from all 13 of the infants tested at 0.4 cpd. All of these infants were between 2 and 3 months of age. The isoluminant point determined from these infants at 0.4 cpd is very similar to that at 0.1 cpd. No assessment at 0.4 cpd was possible with the youngest infants due to unto-operation following the initial trials.
5 0
60
5 40 -
1 20
I
0 1
1.0
0.9
1 .l *
INCREASING
GREEN RED
TO GREEN
INCREASING
RED
RATIO
Fig. 2. Distribution of red to green luminance ratios giving isoluminance at 0.1 cpd for 29 human infants aged between 22 and 98 days. lsoluminant ratios for 3-month-olds (n = 12) CO), 2-month-olds (n = 12) to), and I-month-olds (n = 5) (0) are shown as a function of age. The mean isoluminant red to green ratio for each age group is represented as the large open squares with error bars showing standard deviation.
214
G.A. Rudduck,
G.F.A. Harding /Intwnutional
can be employed to determine isoluminance in infants between 1 and 3 months of age. Fig. 2 shows that the distribution of isoluminant points for the group of individuals tested is very narrow with all points lying within 0.1 log units of 1.0 (photometric isoluminance). Colour deficiency has been shown to shift the isoluminant point in adults [8], but due to the small variation in our sample group it was presumed that no infant tested was colour deficient. The isoluminant points of the infants and the two adult controls were similar. This finding is in agreement with other studies who have used large fields to determine the isoluminant ratio by the minimum motion technique in infants and adults [S]. This similarity has been attributed to the contribution of the peripheral retina and hence does not reflect the difference found between adult and infant spectral sensitivity curves [24]. Due to the limitation of screen size (i.e. 20’) imposed by the width of the gratings necessary to produce a spatial frequency of 0.1 cpd at a distance that the procedure could reliably be undertaken it was accepted that the similarity between our findings for both adults and infants may also be due to large peripheral retinal contribution. The isoluminant point determined for the sample at 0.4 cpd was highly similar to that at 0.1 cpd. This is in agreement with the work of Mullen [7] who showed isoluminance to be unaffected by change of low spatial frequencies. Peeples and Teller [25] discovered that infants can distinguish a 0.1 log unit brightness difference between two components of a stimulus. Although the variation of isoluminant ratios across the whole range of infants tested is less than 0.1 log units it may be unsound to assume that the brightness match for one individual infant will be sufficient for another. This suggests that isoluminance should be determined for each individual. However, this technique is very time consuming and causes a reduction in subject co-operation with prolonged testing. This would thus make any subsequent clinical VEP assessment very difficult. In view of this, an alternative solution to the luminance problem in chromatic studies was sought for the VEP study.
Journul of Psychophysiology
3. Chromatic
16 (1094) 209-218
transient
VEP study
3. I. introduction Due to the problems that present with isoluminance determination, the paradigm of the brightness control developed by Peeples and Teller [25] was adopted for use in the study of the transient VEP. They found that a white bar presented on a white screen was indistinguishable over a range of 0.1 log units for the infant. From this they concluded that infants are insensitive to brightness differences of less than 0.1 log units. Systematic variation of luminance around an approximation to an infant brightness match should present the infant at some point with luminances of red and green that differ indiscriminately, as long as the variation step is less that 0.1 log units [26]. If a visual evoked potential can be elicited for all the relative luminances of red to green one may assume that the infant has the ability to preserve wavelength information to a cortical level. 3.2. Muter-i& and methods Forty infants between the ages of 1 and 13 weeks post-term age (PTA), i.e. assuming term to be 40 weeks post-menstrual age, were recruited into the study. Written consent was obtained from the parents of all infants entering the study. In this study, the first order approximation to infant brightness match was taken as objective photometric isoluminance (i.e. I .(I). The preccding study had demonstrated that all the sample of 29 infants had isoluminant points that fell less than 0.1 log units away from this point. Seven red/green luminance ratios, symmetrically distributed around and including this first match approximation of 1.0 were chosen. Five of these points were separated by less than 0.1 log units with the overall brightness difference range covering 0.5 log units. The red to green luminance ratios arbitrarily selected were 0.5, 0.8, 0.0, 1.0, 1.1, 1.2, 1.5. Pattern reversal checkerboards were produced by the Neuroscientific Venus system. This was controlled by application software which allowed the independent control of the red, green
G.A. Rudduck,
G. F.A. Harding/International
and blue guns and the output of the blue gun was removed for the red/green chromatic stimuli. Large checks have been shown to elicit clear pattern reversal VEPs to achromatic stimuli in infants as young as 30.5 weeks post-menstrual age [l]. Previous investigation of the steady state VEP to chromatic stimuli have shown low chromatic acuity in infants [61 and the effects of chromatic aberration is negligible at low spatial frequencies [27]. For these reasons, checkerboards composed of checks of 2” angular subtense (0.25 cpd) were chosen to record the pattern reversal VEP. Independent control of the red and green guns enabled the production of a series of checkerboard stimuli at a range of red/green luminance ratios. These were programmed and stored on disk for instant retrieval and presentation. The mean luminance of the chromatic stimulus was constant at 10 cdmm2 but relative red to green luminances could be adjusted to produce a series of red/green luminance ratios. A high chromatic contrast (amplitude of modulation) level of 90% was used. This could be further increased up to 100% in the cases where no response could be elicited to a chromatic stimulus. A black/white checkerboard of 90% Michelson contrast and white square luminance 20 cdmp2, was included throughout the recording sessions to assess the presence, quality and repeatability of the pattern reversal response. The pattern reversed at a rate of 1 cycle/s (1 Hz or 2 reversal/s) and a trigger pulse was developed at the beginning of each cycle to initiate the averaging procedure. Thirty responses were averaged using the Biologic Traveller with pass band of 0.3-30 Hz. Thirty sweeps were averaged for the infants as this was found to be adequate to record the response. There was no significant improvement if more than 30 responses were averaged. The analysis time was 800 ms with an Inter-stimulus interval of 1 s. Two to three averages were recorded for each stimulus in order to estimate the reliability of the response and stimulus runs were compared to non-stimulus runs. Pattern reversal VEPs were recorded using silver/silver chloride electrodes which were positioned in accordance with the International lo-20 System. Two channels were recorded from 02
Journal
ofPsychophysiolo~
16 (1994) 209-218
215
(active) referenced to C4 and 01 (active) referenced to C3. Cz was used as the earth. In order to reduce skin resistances, Omniprep was applied to the electrode position prior to electrode placement. Resistances of 5 kR or less were always possible and once the electrodes were allowed to settle on the scalp, resistances tended to decrease further. Electrodes were initially held in place by Blenderm tape but where recording sessions were prolonged, the electrodes tended to slip. To prevent this, collodion glue was applied to the tape and the electrode re-placed. All infants were initially assessed with the achromatic stimulus and only if a clear response could be obtained from this stimulus was recording continued with chromatic stimuli. As much of a young infants time is spent sleeping some recording sessions had to be interrupted to allow for this. Recording by the averager could also be interrupted during a run if the cornea1 reflection of the stimulus was not centred on the infant’s pupil. This was observed throughout the recording sessions by a separate observer who also aided in keeping the infant’s attention on the screen. A minimum of three different chromatic stimulus averages (2-3 runs at each) was accepted as a completed data set but where infant co-operation allowed, up to seven chromatic stimuli were used. Stimulus runs were compared to non-stimulus runs and this was of particular value when determining the presence or absence of a response. Fig. 3 shows two stimulus runs and a non-stimulus run demonstrating the presence and absence of a response to chromatic stimuli. 3.3. Results The wave form of the chromatic response showed a simple morphology of major positive component (Pl), comparable to that of the achromatic response, in the youngest infants. These responses were symmetrical across both channels and all chromatic responses were delayed and reduced in comparison to the achromatic. As the chromatic luminance ratio approached 1.0, this component consistently demonstrated an increased latency and reduced amplitude across the whole sample group. Fig. 4 shows the change of
G.A. Rudduck, G.F.A. Harding/International
216
latency and amplitude of the Pl component for one seven week old infant KB with respect to luminance ratio of the red and green checks. For all of those infants younger than 7 postterm weeks at the time of recording (n = ll>, this Pl component disappeared at a luminance ratio at, or close, to 1.0. No consistent VEP could be elicited to a red/green checkerboard at this ratio although all other stimuli produced clear responses. Fig. 5 shows averaged responses from one channel for a 6-week-old infant for a black/white stimulus and seven different ratios of red to green. As the luminance ratio approached isoluminance the response became delayed and reduced and eventually disappeared, reappearing only when the luminance ratio of the component checks was increased. If no response was obtained at a chromatic amplitude of modulation of 90% the run was repeated at 100%. In all cases increasing amplitude of modulation had no effect
Journal of Psychophysiology 16 (1994) 209-218
310 300
1
290 -j 280 270 260
1
250 + 240t
Pl
(R:G ratio0.5)
.I
.I
0.4
.I
0.6
.,.I
0 6
BW02
I.
8.
04
.I
1.0
R:G
ot a) chromatic response
,
BW 0 2
1.2
I
I
08
R:G
.I
.,
1.6
1 6
2 0
RATIO
‘06
.,
14
.
I
10
I
1.2
I
14
I
1.6
I
16
t
20
RATIO
Fig. 4. Line plot to show of latency and amplitude of the Pl component of the chromatic VEP to stimuli of different red/green luminance ratios for a 7-week-old infant (KB). The achromatic response is included for comparison. It can be seen that latency of the PI component increases at isoluminame with a corresponding decrease in amplitude. b)nochromatic response (R.G ratio1.0)
topv +
L
200 InSfC
Fig. 3. Comparison of chromatic stimulus runs producing (a) a pattern reversal VEP, (b) no response, and (c) a non-stimulus run.
A total of 39 infants over the age range 1-13 weeks were successfully assessed using this technique. Of these, 19 infants showed no reliable VEPs to red/green stimuli whose component checks had a luminance ratio close to that of objective isoluminance. The distribution of infants with and without response to one or more chromatic stimuli with respect to post-term age is shown in Fig. 6. All infants older than 7 weeks post-term (n = 1.5) demonstrated clear, repeatable responses to all chromatic stimuli including isoluminance. Five of the 13 7-week-old infants assessed demonstrated clear VEPs to all stimuli whereas no VEP could be recorded to one or
G.A. Rudduck, G.F.A. Harding/
International Journal of Psychophysiology 16 (1994) 209-218
217
more chromatic stimuli for the other 8 infants. It can be clearly seen that no infant demonstrated a transient chromatic visual evoked potential prior to their 7th post-term week and equally all infants older than 8 weeks produced chromatic potentials. 3.4. Discussion Previous studies have not reported on the transient isoluminant VEP in infants. The only previous studies were those of Allen et al. [.5] and Morrone et al. [6] but both of these groups used steady-state potentials. The transient VEP technique has the advantage that measurement can be made of both the amplitude and latency of the
fi
2
3
4
5 POST
6
7
8
TERM
AGE
9
10
11
12
13
(WEEKS)
Fig. 6. Distribution with age (PTA in weeks) of those infants showing responses to all chromatic stimuli presented (solid columns) (n = 20) and those who failed to show a response at one or more red/green luminance ratios (striped columns) (n = 19).
Black/white
0.9
1.0
-b---
1
1.1
Pl
1.2
1.5
Fig. 5. Change in morphology of chromatic VEP with change of R:G ratio for a 6-week-old infant. The achromatic response is shown above (BW) for comparison.
response and it can be seen from our results that latency is characteristically increased around isoluminance as the amplitude of the response decreases. It would therefore be possible to determine the isoluminant point using the VEP alone and thus future studies would not need the preceding FPL technique that we utilised. What was particularly startling was the remarkably narrow age range (6-8 weeks) at which a change from absence to presence of an isoluminant response took place. No infant less than 7 weeks showed an isoluminant response and yet all infants at 8 weeks produced a clear and consistent response at isoluminance. This finding would be entirely consistent with other studies of the development of colour vision. Two month olds tested behaviourally demonstrate chromatic discrimination and will reliably discriminate a range of colours from white [11,12,25,28], but few studies have convincingly demonstrated chromatic discrimination before this age [5,29]. However, this startling “switch-on” of the chromatic visual system has not previously been shown even allowing for the fact that the VEP represents a mass response. It is indeed surprising that this change over is so well defined but this could be consistent with some theories of the development of the visual system [30,31].
21x
GA
Rudduck,
G.F.A.
Hardinfi/Interncrtional
Acknowledgements The authors are very grateful to Dr. E. O’Brien and the health visitors at Newtown and Aston Health Centres, Birmingham, for their assistance in subject recruitment.
References [I] Harding,
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[IO]
[II]
[12]
[I31
[I41
G.F.A., Grose, .I., Wilton, A.Y. and Bissenden, J.G. (1989) The pattern reversal VEP in short gestation infants. Electroenceph. Clin. Neurophysiol., 74: 76-80. Grose, J., Harding, G.F.A., Wilton, A.Y. and Bissenden, J.G. (1989) The maturation of pattern reversal VEP and flash ERG in preterm infants. Clin. Vision Sci., 4: 239246. Dobson. V. (1976) Spectral sensitivity of the 2 month old infant as measured by the visually evoked cortical potential. Vision Res.. 16: 367-374. Moskowitz-Cook. A. (1979) Development of photopic spectral sensitivity in human infants. Vision Res., 19: 1133-1142. Allen, D.. Banks, MS., Norcia. A.M. and Shannon, L. (1990) Human infants VEP responses to isoluminant stimuli. Invest, Ophthalmol. Vis. Sci., 31 (SuppI.): 47. Morrone, M.C. and Burr, D.C. and Fiorentini, A. (1990) Development of contrast sensitivity and acuity of the infant colour system. Proc. R. Sot. Lond. Ser. B., 242: 1344139. Mullen, K.T. (1985) The contrast sensitivity of human colour vision to red-green and blue-yellow chromatic gratings. J. Physiol., 359: 381l400. Maurer. D., Lewis, T.L., Cavanagh, P. and Anstis, S. (1989) A new test of luminance efficiency for babies. Invest. Ophthalmol. Vis. Sci.. 30: 297-303. Anstis, S. and Cavanagh, P. (1983) A minimum motion technique for judging equiluminance. In: Mollen, J.D. and Sharpe, L.T. (Eds.), Colour Vision: Physiology and psychophysics, Academic Press, London/New York. Teller, D.Y. (1979) The forced choice preferential looking procedure: a psychophysical technique for use with human infants. Infant Behav. Dev., 2: 135-153. Hamer, R.D. Alexander, K.R. and Teller, D.Y. (1982) Rayleigh discrimination in young human infants. Vision Res.. 22: 575-588. Clavadetscher. J.E., Brown, A.M., Ankrum, C. and Teller, D.Y. (198X) Spectral sensitivity and chromatic discriminations in 3- and 7-week-old human infants. J. Opt. Sot. Am. [A], 5: 209332105. Held, R., Birch, E. and Gwiazda, J. (1980) Stereoacuity of human infants. Proc. Natl. Acad. Sci. USA, 77: 5.5725574. Birch, E.E., Gwiadza, J. and Held, R. (1982) Stereoacuity development for crossed and uncrossed disparities in human infants. Vision Res.. 23: 1019-1024.
Journul
of Psychophysiology
16 (19941 209-218
[15] Atkinson, J., Braddick, O.J. and Moar, K. (1977) Development of contrast sensitivity over the first 3 months of life in the human infant. Vision Res., 17: 1037-1044. [I61 Atkinson, J., Braddick, O.J. and Moar, K. (1977) Contrast sensitivity of the human infant for moving and static patterns. Vision Res., 17: 1045~1047. [17] Banks, M.S. and Salapatek, P. (1978) Acuity and contrast sensitivity in I-. 2- and 3-month-old human infants. Invest. Ophthalmol. Vis. Sci., 17: 361-365. [lS] Thompson, C.M. and Drasdo, N. (1984) The VEP and complementary techniques in paediatric visual assess ment. In: W.N. Charman (Ed.), The Frontiers of Optometry, Transactions of 1st Int. Congr., The British College of Ophthalmic Opticians (Optometrists), London. Vol. 2. pp. 63-78. 1191 Thompson, CM. and Drasdo, N. (1988) Clinical experience with preferential looking acuity tests in infants and young children. Ophthalmol. Physiol. Opt., 8: 309-321. [20] Rudduck, G.A., Whitaker, D. and Harding. G.F.A. (1992) Use of preferential looking to determine isoluminance in infants. Invest. Ophthalmol. Vis. Sci.. 33 (Suppl.): 112. [21] Yamamoto, M. and Brown, A.M. (1985) Vision testing of premature infants using the grating acuity cards. Fol. Ophthalmol. Jap., 36: 796-799. [22] Dohson, V.. McDonald. M.A., Kohl, P.. Stern. N., Samek, M. and Preston. K. (19X6) Visual acuity screening of infants and young children with the acuity card procedure. J. Am. Optom. Assoc.. 57: 2844289. [23] Braddick, O.J., Atkinson, J., French, J. and Howland, H.C. (1979) A photorefraction study of infant accommodation. Vis. Res., 19: 1319-1330. [24] Teller, D.Y. and Lindsey, D.T. (1989) Motion nulls for white versus isochromatic gratings in infants and adults. J. Opt. Sot. Am. [A], 6: 1945-1954 [25] Peeples. D.R. and Teller, D.Y. (1975) Colour vision and brightness discrimination in 2 month old human infants. Science, 18Y: 1102-l 103. [26] Teller, D.Y. and Bornstein, M.H. (lY87) Infant colour vision and colour perception. In: Salapatek. P. and Cohen, L. (Eds.), Handbook of Infant Perception. Academic Press, London/New York. [27] Flitcroft, D.I. (1989) The interactions between chromatic aberration, defocus and stimulus chromaticity: implications for visual physiology and calorimetry. Vision Res., 29: 3499360 [28] Allen. D., Banks, MS. and Schefrin, B. (1988) Chromatic discrimination in infants: a reexamination. Invest. Ophthalmol. Vis. Sci., 29 (SuppI.): 25. [29] Adams, R.J., Maurer, D. and Davis. M. (1986) Newborns’ discrimination of chromatic from achromatic stimuli. J. Exp. Child Psycho]., 41: 261-281. [30] Bronson, G. (lY74) The postnatal growth of visual capacity. Child Dev., 45: 873-890. [31] Atkinson, J. (1992) Early visual development: differential functioning of parvocellular and magnocellular pathways. Eye, 6: 1299135.