Over the past 25 years, there has been an explosion of new information about the maturation of vision during infancy. A growing appreciation of the role of visual experience during infancy in shaping the functional organization of the maturing visual system has driven a research effort to define the normal course of maturation and its susceptibility to disruption by early abnormal visual experience. As this research effort has bridged the transition from the laboratory to the clinical setting, the success of various treatment protocols for remediation following abnormal visual experience has come under evaluation, both in clinical trials and in clinical care of individual patients.

The study of infant visual maturation includes multiple approaches, including electrophysiology and psychophysics. Even in assessing a single visual dimension, such as visual acuity, the combination of multiple tests can provide more complete information about the infant's status than any one test alone. For example, while it is difficult to record pattern visual evoked potentials (VEPs) in patients with nystagmus or shunts, psychophysical assessment of acuity via preferential looking has a high success rate among these patients.^2,5,14 On the other hand, psychophysical testing may be insensitive to macular dysfunction and strabismic amblyopia,^9,24–26 while pattern VEPs are uniquely sensitive to these disorders because of the predominance of the central visual field in the responses.

Infant visual acuity testing is unusual in that it has bridged the gap between the laboratory and the clinic. Availability of the Teller Acuity Cards⁴⁵ has enabled psychophysical assessment of infant visual acuity during routine office examinations. While the additional expense and expertise associated with electrophysiological approaches to infant acuity assessment have limited their availability primarily to medical school hospitals and clinics, infant VEP acuity is now widely used both as an outcome measure for randomized clinical trials and for individual patient assessment. The focus of electrophysiological testing of infant visual acuity has shifted toward the use of the sweep VEP³⁰ in recent years. The sweep VEP is based on Fourier analysis of steady-state VEPs elicited by periodic stimulus changes, such as pattern reversal of a sine wave grating pattern. The steady-state VEP response is composed of a series of harmonically related components that are multiples of the frequency at which contrast reversal or pattern onset occurs.

To measure acuity by using the steady-state VEP, for example, the test protocol includes a series of checkerboard patterns or sine wave gratings that range from coarse to fine (low to high spatial frequency). For each pattern, 50–100 steady-state VEP responses are recorded, and the average amplitude is determined. By examining the relationship between check size (or spatial frequency) and response amplitude, the check size that corresponds to zero amplitude can be extrapolated as an acuity estimate. Ideally, many check sizes (or spatial frequencies) would be evaluated to pinpoint the exact size at which a reliable VEP could no longer be recorded. With the limited attention span of infants, this is rarely possible. In fact, in the now classic studies of the maturation of visual acuity during infancy that were conducted during the mid-1970s by Sokol⁴¹ and by Marg et al.,²³ only four to six large-to-moderate pattern element sizes were included in each acuity test.

On the other hand, sweep VEP protocols usually present 10–20 spatial frequencies to the infant in rapid succession during a 10-second sweep. By using Fourier analytic techniques to extract the VEP responses to each of the brief stimuli (specifically, the amplitude and phase of the harmonics of the stimulation rate), sufficient information can be obtained from the VEP records to estimate visual acuity from only a few brief test trials. This technique has three significant advantages over the more classical methods. First, test time is greatly reduced. The classic steady-state VEP acuity paradigms often required an hour or more to obtain data for just a few check sizes, while data for 10–20 spatial frequencies can be collected within 15 minutes by using the sweep protocol. Second, the infant's behavioral state changes little during the recording session; since behavioral state can influence the quality of the data obtained, brief test protocols provide a major advantage. Third, since many more spatial frequencies can be included in the test protocol, close spacing in the series can be used, and a more accurate estimate of acuity can be obtained. Figure 23.1 illustrates two potential pitfalls in acuity estimation associated with the classic VEP acuity protocols that are overcome by the improved sampling of the amplitude versus spatial frequency function of the sweep VEP method.

Figure 23.1.  

Potential pitfalls in acuity estimation caused by limited sampling of the amplitude versus spatial frequency function in classic VEP acuity protocols (panels A and C) that are overcome by the dense sampling and extended sampling range of the sweep VEP (panels B and D). Amplitude and phase of the VEP response (solid circles) are plotted along with amplitude at adjacent nonharmonic noise frequencies (open circles). Panel A illustrates a potential error in the acuity estimate that may occur if only low to moderate spatial frequencies are sampled (e.g., ≤8c/deg) and the slope of the descending limb of the amplitude versus spatial frequency function is not constant. Panel B shows the more accurate acuity estimate that is derived when data are available for a larger range of spatial frequencies (1–20c/deg). Panel C illustrates a limited sampling range (e.g., ≤8c/deg) may also result in a poor acuity estimate because of missing a second peak in the amplitude versus spatial frequency function. Panel D shows the more accurate acuity estimate that is derived when data are available for a larger range of spatial frequencies (1–20c/deg).