Faculty Bibliography

Temporal modulation transfer functions (MTF's) were recorded from the macula of nine normal subjects using focal electroretinography (FERG). An array of light emitting diodes (LED's) was used to experimentally manipulate stimulus temporal frequency, modulation depth, and mean luminance values. Two techniques were used to derive FERG modulation thresholds at several temporal frequencies: conventional averaging with extrapolation to a criterion amplitude, and a swept stimulus lock-in retrieval method. These two methods produced comparable results. The electrophysiologically derived MTF's were similar in shape to those obtained psychophysically. Six patients with retinal disease were also examined; all patients showed sensitivity losses which were most marked at the higher frequencies. Such losses tended to be greater in patients with poorer visual acuity

Contrast thresholds and acuity limits were measured in 4 observers with the swept visual evoked potential (VEP) technique. In this technique, grating contrast or grating spatial frequency is electronically varied while the subject's evoked response is retrieved in real time (without averaging). Contrast or spatial frequency variation make the stimulus vary in intensity; zero VEP response amplitude indicates the threshold intensity. Large shifts occur in the indicated threshold when stimulus sweep direction is reversed. Thresholds are always relatively elevated when the run begins with the strongest stimulus value. These shifts do not have a technical origin in the delay of the instrument (Nelson et al. 1984b). Here, it is shown that the shifts are due to orientation and spatial frequency selective adaptation, probably of cortical origin. Measureable adaptation is produced by momentary exposure to contrasts as low as 1.25%; nearly maximum adaptation (0.6 log units) is reached with 20% contrast. These findings support the concept of a contrast gain control mechanism in visual cortex, and pose practical problems for visual assessment with the evoked potential

We discuss the use of synchronous-demodulation (lock-in) techniques for evoked potential retrieval. Application to electronically swept visual displays is emphasized. These techniques permit a visual threshold to be specified in 20 s, but their application to visual assessment requires careful consideration of several problems, notably alleged delay in the instrumentation, specification of the baseline response level, and the nature of EEG interference. In addition, since stimulus waveform information is lost in all lock-in methods, questions concerning what activity is contributing to the measured response must be answered. A technique addressing these issues and combining phase-sensitive detection and vector computation is presented

Multiple sclerosis can produce highly selective losses in visual function. Psychophysical studies have demonstrated contrast sensitivity deficits for spatial frequencies or for stimulus orientations. Using real-time lock-in retrieval of the visual evoked potential, the authors measured contrast sensitivity in 15 cases with probable or definite multiple sclerosis and acuities of 20/40 or better. Sine-wave grating contrast threshold determinations for three spatial frequencies (1, 4, and 8 cycles/deg) and four orientations (0, 45, 90, and 135 deg) revealed contrast deficits in at least one spatial frequency and orientation in every case. In most cases the visual losses were spotty or multifocal, and not the same in both eyes. Some cases with highly selective patterns of orientation or spatial frequency losses were observed and are discussed in terms of involvement of cortical functional architecture in the disease

Electrophysiologic contrast sensitivity functions (CSF) have been estimated using lock-in amplifier signal retrieval of the visually evoked response (VER). These CSFs were compared with CSFs obtained psychophysically using the same stimulus conditions. The two measures of contrast sensitivity behave similarly in response to variations of temporal and spatial frequency. The major advantage of using real-time retrieval is speed. Threshold for a single spatiotemporal condition can be estimated in as little as 20 sec, making the application of electrophysiologic contrast sensitivity testing feasible for clinical populations

Reversing sine wave gratings were electronically swept in spatial frequency and contrast. The acuity limits and contrast thresholds of 4 observers were inferred from evoked potential stimulus-response functions elicited by these stimuli and retrieved with a quadrature lock-in amplifier. The evoked potential functions, linearized in the case of contrast by increasing contrast logarithmically with time, were extrapolated to the point of zero response. This point provides an electrophysiologically defined threshold value for acuity and for contrast. An oblique effect (superior sensitivity for HV-oriented gratings) could reliably be demonstrated in both acuity and contrast threshold performance. This oblique effect could readily be abolished under low spatial/high temporal frequency conditions. The findings are discussed in terms of shifting relative strengths of X and Y contributions to the steady-state evoked potential

Pattern evoked electroretinograms (PERG), diffuse flash electroretinograms (ERG) and visual evoked potentials were studied in patients with unilateral optic nerve disease. Patients with Snellen acuities of less than 6/30 did not have recordable PERGs in their affected eye, whereas their diffuse flash ERGs were normal. The VEPs were correspondingly reduced or absent when recorded from the poorer seeing eyes. A second group of patients with Snellen acuity between 6/6 and 6/30 in the involved eye showed reductions in the mean PERG amplitude of the affected as compared with the normal eyes. All affected eyes showed an abnormal contrast threshold measured with the PERG amplitude. Such results underscore the diagnostic value of the PERG in detecting even mildly affected cases of optic nerve disease