458 AM J OPTOM & PHYSIOL OPTICS

Vol. 65, No. 6

FIG. 3. Representative convergence performance

for each session of patient given automated convergence tralning using ADS (top panel) and a matched control patient exposed to identical target without increasing convergence demand (bottom panel). It is readily apparent that vergence training increases fu slonal ranges for both RDS and vectograms, whereas control therapy does not result in an increase in fusional ranges. Vectogram arid RDS performance are presented In prism diopters. After control therapy the patient received convergence therapy. (Reprinted from Am J Optom Physiol Opt 1980;57:21 1.)

found previously in normal patients’ Therefore, Cooper et al. concluded that the convergence insufficiency patient must be treated with a variety of vergence stimuli to obtain transfer. They also found a decrease in asthenopia on a scaled questionnaire as well as a flattening of the patient’s fixation disparity curve after vergence training. During placebo therapy no such improvement occurred. They concluded that orthoptics was effective in remedying convergence insufficiency with its accompanying symptoms (Figs. 4 and 5).

Cooper et al.8 have recently used a similar automated A-B crossover design to determine if monocular accommodative therapy results in improved accommodative abilities. Their patients demonstrated a statistically significant

improvement in accommodative facility, an in crease in accommodative amplitude, and a re duction in asthenopic symptoms. Within a short period of time a 55% improvement in amplitude and a reduction in symptoms occurred. Again, the experimental design controlled for effects that were coincidental or due to experimental bias or placebo (Fig. 6). Improvement in accommodative facility is important in the convergence insufficiency population because the majority of patients with convergence insufficiency have a secondary accommodative anomaly.7

Kertesz9 showed that automated training with microprocessor produced anaglyphic, large target, vergence stimuli that resulted in an im provement in vergence ranges and a reduction in asthenopia in patients who had a convergence insufficiency. All their convergence patients had previously failed to benefit from traditional orthoptics. Of the 29 convergence insufficiency patients treated, 23 increased their fusional ranges with a concurrent alleviation of symptoms. Treatment included slowly separating 57 dichoptic targets and RDS which were presented in both convergent and divergent directions. Therapy required 5 to 15 sessions. Kertesz and Kertesz concluded that computer-generated, large stimuli are more effective in remedying convergence insufficiency than traditional orthoptic techniques. However, Kertesz and Kertesz did not control for stimulus parameters (large vs. small, stereo vs. flat), motivation, skill of the therapist, and/or speed of vergence. Thus, their success may have been due to extraneous factors. Somers et al.10 used microprocessor-generated stimuli to treat patients with binocular anomalies. They reported that patients treated with computer-produced vergence stimuli showed more rapid and complete improvement than traditional techniques. Griffin reported that microprocessor-produced anaglyphs re sulted in an improvement in convergence ranges similar to traditional methods and a greater improvement in divergence ranges than traditional methods.

The above studies have shown the clinical effectiveness of automated microprocessor-generated anaglyphs in increasing fusional ranges. Methods which incorporated operant conditioning seemed to be the most effective. However, many of these research studies utilized sophis ticated computer equipment and techniques not yet available to the clinician. Cooper and Citron12 demonstrated that a personal computer (PC) could produce sophisticated anaglyphs, which could be moved to create a variety of vergence stimuli.

With the advent of small, powerful PC’s that can produce sophisticated anaglyphic targets, commercially available computerized vision