Two experiments were conducted to investigate the probabilistic independency of the feature extraction for the multi-element visual display, In experiment I, subjects reported the number of dots which were displayed simultaneously for brief exposure time. In Experiment II, subjects answered whether all figures were the same or not, when they were shown simultaneously. The principal experimental variables were the number of items in the display and exposure duration. If the stimulus composed of N items was processed for 50%, and the features were extracted independently, each item must be processed for √^N<0.5>. From these experiments, it was concluded that features were extracted independently between items. Finally, we examined the probabilistic independency for the more complex display and several implications were discussed.
In a Sidman-avoidance procedure, Galizio (1979) investigated the nature of instruction-following behavior of human subjects as rule-governed behavior. The present study, in contrast to Galizio's experiment, used a schedule of positive reinforcement (three-component multiple DRL schedule of reinforcement) as experimental settings and investigated the effects of accurate or inaccurate instructions (Experiment I, II) and also examined whether instruction-following behavior comes under the external discriminative stimulus or not (Experiment III). In Experiment I and Experiment II, subjects followed instructions as far as the instructions were accurate, but didn't always followed the instructions when they were inaccurate. Especially, in the case following instructions resulted in the aversive consequences, subjects responded according to ongoing schedule of reinforcement not to instructions. Experiment III demonstrated that added novel stimulus controlled instruction-following behavior. Consistent with Galizio's findings, these results support Skinner's view of rule-governed behavior.
The development of visual acuities in infancy assessed by preferential looking (PL) is discussed with regard to methodological problems and clinical applications. Grating acuity develops up to about 12 cpd in the first year of life. The oblique effect was found at an age younger than 6 months. There was no evidence for meridional amblyopia in the first year of life. Monocular measures of acuity revealed the suppressive interaction between the eyes in strabismic amblyopia and monocular deprivation. The discrepancy between the results obtained by PL and those by VEP technique is also discussed. The development of stereopsis, in contrast to grating acuity, is marked by a sudden onset at an average age of 4 months followed by a rapid increase in stereoacuity. Another hyperacuity, vernier acuity, was also measured recently by PL using a vernier-motion-sound display. It was found to be superior to grating acuity by 4 months of age, about the same as the onset age of stereopsis. Thus, a common cortical basis for the development of hyperacuities was suggested.
Two experiments were performed to compare the pursuit and the compensatory tracking behavior with either a proportional or an integral control system. The first experiment investigated the effects of the target characteristics on each behavior. It was found that while with the proportional control system the superiority of the pursuit over the compensatory tracking behavior became greater with increasing target frequencies, with the integral control system the two types of tracking behavior were not significantly different in scores for any target conditions. The second experiment investigated the learning characteristics and the transfer effects of each behavior. It was observed that while with the proportional control system the subjects who transferred from the compensatory to the pursuit tracking behavior had lower error scores on transfer trials than the subjects transferred in the reverse order, with the integral control system error scores on transfer trials increased when the subjects transferred to the behavior differring from the learning behavior. These results were discussed in terms of the difference of the optimal response patterns of these control systems.