Orientation discrimination as a function of orientation and spatial frequency

Abstract

Orien~tion discrimination was measured with a yes-no signal detection procedure at two reference orientations, 90° and 45°. For the first of two experiments, the stimuli were disk-shaped gratings, 2.5 ° in diameter. Their spatial frequencies ranged from 1.8 cpd to 17.6 cpd. For the second experiment, the stimuli were either gratings or bars consisting of the center cycle of the corresponding grating. As expected, the results showed a highly significant reference orientation factor, with smaller just noticeable differences at 90° than at 45° (the oblique effect). However, orientation discrimination was independent of spatial frequency and bar width. Implications of the findings are discussed in relation to a holographic model that relates the oblique effect to meridional differences in the number of orientation-tuned cortical cells.In a variety of sensory, perceptual, and cognitive visual tasks, performance is superior when stimuli are oriented along the horizontal or vertical meridian rather than along the oblique meridians. One manifestation of this oblique effect was studied in the present experiments, in which we measured just noticeable differences (JNDs) for slant. Our research was guided by a model that assumes a proportionality between orientation sensitivity for a given stimulus and the number of orientation-tuned cortical cells that respond to it. According to this theory, first considered by Bouma and Andriessen (1968), meridional differences in orientation sensitivity result from meridional anisotropies in the numbers of such cells, with fewer neurons tuned to the oblique meridians than to the main meridians. (See Figure 9b and the related discussion in Bouma & Andriessen, 1968, for a description of their model.!)Since the number-of-neurons theory was proposed, a substantial body of physiological evidence has accumulated that supports the existence of the postulated neural anisotropy (De Valois, Yond, & Hepler, 1982; Mansfield, 1974; Mansfield & Ronner, 1978; Orban & Kennedy, 1981;Pettigrew, Nikara, & Bishop, 1968; but see Finlay, Schiller, &Volman, 1976, and Poggio, Doty, &Talbot, 1977). Moreover, the theory itself was recentlyexpanded and modified to incorporate the results of a study of orientation sensitivity in short-term memory (Thoms, 1981). In that study, memory for orientation decayed more rapidly at 90° , where sensitivity was greater, than at 45 ° .Thoms related this memorial manifestation of the oblique effect to the theory about the neural basis of the corresponding sensory effect by making the following as- Matin and Thomas (1984) further elaborated the preceding suggestions about sensitivity and amount of stored information in terms of the number of responding neurons by relating them to ideas about holographic methods of information storage. The linkage was established through the following property of a holographic image: storage of the object as a whole is retained while the total amount of information (detail and resolving power) is decreased if pieces of the hologram are replace...

Human perceptual learning in identifying the oblique orientation: retinotopy, orientation specificity and monocularity.

“…We observed that performance in identification of the oblique orientation could be improved dramatically with practice. Initial thresholds were similar to those reported by others, and higher than reported for vertical or horizontal references (Vogels & Orban, 1985;Matin, Rubsamen & Vannata, 1987), in line with the 'oblique effect' reported for line orientation discrimination and grating contrast sensitivity (Andrews, 1967;Appelle, 1972;Orban et al 1984). Our finding of a dramatic decrease in threshold for the oblique orientation with practice may not apply to the vertical and horizontal orientation.…”

“…Initial thresholds were similar to those reported by others, and higher than reported for vertical or horizontal references (Vogels & Orban, 1985;Matin, Rubsamen & Vannata, 1987), in line with the 'oblique effect' reported for line orientation discrimination and grating contrast sensitivity (Andrews, 1967;Appelle, 1972;Orban et al 1984 Orban (1985) that the effects of practice upon orientation discrimination originate at the sensory level and not in the decision process, and cannot be attributed to a change in memory noise, attention or accommodation. Our study elaborated upon these findings to investigate the localization of this form of human perceptual learning in the visual pathway.…”