Perceptualand Motor Skills, 1991, 73, 765-766. O Perceptual and Motor Skills 1991
DIFFERENTIAL MOTION THRESHOLDS TO SINUSOIDAL GRATINGS AT TWO ECCENTRICITIES MARK C. CHORLTON, DAVID C. FINLAY, MARK L. MANNING, W. ROSS FULHAM, AND JOHN BOULTON
The University of Newcastle Summary.-Differentia1 motion thresholds were measured at eccentricities of 9' and 16.6' using computer-generated sinusoidal gratings. Three spatial frequencies (0.51, 0.25, and 0.13 cydesldeg) were examined at reference velocities of 2, 4, 8, 16, 32, and 48 deglsec. Minimum differential velocity thresholds were between 20 and 30% of the reference velocities for the three spatial frequencies at both eccentricities Increasing eccentricity produced an increase in the velocity at which minimum velocity discrimination occurred. Temporal frequency tuning was between 4 and 8 Hz, regardless of eccentricity. Psvcho~hvsical examinations indicate that motion-detector units in the human visual cor. - . tex are selective for velocity, spatial frequency, temporal frequency, and orientation (1, 2, 3, 4, 6). Differential velocity discrimination has been described as a U-shaped function for velocities between 4 and 32 deglsec. (4). The minima of these functions occur at lower velocities for the fovea than for peripheral regions (2, 4, 5). The experiment used two independent groups of subjects to examine thresholds at 9O and 16.6O eccentricity. The f i s t group of ten subjects included eight men and two women, whilst the second group of seven subjects included four women and three men. The stimuli were a pair of computer-generated (Commodore Amiga AlOOOs and Commodore 1081 RGB monitors) vertical sinusoidal gratings which both moved horizontally to either the right or left at precisely controlled velocities. Three spatial frequencies (0.51, 0.25, and 0.13 cycles/deg) and six reference velocities (2, 4, 8, 16, 32, and 48 deglsec.) were examined. Direction of motion was varied randomlv from trial to trial. On any trial the subject was presented two gratings of the same spatial frequency, both moving either to the left or right. One stimulus was moving at one of the fixed reference velocities, whilst the other moved within a range of 60% of the reference velocity. The subject was required to indicate if a discriminable difference in velocities between the stimulus pair was perceivable. An eccentricity groups (2) x spatial frequency (3) x reference velocity (4) x comparison velocity (9) mixed-measures analysis of variance was carried out on the number of "different" responses for the f i s t four velocities (2, 4, 8, and 16 deglsec.). At both eccentricities thresholds for the lower velocities (4 and 8 deglsec.) were elevated in comparison to those at higher velocities (16, 32, and 48 deglsec.). The general findings of this experiment may be surnmarised. First, the differential velocitv discrimination thresholds found here were qualitativelv similar to those obtained by other investigators using different techniques. Second, the minima for the differential velocity discrimination thresholds were of the same magnitude regardless of eccentricity; however, minimum thresholds for the mote eccentric stimuli occurred at higher velocities. Third, quantitative differences exist in the magnitudes of the thresholds reported here (typically 20 to 30% of the reference velocity) and those reported elsewhere (typically 5 to 10%) ( 2 , 3, 4), although they are s i d a r to those reported by Thompson (6). Fourth, thresholds for
*
'This research was supported by NH&MRC Grant No. 35 280 107 to D. Finlay and J. Boulton. 'Correspondence concerning this pa er should be addressed to Mr. Mark C. Chorlton, Department of Psychology, University o?~ewcastle, NSW 2308, Austraha.
766
M. C. CHORLTON, ETAL.
the lower spatial frequency (0.13 cycles/deg) were considerably larger than those for the other two spatial frequencies (0.51 and 0.25 cycles/deg) at both eccentricities. This finding suggests that, for lower spatial frequencies, higher velocities are required for precise discrimination of velocity. Fifth, describing the data in terms of temporal frequency (see Fig. 1) resulted in a tuning of thresholds between 4 and 8 Hz for all spatial frequencies and both eccentricities.
m
0
1
'
2
a
3
'
4
L
5
'
0
'
7
'
8
Temporal Frequency lHzl
FIG.1. Differential motion thresholds for spatial frequency (cyclesldeg) as a function of temporal frequency (Hz) for both eccentricities (9 and 16.6'). rn = .51 c/deg (9 deg), = .25 c/deg (9 deg), A = .13 c/deg (9 deg), n = .51 c/deg (16.6 deg), o = .25 c/deg (16.6 deg), A = .13 c/deg (16.6 deg) Sixth, as eccentricity of the stimuli increased, there was a corresponding decrease in sensitivity to the slower velocities. This finding agrees with Orban, Van Calenbergh, De Bmyn, and Maes (4) who suggest that the number of neurons selective for slow velocities decreases with increasing eccentricity.
REFERENCES 1. ANDERSON, S. J., & BURR,D. C. (1987) Receptive field size of human motion detection units. Vision Research, 27, 621-635. 2. MCKEE, S. I?, & NAKAYAMA, K. (1984) The detection of motion in the peripheral visual field. Vision Research, 24, 25-32. 3. MCKEE, S. P, SILVERMAN, G. H., & NAKAYAMA, K. (1986) Precise velocity &scrimination despite random variations in temporal frequency and contrast. Vision Research, 26, 609619. 4. ORBAN, G. A , , VANCALENBERGI-I, F., DE BRUYN,B., & WS,H. (1985) Velocity discrirnination in central and ~ e r i ~ h e rvisual al field. Journal of the Optical Society of America, 2 , 1836-1847. 5. PANISH,S. C. (1988) Velocity discrimination at constant multiples of threshold contrast. Vision Research, 28, 193-201. 6. THOMPSON, l? G. (1983) Discrimination of moving gratings at and above detection threshold. Vision Research, 23, 1533-1538.
Accepted November 8, 1991
DIFFERENTIAL MOTION THRESHOLDS TO SINUSOIDAL GRATINGS AT TWO ECCENTRICITIES MARK C. CHORLTON, DAVID C. FINLAY, MARK L. MANNING, W. ROSS FULHAM, AND JOHN BOULTON
The University of Newcastle Summary.-Differentia1 motion thresholds were measured at eccentricities of 9' and 16.6' using computer-generated sinusoidal gratings. Three spatial frequencies (0.51, 0.25, and 0.13 cydesldeg) were examined at reference velocities of 2, 4, 8, 16, 32, and 48 deglsec. Minimum differential velocity thresholds were between 20 and 30% of the reference velocities for the three spatial frequencies at both eccentricities Increasing eccentricity produced an increase in the velocity at which minimum velocity discrimination occurred. Temporal frequency tuning was between 4 and 8 Hz, regardless of eccentricity. Psvcho~hvsical examinations indicate that motion-detector units in the human visual cor. - . tex are selective for velocity, spatial frequency, temporal frequency, and orientation (1, 2, 3, 4, 6). Differential velocity discrimination has been described as a U-shaped function for velocities between 4 and 32 deglsec. (4). The minima of these functions occur at lower velocities for the fovea than for peripheral regions (2, 4, 5). The experiment used two independent groups of subjects to examine thresholds at 9O and 16.6O eccentricity. The f i s t group of ten subjects included eight men and two women, whilst the second group of seven subjects included four women and three men. The stimuli were a pair of computer-generated (Commodore Amiga AlOOOs and Commodore 1081 RGB monitors) vertical sinusoidal gratings which both moved horizontally to either the right or left at precisely controlled velocities. Three spatial frequencies (0.51, 0.25, and 0.13 cycles/deg) and six reference velocities (2, 4, 8, 16, 32, and 48 deglsec.) were examined. Direction of motion was varied randomlv from trial to trial. On any trial the subject was presented two gratings of the same spatial frequency, both moving either to the left or right. One stimulus was moving at one of the fixed reference velocities, whilst the other moved within a range of 60% of the reference velocity. The subject was required to indicate if a discriminable difference in velocities between the stimulus pair was perceivable. An eccentricity groups (2) x spatial frequency (3) x reference velocity (4) x comparison velocity (9) mixed-measures analysis of variance was carried out on the number of "different" responses for the f i s t four velocities (2, 4, 8, and 16 deglsec.). At both eccentricities thresholds for the lower velocities (4 and 8 deglsec.) were elevated in comparison to those at higher velocities (16, 32, and 48 deglsec.). The general findings of this experiment may be surnmarised. First, the differential velocitv discrimination thresholds found here were qualitativelv similar to those obtained by other investigators using different techniques. Second, the minima for the differential velocity discrimination thresholds were of the same magnitude regardless of eccentricity; however, minimum thresholds for the mote eccentric stimuli occurred at higher velocities. Third, quantitative differences exist in the magnitudes of the thresholds reported here (typically 20 to 30% of the reference velocity) and those reported elsewhere (typically 5 to 10%) ( 2 , 3, 4), although they are s i d a r to those reported by Thompson (6). Fourth, thresholds for
*
'This research was supported by NH&MRC Grant No. 35 280 107 to D. Finlay and J. Boulton. 'Correspondence concerning this pa er should be addressed to Mr. Mark C. Chorlton, Department of Psychology, University o?~ewcastle, NSW 2308, Austraha.
766
M. C. CHORLTON, ETAL.
the lower spatial frequency (0.13 cycles/deg) were considerably larger than those for the other two spatial frequencies (0.51 and 0.25 cycles/deg) at both eccentricities. This finding suggests that, for lower spatial frequencies, higher velocities are required for precise discrimination of velocity. Fifth, describing the data in terms of temporal frequency (see Fig. 1) resulted in a tuning of thresholds between 4 and 8 Hz for all spatial frequencies and both eccentricities.
m
0
1
'
2
a
3
'
4
L
5
'
0
'
7
'
8
Temporal Frequency lHzl
FIG.1. Differential motion thresholds for spatial frequency (cyclesldeg) as a function of temporal frequency (Hz) for both eccentricities (9 and 16.6'). rn = .51 c/deg (9 deg), = .25 c/deg (9 deg), A = .13 c/deg (9 deg), n = .51 c/deg (16.6 deg), o = .25 c/deg (16.6 deg), A = .13 c/deg (16.6 deg) Sixth, as eccentricity of the stimuli increased, there was a corresponding decrease in sensitivity to the slower velocities. This finding agrees with Orban, Van Calenbergh, De Bmyn, and Maes (4) who suggest that the number of neurons selective for slow velocities decreases with increasing eccentricity.
REFERENCES 1. ANDERSON, S. J., & BURR,D. C. (1987) Receptive field size of human motion detection units. Vision Research, 27, 621-635. 2. MCKEE, S. I?, & NAKAYAMA, K. (1984) The detection of motion in the peripheral visual field. Vision Research, 24, 25-32. 3. MCKEE, S. P, SILVERMAN, G. H., & NAKAYAMA, K. (1986) Precise velocity &scrimination despite random variations in temporal frequency and contrast. Vision Research, 26, 609619. 4. ORBAN, G. A , , VANCALENBERGI-I, F., DE BRUYN,B., & WS,H. (1985) Velocity discrirnination in central and ~ e r i ~ h e rvisual al field. Journal of the Optical Society of America, 2 , 1836-1847. 5. PANISH,S. C. (1988) Velocity discrimination at constant multiples of threshold contrast. Vision Research, 28, 193-201. 6. THOMPSON, l? G. (1983) Discrimination of moving gratings at and above detection threshold. Vision Research, 23, 1533-1538.
Accepted November 8, 1991
