Alexa Ruppertsberg Sophie Wuerger Global Motion Processing: The Red - Green Mechanism The MacKay Institute of Communication and Neuroscience School of Life Sciences Keele University United Kingdom Dep. of Psychology University of Liverpool United Kingdom www.keele.ac.uk/depts/co/HMP Marco Bertamini •Derrington, A.M., Krauskopf, J., and Lennie, P. 1984. Chromatic Mechanisms in lateral geniculate nucleus of macaque. Journal of Phsyiology , 357, 241-265. •Edwards, M. and Badcock, D. R. 1996. Global-motion perception: Interaction of chromatic and luminance signals. Vision Research, 36 (16), 2423-2431. • Snowden, R. J. and Edmunds, R. 1999 . Colour and polarity contributions to global motion perception. Vision Research, 39, 1813- 1822. • Stromeyer III, C. F., Kronauer, R. E., Ryu, A., Chaparro, A., and Eskew Jr., R. T. 1995.Contribution of human long-wave and middle-wave cones to motion. Journal of Physiology, 485 (1), 221-243. • Wuerger, S. M. and Landy, M. S. 1993. Role of chromatic and luminance contrast in inferring structure from motion. JOSA A, 10 (6), 1363-1372. REFERENCES INTRODUCTION STIMULI & METHODS EXPERIMENT 1: BASELINE Colour Contrast thresholds EXPERIMENT 4: MIXING RED AND GREEN EXPERIMENT 2: ADDING CHROMATIC NOISE EXPERIMENT 2: RESULTS CONCLUSION The interaction of colour and motion cues for global motion integration across space has only recently been studied. To establish the tolerance limits of global motion we first determine global motion detection thresholds (81%) as a function of the chromatic contrast in the isoluminant cone-opponent colour-space (S-(M- - L) space). We further test whether global motion in the isoluminant plane is mediated by more than one chromatic mechanism. S S M -L M -L Lum Lum WP WP 0° 90° 180 ° 270 ° -90° +90° 270 270 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 AIRF 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 CXVF 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 DMSM 0.5 1 1.5 30 210 60 240 90 120 300 150 330 180 0 ELSF 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 TJHM 0.5 1 1.5 30 210 60 240 90 120 300 150 330 180 0 HYMM 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 MXTM 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 SLTF 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 DXMF EXDF 0.5 1 1.5 30 210 60 240 90 270 120 300 150 330 180 0 2 out of 10 participants seem to use S-cone input. 8 out of 10 participants seem not to be able to use S-cone input to extract global motion: a pure Red-Green mechanism! If for 8 out of 10 participants only the Red-Green mechanism mediates global motion processing, then any colour giving the same stimulation to the mechanism should yield the same performance. M-L Projection onto M-L-axis is the same for every colour from the distribution. • Same task as in Exp. 1. • Mean colour = mean (Projections Exp.1) • Coherence Level thresholds as a function of distribution width • Motion coherence and colour are uncorrelated ! S EXPERIMENT 3: BASELINE AT OBSERVER ISOLUMINANCE 0 1 2 3 0 1 2 3 0 1 2 3 Threshold Coherence Level [%] 0 1 2 3 0 10 20 30 40 50 60 70 DMSM 0 1 2 3 0 10 20 30 40 50 60 70 ELSF 0 10 20 30 40 50 60 70 HYMM 0 1 2 3 0 10 20 30 40 50 60 70 MXTM 0 1 2 3 0 10 20 30 40 50 60 70 SLTF AIRF CXVF DMSM ELSF HYMM MXTM SLTF TJHM -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0deg 180deg Regression coefficients and their 95% confidence intervals Sigma of Colour Noise Distribution Adding chromatic noise does not change the performance. 30 210 60 240 90 270 120 300 150 330 180 0 CXVF 30 210 60 240 90 270 120 300 150 330 180 0 AIRF 30 210 60 240 90 270 120 300 150 330 180 0 JCSF 30 210 60 240 90 270 120 300 150 330 180 0 BRIM 30 210 60 240 90 270 120 300 150 330 180 0 AIRF 30 210 60 240 90 270 120 300 150 330 180 0 BRIM 30 210 60 240 90 270 120 300 150 330 180 0 CXVF 30 210 60 240 90 270 120 300 150 330 180 0 JCSF To evaluate possible luminance artefacts we establish observers’ individual isoluminance by heterochromatic flicker and rerun the baseline experiment. We find: - all observers require more luminance for green -confirmation of the results from Experiment 1. Nominal Isoluminance Observer Isoluminance 0 10 20 30 40 50 60 70 AIRF 0 1 2 3 0 10 20 30 40 50 60 70 CXVF 0 10 20 30 40 50 60 70 TJHM To determine whether this is a single R/G mechanism or two different ones (a red and a green one), half the dots in the display are green and the other half red. Testing three models: a) Linear summation b) Probability summation c) Independent decision (MAX rule) θ θR/G40%? θ θ R20% θ θ G20% θ θ R40% θ θ G40% a) b) c) The S cone input in our global motion task is negligible for the majority of observers. The observed sensitivity to global motion is predicted by the projection onto the M - L-axis and is not due to luminance artefacts. There is evidence that global motion extraction in the observer’s isoluminant plane is governed by two mechanisms operating along the M - L-axis: a red one and a green one. θ R20 θ G20 θ R/G40 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 a) b) c) AIRF 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 θ R/G60 a) b) c) θ R30 θ G30 CXVF +7.3% +4.7% +4.0% +4.0% ⇒ ⇒ Global motion seems to be mediated by a Red-green mechanism. a), b), c) = predictions of the three different models One observer’s data are closer to the predictions of the linear summation model, the other observer’s data are more in line with a probability summation model. More data needs to be collected. :-) ϕ VSS02 # 430 DKL Cone-Opponent Space Whitepoint: s: 0.0211 m-l: 0.3516 Lum: 50 cd/m 2 random dot kinematograms (RDK): 300 coloured gaussian blobs 0.22 ° °, 1 ° °/s, 5.1° °x 4° °, 200cm viewing distance Observers had to distinguish between an interval with random motion and an interval with 40% of the blobs moving either left or right (2IFC).