What effect does color palette have on the strength of the Hermann illusion?

Cesar Duran

The Hermann-grid illusion is a widely known illusion in which the intersection of lines appear to be a different shading than the lines themselves. One classical explanation for this phenomenon is the importance of on-center/off-center retinal ganglion cells. The receptive field is excited less at the intersections, and therefore leading to dark patches at the intersections. This process occurs in the periphery/semi-periphery. because the size of the receptive field is smaller at the fovea, making it less susceptible to the illusion. It should be noted that refutations to this explanation exist, as shown here.

For my variation of this illusion, I chose to look at whether color palette would play any role in the experience of the Hermann grid. I was curious to not only see if the illusion would occur for colors of the same luminance, but also whether there were certain color pairs that would cancel the illusion. My hypothesis was that the introduction of color into the illusion would not disrupt the experience, but the illusion would be less strong because the amount of rods in the periphery outnumber the cones in the periphery, which is what is sensitive to color. For two of the images, I used a purple/yellow pair (188,0,255)/(240,208,0) which contain roughly equal luminance. For the other two images, I used a green-orange pair (141,188,124)/(244,148,101), which I found online to be colorblind-unfriendly (w/ red-green opponent channel).

Purple Squares with yellow lines:


Yellow Squares with purple lines:


Green Squares with red-orange lines (colorblind-unfriendly):


Red-orange Squares with green lines (colorblind-unfriendly):


Hermann L (1870) Eine Erscheinung simultanen Contrastes. Pflügers Archiv für die gesamte Physiologie 3:13–15

Geier J, Bernáth L, Hudák M, Séra L (2008) Straightness as the main factor of the Hermann grid illusion. Perception 37:651–665

Comments


Cesar Duran

Thank you all for your feedback!

c) The responses were as expected. Even though the purple square has a similar luminance to the yellow lines, the illusion effect still occurs. This means that the illusion is consistent with both rods and cones. The disappearance of the illusion in the red-green case shows that the illusion is created before color-opponency occurs. The illusion linked in the description also shows that the receptive fields are not the sole explanation for the illusion.

d) The next steps for research would be into modification of the lines to see if there's a way to break the illusion in different circumstances. Perhaps using diagonal lines or acute angled intersections, the illusion will be perceived differently.

Griffin Leoanrd

a) The purple squares created a slightly weaker effect than the yellow squares, although both were strong. The illusion completely disappeared for me with the red and green squares! Extremely cool effect!
b) This illusion perfectly answers the question. The color of the the grid affects whether or not the Hermann illusion is perceived. Specifically, it seems that using opponent colors makes the illusion disappear!

Maddie C

Strength of hermann illusion for me:
Purple Squares > Yellow Squares > Green Squares ~ Red Squares
It's super intense with the Purple squares. It's almost completely absent in the red and green ones.

Maria Fernanda De La Torre

a) In the first grid (purple squares with yellow lines), the light-purple circles in the non-fixated corners appeared very strongly. This effect continued in the second inverted grid, although less strongly. The third and fourth grid (red-orange on green and inverted) made the illusion disappear for me. Although, in the third one I could still see the effect if I changed the location of my eye-fixation very rapidly.
b) This experiment answers the posed question, it shows that the magnitude and the existence of the Hermann illusion is affected by the colors of the grid. It is very interesting that a pair of opponent colors seem to make the illusion disappear, I would be curious to see the same grid with a different pair of opponent colors.

Ben Radovitzky

a) I noticed that when the squares were the "darker" color (purple in the first case, green in the second), I noticed that it was slightly easier to see the effect than in the inverse scenario. It still worked in all cases, but it was significantly more pronounced in the former. Also, the effect was overall stronger in the purple and yellow case than in the red and green one.
b) This demo perfectly answered the question asked in my opinion. The effect of a color palette on the Hermann Illusion (from my experience) is that it is less pronounced when more "similar" colors are used in terms of their "lightness".