Limits...
Brightness/darkness induction and the genesis of a contour.

Roncato S - Front Hum Neurosci (2014)

Bottom Line: Particular configurations have been introduced that allow us to observe the induction effects of one contour taken in isolation.This effect weakens or s when the contour of the invariant CP separates surfaces filled with different gray shades.These conflicting results stimulate a deeper exploration of the induction phenomena and their role in the computation of brightness contrast.

View Article: PubMed Central - PubMed

Affiliation: Dipartimento Psicologia Generale, UniversitĂ  Padova Padova, Italy.

ABSTRACT
Visual contours often result from the integration or interpolation of fragmented edges. The strength of the completion increases when the edges share the same contrast polarity (CP). Here we demonstrate that the appearance in the perceptual field of this integrated unit, or contour of invariant CP, is concomitant with a vivid brightness alteration of the surfaces at its opposite sides. To observe this effect requires some stratagems because the formation in the visual field of a contour of invariant CP normally engenders the formation of a second contour and then the rise of two streams of induction signals that interfere in different ways. Particular configurations have been introduced that allow us to observe the induction effects of one contour taken in isolation. I documented these effects by phenomenological observations and psychophysical measurement of the brightness alteration in relation to luminance contrast. When the edges of the same CP complete to form a contour, the background of homogeneous luminance appears to dim at one side and to brighten at the opposite side (in accord with the CP). The strength of the phenomenon is proportional to the local luminance contrast. This effect weakens or s when the contour of the invariant CP separates surfaces filled with different gray shades. These conflicting results stimulate a deeper exploration of the induction phenomena and their role in the computation of brightness contrast. An alternative perspective is offered to account for some brightness illusions and their relation to the phenomenal transparency. The main assumption asserts that, when in the same region induction signals of opposite CP overlap, the filling-in is blocked unless the image is stratified into different layers, one for each signal of the same polarity. Phenomenological observations document this "solution" by the visual system.

No MeSH data available.


Related in: MedlinePlus

(A) The iso-contrast version of the “snake illusion”: the two axes (straight and wavy) have opposite CP as illustrated in the outline patterns on the right. The background between the columns of inducers is traversed by induction signals of the opposite sign. Three elongated ellipsoids drawn in the central portion of the columns serve as targets. They are identical in gray shade but the central one looks darker. (B) Same as in (A) but with smaller inducers inside the semi-ellipses that generate vertical axes of the opposite sign with respect to (A) as illustrated in the outline on the right. The background is traversed by induction signals of the same sign. In this case the interlaced axes sum their effects. Note the target brightness; the differences are negligible with respect to that observed in (A).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4202701&req=5

Figure 6: (A) The iso-contrast version of the “snake illusion”: the two axes (straight and wavy) have opposite CP as illustrated in the outline patterns on the right. The background between the columns of inducers is traversed by induction signals of the opposite sign. Three elongated ellipsoids drawn in the central portion of the columns serve as targets. They are identical in gray shade but the central one looks darker. (B) Same as in (A) but with smaller inducers inside the semi-ellipses that generate vertical axes of the opposite sign with respect to (A) as illustrated in the outline on the right. The background is traversed by induction signals of the same sign. In this case the interlaced axes sum their effects. Note the target brightness; the differences are negligible with respect to that observed in (A).

Mentions: In Figure 6A we have reproduced a variation of the “snake-illusion” in which the background is a homogeneous gray. The axes generate inductions of opposite sign on the background as illustrated in Figure 3. Figure 6B has been obtained by inverting the CP of the vertical axis so as to give two interlacing edges of the same CP orientation. With this combination of axes, the background between two columns of inducers is traversed by two streams of induction signals of the same sign (see outline pattern in Figure 6). We expect no layering to occur. If we compare the brightness alteration of the three targets (the ellipses have an identical gray shade) we observe a marked difference in the upper configuration but a negligible effect in the lower configuration. Note that in this latter case the background is viewed as split into vertical light and dim bands, in agreement with predictions that can be made considering the CP of the vertical axes. It is interesting to note that the brightness alterations are more vivid in the layered image (Figure 6A).


Brightness/darkness induction and the genesis of a contour.

Roncato S - Front Hum Neurosci (2014)

(A) The iso-contrast version of the “snake illusion”: the two axes (straight and wavy) have opposite CP as illustrated in the outline patterns on the right. The background between the columns of inducers is traversed by induction signals of the opposite sign. Three elongated ellipsoids drawn in the central portion of the columns serve as targets. They are identical in gray shade but the central one looks darker. (B) Same as in (A) but with smaller inducers inside the semi-ellipses that generate vertical axes of the opposite sign with respect to (A) as illustrated in the outline on the right. The background is traversed by induction signals of the same sign. In this case the interlaced axes sum their effects. Note the target brightness; the differences are negligible with respect to that observed in (A).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4202701&req=5

Figure 6: (A) The iso-contrast version of the “snake illusion”: the two axes (straight and wavy) have opposite CP as illustrated in the outline patterns on the right. The background between the columns of inducers is traversed by induction signals of the opposite sign. Three elongated ellipsoids drawn in the central portion of the columns serve as targets. They are identical in gray shade but the central one looks darker. (B) Same as in (A) but with smaller inducers inside the semi-ellipses that generate vertical axes of the opposite sign with respect to (A) as illustrated in the outline on the right. The background is traversed by induction signals of the same sign. In this case the interlaced axes sum their effects. Note the target brightness; the differences are negligible with respect to that observed in (A).
Mentions: In Figure 6A we have reproduced a variation of the “snake-illusion” in which the background is a homogeneous gray. The axes generate inductions of opposite sign on the background as illustrated in Figure 3. Figure 6B has been obtained by inverting the CP of the vertical axis so as to give two interlacing edges of the same CP orientation. With this combination of axes, the background between two columns of inducers is traversed by two streams of induction signals of the same sign (see outline pattern in Figure 6). We expect no layering to occur. If we compare the brightness alteration of the three targets (the ellipses have an identical gray shade) we observe a marked difference in the upper configuration but a negligible effect in the lower configuration. Note that in this latter case the background is viewed as split into vertical light and dim bands, in agreement with predictions that can be made considering the CP of the vertical axes. It is interesting to note that the brightness alterations are more vivid in the layered image (Figure 6A).

Bottom Line: Particular configurations have been introduced that allow us to observe the induction effects of one contour taken in isolation.This effect weakens or s when the contour of the invariant CP separates surfaces filled with different gray shades.These conflicting results stimulate a deeper exploration of the induction phenomena and their role in the computation of brightness contrast.

View Article: PubMed Central - PubMed

Affiliation: Dipartimento Psicologia Generale, UniversitĂ  Padova Padova, Italy.

ABSTRACT
Visual contours often result from the integration or interpolation of fragmented edges. The strength of the completion increases when the edges share the same contrast polarity (CP). Here we demonstrate that the appearance in the perceptual field of this integrated unit, or contour of invariant CP, is concomitant with a vivid brightness alteration of the surfaces at its opposite sides. To observe this effect requires some stratagems because the formation in the visual field of a contour of invariant CP normally engenders the formation of a second contour and then the rise of two streams of induction signals that interfere in different ways. Particular configurations have been introduced that allow us to observe the induction effects of one contour taken in isolation. I documented these effects by phenomenological observations and psychophysical measurement of the brightness alteration in relation to luminance contrast. When the edges of the same CP complete to form a contour, the background of homogeneous luminance appears to dim at one side and to brighten at the opposite side (in accord with the CP). The strength of the phenomenon is proportional to the local luminance contrast. This effect weakens or s when the contour of the invariant CP separates surfaces filled with different gray shades. These conflicting results stimulate a deeper exploration of the induction phenomena and their role in the computation of brightness contrast. An alternative perspective is offered to account for some brightness illusions and their relation to the phenomenal transparency. The main assumption asserts that, when in the same region induction signals of opposite CP overlap, the filling-in is blocked unless the image is stratified into different layers, one for each signal of the same polarity. Phenomenological observations document this "solution" by the visual system.

No MeSH data available.


Related in: MedlinePlus