Limits...
Orientation tuning of a two-stimulus afterimage: Implications for theories of filling-in.

Van Horn DR, Francis G - Adv Cogn Psychol (2008)

Bottom Line: From the analysis, we show that the model must predict a rapid drop in afterimage occurrence as the gratings deviate from orthogonal.We then report on 2 experiments that test the properties of the model and find that the experimental data are strikingly different from the model predictions.From these discrepancies we identify the key deficits of the current version of the model.

View Article: PubMed Central - PubMed

Affiliation: Psychological Sciences, Purdue University,West Lafayette, IN, USA.

ABSTRACT
Sequential viewing of 2 orthogonally related gratings produces an afterimage related to the firstgrating (Vidyasagar, Buzas, Kisyarday, & Eysel, 1999; Francis & Rothmayer, 2003). We investigated how the appearance of the afterimage depended on the relative orientations of the 2 stimulus gratings. We firstanalyzethetheoretical explanation of the appearance of the afterimage that was proposed by Francis and Rothameyer (2003). From the analysis, we show that the model must predict a rapid drop in afterimage occurrence as the gratings deviate from orthogonal. We also show that the model predicts that the shape of the afterimage should always be orthogonal to the second grating. We then report on 2 experiments that test the properties of the model and find that the experimental data are strikingly different from the model predictions. From these discrepancies we identify the key deficits of the current version of the model.

No MeSH data available.


The spatial structure of the model-produced afterimage for various S1 and							S2 orientation differences. The model always predicts that the							afterimage is orthogonal to S2, regardless of the orientation of S1.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2864993&req=5

Figure 6: The spatial structure of the model-produced afterimage for various S1 and S2 orientation differences. The model always predicts that the afterimage is orthogonal to S2, regardless of the orientation of S1.

Mentions: Figure 6 shows the spatial structure of the predicted afterimage percept for different combinations of S1 and S2 relative orientations. When S1 is orthogonal to S2, the afterimage percept consists of alternating vertical bars. When S1 is rotated only 5° clockwise from orthogonal, the afterimage percept is more muddled. This is because of the spatial interactions of the orientation after-responses from S2 and the color after-responses from S1. The orientation after-responses from S2 are all vertical and constrain whatever color after-responses exist to only flow up and down, not left or right. Because of the orientation of S1 and the thickness of the bars, there are differences in the proportion of black and white after-responses in different vertical columns. When there is more white than black in a column, the afterimage percept at that column will be light gray. Similarly, other columns will have an afterimage percept of dark gray, when there are more black than white after-responses. For further rotations (and smaller angle differences between S1 and S2) the number of black and white after-responses in a column tend to balance out with only small differences being present between different columns.


Orientation tuning of a two-stimulus afterimage: Implications for theories of filling-in.

Van Horn DR, Francis G - Adv Cogn Psychol (2008)

The spatial structure of the model-produced afterimage for various S1 and							S2 orientation differences. The model always predicts that the							afterimage is orthogonal to S2, regardless of the orientation of S1.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2864993&req=5

Figure 6: The spatial structure of the model-produced afterimage for various S1 and S2 orientation differences. The model always predicts that the afterimage is orthogonal to S2, regardless of the orientation of S1.
Mentions: Figure 6 shows the spatial structure of the predicted afterimage percept for different combinations of S1 and S2 relative orientations. When S1 is orthogonal to S2, the afterimage percept consists of alternating vertical bars. When S1 is rotated only 5° clockwise from orthogonal, the afterimage percept is more muddled. This is because of the spatial interactions of the orientation after-responses from S2 and the color after-responses from S1. The orientation after-responses from S2 are all vertical and constrain whatever color after-responses exist to only flow up and down, not left or right. Because of the orientation of S1 and the thickness of the bars, there are differences in the proportion of black and white after-responses in different vertical columns. When there is more white than black in a column, the afterimage percept at that column will be light gray. Similarly, other columns will have an afterimage percept of dark gray, when there are more black than white after-responses. For further rotations (and smaller angle differences between S1 and S2) the number of black and white after-responses in a column tend to balance out with only small differences being present between different columns.

Bottom Line: From the analysis, we show that the model must predict a rapid drop in afterimage occurrence as the gratings deviate from orthogonal.We then report on 2 experiments that test the properties of the model and find that the experimental data are strikingly different from the model predictions.From these discrepancies we identify the key deficits of the current version of the model.

View Article: PubMed Central - PubMed

Affiliation: Psychological Sciences, Purdue University,West Lafayette, IN, USA.

ABSTRACT
Sequential viewing of 2 orthogonally related gratings produces an afterimage related to the firstgrating (Vidyasagar, Buzas, Kisyarday, & Eysel, 1999; Francis & Rothmayer, 2003). We investigated how the appearance of the afterimage depended on the relative orientations of the 2 stimulus gratings. We firstanalyzethetheoretical explanation of the appearance of the afterimage that was proposed by Francis and Rothameyer (2003). From the analysis, we show that the model must predict a rapid drop in afterimage occurrence as the gratings deviate from orthogonal. We also show that the model predicts that the shape of the afterimage should always be orthogonal to the second grating. We then report on 2 experiments that test the properties of the model and find that the experimental data are strikingly different from the model predictions. From these discrepancies we identify the key deficits of the current version of the model.

No MeSH data available.