The Strength of the Ponzo Illusion in the Central Versus Peripheral Vision

Russell Anderson, anderruss@hotmail.com, Alma College

Jason Garvin, craz53night@aol.com, Alma College

      Much work has been done, to understand the effects of various illusions at work within the deception of the Ponzo illusion (Prinzmetal and Shimamure, 2001; Pressey and Epp, 1992; Jordan and Randall, 1987; Jaeger, Treiber, and Pollack, 1980; Libet, Pollack, and Malatesta, 1980; Thompson and Schiffman, 1979) and much been done assessing the differences in the vision of the Fovea, which is primarily cones, and peripheral vision, which is attributed mostly to rods (Lit, Young, and Shaffer, 1971; Lit and Hamm, 1966; Michaels, 1953; Doesschate, 1946: Adrain,1944).

      Due to the less detailed visual reception of the rods, it is reasonable to hypothesize that the illusion may not have as strong of an effect in the right peripheral field. This would produce a more uniform result across both the control and the Ponzo illusion in the right fixation trials. Also, the left converging line in the Ponzo illusion would be ineffective due to the distance from the center of the visual field.

      Trials assessing Ponzo illusion stimuli without fixation and with center fixation should show stronger illusion, as well as greater variability (people are effected at different magnitudes by the illusion). No fixation data should be slightly more accurate than center fixation data, because participants can focus on the ends of the comparison lines and visually line them up (as if creating an imaginary box). Standard deviations should appear very closely between these two stimuli because their range of fixations do not differ greatly in degrees of visual angle.

      This experiment is intended to gain insight into how the Ponzo pattern is perceived by the rods and how that differs from central vision.

Paricipants
      Participants were 22 college students including 11 males and 11 females, aged 18-22 years, averaging 19.45 years of age.

Apparatus
      Figures were displayed on a 17 inch color monitor. The figures were constructed and run in the Eye Lines program (Beagley, 1991). Fixation points were black dots. Center fixation points were positioned at equal distance between the two lines at the point perpendicular to the midpoint of the lines. Right fixation points were placed an equal distance from each of the parallel lines, at a horizontal distance of 9.04mm past their endpoints. Using these measurements, and setting the participants face at a set distance of 680 mm from the center of the screen (using a string of the pre-marked distance), the right fixation points were 4.56 degrees of visual angle to the right of center. Participants were read a set of directions, explaining the usage of the fixation points to evaluate peripheral vision and how to perform their trial, but no insight of the illusion was given. No notion of time restraint or importance of pace was given.

Stimuli
      The general stimuli are shown in figure 1. The stimuli were presented in random order. Twelve adjustments were presented to each participant. Each set of twelve consisted of each stimuli shown twice; one adjustment began 9.04mm (10%) longer than the comparison line and the other adjustment began 9.04mm (10%) shorter. Adjusting, lengthed or shortened the line from both ends, at equal rates.

Figure 1. Stimuli are shown here at 25% of actual size. In presentation, the comparison line was 90.38mm long, set 6.24mm above the line of adjustment, and 92.11mm from the top of the illusion. In the alterations of Ponzo illusion the converging lines were 271.1mm long, set 26.11mm apart (at the top), and set to converge at angles 25 degrees from perpendicual. Fixation dots were constructed as 0.6mm long lines.

Procedure
      The observers were seated in front of the monitor and the experimental directions were read to them. After any questions were answered, their face was positioned. This distance was observed by the experimenters to remain constant throughout the experiment and be repositioned if changed. Experimenters also pointed out when the fixation points appeared in the trial, so that the participants were able to fixate as quickly as possible. Experimenters also made sure that the cursor was kept away from the trials (it was positioned at the edge of the screen), and that the participants kept their hands on or below the table. Participants proceeded in self-paced manner.

Figure 2. Mean illusion mangnitude of the Ponzo and Control stimuli as a function of no, central, and right fixation.

Figure 3. Standard deviation of mean illusion magnitude as a function of no, center, and right fixation on the Ponzo and control stimuli.

      A t-test showed the statistical significance (T(21)=1.415 P=0.047) of the difference between right and center fixations.

      All of this data supports the revised assimilation theory presented by Pressey and Wilson (1980). This theory proposes that there is a contrast field surrounding the central, attentive field. In the central attentive field the Ponzo illusion causes an underestimation of the adjustment line, as is normally expected. But in the (surrounding) contrast field, test stimulus (which is defined in the attentive field) is distorted in the opposite direction of the context of that stimulus. This theory has been supported by the findings of Pressey, Butchard, and Scrivner (1971), Fisher (1968), and Quina and Pollack (1972), as well.

      Center fixation is most influenced because it keeps the central and most influential part of the illusion in the attentive field and puts the part of the illusion in the contrast field that is almost neutral (as it contributes little to the over-all effect of the illusion).

      The no fixation stimulus has illusion magnitude between center and right fixation because it takes the greatest portion of the entire figure in the attentive field, as observers were free to scan the entire field. In perceiving all of the figure, participants judged on a much lower concentration of the illusion effects than in center fixation, and a much greater concentration than in right fixation. This is, proposing that information from the contrast field is secondary to the information picked up in the attentive field, while scanning the small area (a visual field with a radius of 17.35 degrees of visual angle).

      Right fixation is most resistant to the effects of the illusion due to the left converging line being completely outside of the attentive field and partially in the contrast field, serving as a counter illusion. The portion of the far left line which is perceived in the circular contrast field is the most influencial part of that line (conter-influencial in the contrast field), being closest to the comparison lines.

      The revised assimilation theory of Pressey and Wilson (1980), implies, through the standard deviation data, that the magnitude of the Ponzo illusion is more consistent across participants in the (left) contrast field.

      Much experimentation could be done in the direction of the assumptions made here; contrasting information gathered in the contrast field and attentive fields, and judging the nature of the fields outlined in the revised assimilation theory (Pressey and Wilson, 1980).

Beagley, W.K. (1991). Eye Lines [Computer Program]. Alma, MI: Alma College.

Fisher, G. H. (1968). Gradients of distortion seen in the context of the Ponzo illusion and other contours. Quarterly Journal of Experimental Psychology, v 20, p 212-217.

Jaeger, T., Treiber, F. A., & Pollack, R. H. (1980). Effect of lightness contrast on Ponzo illusions. Bulletin of the Psychonomic Society, v 15, p 1-4.

Jordan, K., & Randall, J. (1987). The effects of framing ratio and obligque length on Ponzo illusion magnitude. Perception & Psychophysics, v 41, p 435-439.

Jordan, K., & Schiano, D.J. (1986). Serial processing and the parellel lines illusion: Length contrast through relative spatial separtaion of contours. Perception & Psychophysics, v 40, p 384-390.

Pressey, A. W., Butchard, N., & Scrivner, L. (1971). Assimulation theory and the Ponzo illusion: Quantitative predictions. Canadian Journal of Psychology, v 25, p 486-497.

Pressey, A. W., & Epp, D. (1992). Spatial attention in Ponzo-like patterns. Perception & Psychophysics, v 52, p 211-221.

Pressey, A. W., & Wilson, A.E. (1980). Assimulation theory and the Baldwin illusion. Italian Journal of Psychology, v 7, p 65-73.

Prinzmetal, W., Shimamura, A. P., & Mikolinski, M. (2001). The Ponzo illusion and the perception of orientation. Perception & Psychophysics, v 63, p 99-114.

Quina, K., & Pollack, R. H. (1978). Effects of test line position and age on the magnitude of the Ponzo illusion. Perception & Psychophysics, v12, p 253-256.

Thompson, J. G., & Schiffman, H. R. (1979). The role of attention in the perception of the Ponzo illusion. Bulletin of the Psychonomic Society, v 13, p 336-338.

More Posters

Web Posters from other courses and projects

Alma College Psychology Department