PSYC111-Depth-Perception PDF

Preview

Summary

Depth Perception and CuesInfluencingthe Ponzo IllusionName: Olivia KingStudent ID Number: 2824719Demonstrator: Millie JohnstonLab Class Number: 113Abstract The aim of the present experiment was to examine whether the number of depth cues in an illusion pattern influences the magnitude of the Ponzo i...

Description

Depth Perception and Cues Influencing the Ponzo Illusion

Name: Olivia King Student ID Number: 2824719 Demonstrator: Millie Johnston Lab Class Number: 113

Abstract The aim of the present experiment was to examine whether the number of depth cues in an illusion pattern influences the magnitude of the Ponzo illusion. It was hypothesised that as the number and type of depth cues in the illusion pattern increases, participants’ estimates of stimulus length will increase (because the magnitude of the Ponzo illusion/perceived size of the top stimulus will increase). Twenty-five psychology students were presented with sixty trials of illusion patterns in which they had to alter the physical length of the bottom stimulus (either a line or a truck) to match the physical length of the top stimulus. The measured variable was the length of altered stimulus as a percentage of the reference stimulus for all stimuli types. The findings showed that participants assumed the length of the stimuli was longer when there were more depth cues. The results supported the hypothesis that as the number and type of depth cues in the illusion pattern increases, participants’ estimates of stimulus length will increase (because the magnitude of the Ponzo illusion/perceived size of the top stimulus will increase).

Depth Perception and Cues Influencing the Ponzo Illusion Our visual system is an expansive tool that allows us to grasp the shape, size and distance of objects. (Laboratory Manual: Psychology 111/112, 2016). When an image comes in, we have a two dimensional image on our retina which is translated to a three dimensional image through depth cues (Laboratory Manual: Psychology 111/112, 2016). This is referred as depth perception; the visual ability to view the world in a three dimensional state and understand the distance of an item (Colombo, 2016). Depth cues help us understand the distance objects are from us and from each other (Laboratory Manual: Psychology 111/112, 2016), and can be either binocular or monocular (Sternberg, 1998). Binocular cues require two forward facing eyes, which results in the two visual fields overlapping (Department of Psychology, 2016). This is found in animals that prey, as it assists with hunting. (Department of Psychology, 2016). There are two binocular depth cues; binocular convergence and binocular disparity (Sternberg, 1998). Binocular convergence is how our eyes rotate inward in order to see the same image of an object directly in front of us (Sternberg, 1998). The more our eyes converge, the perceived object is assumed to be closer (Sternberg, 1998). Binocular disparity is a cue based on the differences between the left and right eyes’ views of objects. The brain has to translate two different lots of information from each vision field, and the closer an object is to us, the more it appears to move. (Sternberg, 1998). Monocular cues depend on one eye (Gleitman, Gross & Reisberg, 2010) and can be represented in two dimensions, such as in a picture (Sternberg, 1998). Monocular cues are made up of five pictorial depth cues and a motion cue (Department of Psychology, 2016). Pictorial cues are patterns that can be shown on a flat surface to create a sense of a threedimensional object, and motion cues come from motion and the distance of an object (Gleitman et al., 2010).

Relative size is a monocular cue which perceives objects further away to be smaller, as the image is smaller on the retina if the object is a greater distance away (Sternberg, 1998). Texture gradient is how a change in texture provides information, with more texture meaning the object is closer to us (Department of Psychology, 2016). Interposition is one object being blocked by another, and the blocked object is perceived to be further away (Gleitman et al., 2010). Linear perspective helps us judge distance based on parallel lines, as those in the distance appear to converge (Sternberg, 1998). Relative height/location in the picture plane makes items closer to the horizon line appear further away (Department of Psychology, 2016). Motion parallax, the motion cue, makes nearby objects appear to move more rapidly than those further away (Department of Psychology, 2016). In order to perceive the depth of an object, objects first need to be recognized in terms of what they are (Gleitman et al., 2010). Perceptual constancy allows us to understand what an object is, as our view of an object remains the same despite changes in the retinal image of the object (Sternberg, 1998). Our brain compensates for the retinal image being different but perception of the object being the same (Laboratory Manual: Psychology 111/112, 2016). We can also perceive shape constancy, allowing us to view an object as remaining the same shape, even when the shape of its retinal image changes (Sternberg, 1998). In addition to perceptual and shape constancy, size constancy refers to accurately perceiving object size (Gleitman et al., 2010). Size constancy enlarges the perceived size of objects in the distance, and therefore we perceive the same size for objects that have the same actual size, but different retinal size (Department of Psychology, 2016). Size constancy influences the Ponzo illusion, a visual illusion triggered from the depth cue involved with converging lines (Sternberg, 1998). Size constancy plays a part as the Ponzo illusion is related to accurately perceiving object size (Sternberg, 1998). With the Ponzo illusion, the more distant object is perceived as larger than the closer object, as more

depth cues make the object appear to be bigger. (Department of Psychology, 2016). Linear perspective affects the illusion as this cue is involved with judging distance based on parallel lines as those in the distance converge (Sternberg, 1998). The Ponzo illusion occurs because in the real three-dimensional word, the further away object would be bigger (Sternberg, 1998). The aim for the current experiment was to examine whether the number of depth cues influences the magnitude of the Ponzo illusion. The aim was tested by presenting participants with sixty trials of illusion patterns, in which they had to alter the physical length of the bottom stimulus (either a line or a truck) until it matched the physical length of the top stimulus. Each pattern was presented ten times, and participants’ average estimate of stimuli length was recorded for each pattern by the computer . It was hypothesised that as the number and type of depth cues in the illusion pattern increases, participants’ estimates of stimulus length will increase (because the magnitude of the Ponzo illusion/perceived size of the top stimulus will increase).

Method Participants The participants were twenty-five students recruited through their PSYC111 laboratory programme at the University of Otago. There were ten males and fifteen females, with an age range of 17 to 28 years old. It was assumed that all participants had good hearing and eyesight. Materials The materials used for this experiment consisted of a computer which presented and recorded the data. The computer presented six illusion patterns; grey background, horizon,

road, road and houses, photographs and road and houses. There were two stimulus types, two lines and two trucks. Procedure The manipulated variable was the type and number of depth cues in the illusion patterns and it had six levels. These were grey background, horizon, road, road and houses, photographs and road and houses again. Each pattern was presented ten times, but the order of presentation was randomized. There were two types of stimuli, either two horizontal lines or two trucks. All patterns contained the line stimuli apart from the road and houses one which had two trucks as the stimuli. The design was within-subjects as each participant experienced all levels of the manipulated variable. Participants were placed with a computer each and completed a computer test. The test had six practice trials and sixty experimental trials. In each trial, participants were presented with an illusion pattern, and altered the physical length of the bottom stimulus (either a line or a truck) until it matched the physical length of the top stimulus. The average estimate of the length of the stimuli presented to each individual was recorded for each illusion by the computer.

Results The measured variable for the present experiment was the length of altered stimulus as a percentage of the reference stimulus for all stimuli types. A value greater than 100% meant that participants overestimated stimuli length. Each individual’s data was calculated then averaged across the group, producing a group mean. Table 1 Mean Percent of Altered Stimulus Length Compared to Reference Stimulus Background Cue

Stimuli

Mean Percentage (%)

Grey

Lines

103.9%

Horizon

Lines

107.0%

Road

Lines

116.5%

Road + Houses

Lines

121.5%

Photograph

Lines

126.6%

Road + Houses

Trucks

131.9%

As shown in Table 1, as the number and type of depth cues in the illusion pattern increased, participants’ estimation of stimuli length increased. The estimation goes up gradually from 103.9% to 131.9%. The figures between those two climb from 107.0%, 116.5%, 121.5% and 126.6%. A dependent t test was performed to determine if the mean results for 'grey' and 'roads + houses' were significantly different. There was a significant difference between grey conditions and road + houses conditions for the line stimulus, where participants estimated the line to be longer in the road + house condition, ( t(24) = 5.947, p < 0.05).

Discussion When participants were presented with an illusion pattern that had more depth cues, they overestimated the stimulus length. This explains the significant difference result found through the independent t test, as a higher amount of depth cues resulted in a difference in the mean percentage for altered stimulus length. The hypothesis that as the number and type of depth cues in the illusion pattern increases, participants’ estimates of stimulus length will increase (because the magnitude of the Ponzo illusion/perceived size of the top stimulus will increase) was supported.

The results from this experiment support the visual concepts of depth perception, constancy and the Ponzo illusion. Participants were able to identify what an object was due to perceptual constancy, which in turn helps to perceive the depth of an object (Gleitman et al., 2010). The concept of size constancy is demonstrated in the current experiment as objects in the distance were perceived to be larger with more depth cues (Sternberg, 1998). Size constancy triggers the Ponzo illusion, in which the further away object is perceived as larger than the closer object (Department of Psychology, 2016). This is heightened with more depth cues, and is shown in this experiment as participants overestimated stimulus length as more depth cues were involved. The results show that when there was a line and truck stimuli rather than just two horizontal lines, participants overestimated stimulus length at a higher level than all other overestimations. This difference could be due to the fact that having a truck and line stimuli meant there was vertical and horizontal stimuli that had to be matched, which proved more difficult than matching horizontal stimuli. This change was expected, as more depth cues was hypothesised to result in an overestimation of stimuli length by participants. A limitation of this experiment could be that participants were using their own methods in order to estimate stimulus length. For example, some may have been looking at the size of the truck wheels when matching truck stimuli length. This could be fixed by asking participants after they have completed the trials if they used a method, and if they felt it impacted their results. Another limitation of this experiment could be the fatigue that participants may have felt. As the trials were tested at night, participants were likely to have been tired and perhaps rushed through the trials due to fatigue and lack of focus. This could be fixed by holding the experiment at a more suitable time of day, or asking participants about their fatigue level.

An implication of this experiment is that more depth cues will make an object appear to be larger, which results in less accurate perception of the object size. This experiment suggests that our visual surroundings have an effect on how we view objects and the information we gather on their size, shape and distance. When people view an object in the distance whilst also presented with several depth cues, they are likely to believe it is larger than it actually is. An application from this is how we view upcoming obstacles with driving. We need to use our depth perception to accurately judge how far a car is from us. An example of how people use this is with the two-second object rule. When driving, in order to judge if you are a safe distance from the car in front of you, when that car passes an object you count to two slowly. If you pass the same object before you reach two, you are too close to that car. This works with the monocular cue of motion parallax, as nearby objects appear to move more rapidly than those further away (Department of Psychology, 2016). Therefore, judging distance based on a close and fast moving object ensures us we are a safe distance from the car ahead. A further application is how instructions are placed on the road for drivers. With markings such as ‘Keep Clear’, ‘Give Way’ or ‘Slow Down’, the first word will be put further away on the road. This is due to size constancy and the Ponzo illusion, as the further away object (in this case, the word) will appear larger even though it is the same size. This means that drivers will read the instruction correctly as they will be drawn to the larger word first. Further research in this area might include more vertical cues than horizontal cues in order to see if this effects the estimation of stimulus length. This could be done with horizontal cues and both horizontal and vertical cues, in order to see the differences between all three groups. Given that the only stimuli which contained both horizontal and vertical cues

provided the highest overestimation (truck stimuli), vertical cues may have an effect on future results. In summary, the present experiment demonstrated the Ponzo illusion and various visual concepts. The results show that when surrounded with more depth cues, people are likely to overestimate the size of objects in the distance. This suggests that in an environment with more cues, our perception of faraway objects’ size is likely to be less accurate. Our visual system is constantly trying to gather information and account for differing environments and situations when viewing objects, but can be easily manipulated.

References Colombo, M. (2016, April 8) Lecture 14: Pattern Perception. Lecture presented in Psychology 111, University of Otago. Department of Psychology (2016, April 13) Depth Perception. PowerPoint slide presented in Psychology 111, University of Otago. Gleitman, H., Gross, J., & Reisberg, D. (2010). Psychology (8th ed.). New York: W. W. Norton & Company, Inc. Laboratory Manual: Psychology 111/112 (2016). Dunedin: Department of Psychology, University of Otago. Sternberg, R. J. (1998). In search of the human mind (2nd ed.). Fort Worth, TX, Harcourt Brace & Company....