PIGEON DISCRIMINATION OF PAINTINGS 1 Pigeon Discrimination of Paintings by Image Sharpness ANONYMOUS Psychology and 20th Century Literature August 8th, 2016
PIGEON DISCRIMINATION OF PAINTINGS 2 Pigeon Discrimination of Paintings by Image Sharpness Pigeon visual perception and stimulus recognition is well documented in laboratoratory research. Bhatt et al. (1988) established in many experiments the pigeon's ability to discriminate and differentiate between many objects not found in the pigeon's natural habitat. Wasserman et al. (1988) continued that research and further developed our conceptual understanding of not only the pigeon's ability to differentiate but the pigeons ability to categorize the stimuli it sees as well. Watanabe, Sakamoto, and Wakita (1995) conducted an experiment with pigeons that takes the concept of classification and categorization of visual stimuli much further by testing pigeon discrimination skills with paintings by different artists. In the experimenters' first study, the pigeons successfully discriminated between various slides, suggesting pigeons behavior can be controlled by visual stimuli and categorization. In one test the experimenters altered the sharpness of the Picasso and Monet images to check for that as a possible aid for the pigeons discrimination, but the exact degree of blurredness used was not noted. Watanabe et al. (1995) concluded that sharp edges did not uniquely control for the discrimination (196). However, I believe there to be a degree of alteration that could inhibit the pigeons ability to discriminate Monet from Picasso. This study aims to look at the exact degree of blurred edges required before the pigeons can no longer successfully discriminate. By completing this experiment, we will now be able to at what point pigeons fail to tell the difference between distinctively different paintings and we could speculate which stimuli pigeons perceive and note the differences of. Future experiments may also include human subjects and by finding the similarities and
PIGEON DISCRIMINATION OF PAINTINGS 3 differences between the perceptions of the pigeon and the human we can determine what stimuli are most effective and important to animal and human observation. METHOD Subjects and Apparatus Eight experimentally naïve pigeons, housed in stainless steel cages, and kept at 80% of their free-feeding weight will be used in this study. The experimental chambers will be two operant chambers that will be identical to one another will be used and constructed of folded sheet metal, with a clear plastic top and sides to allow for visibility, the chambers will measure 40 cm wide by 40 cm long by 40 cm high. Each chamber will have a frosted glass screen attached to the chamber above the key, which will be 8 cm off the floor of the chamber and in the direst center. The frosted screen will be used for the photographs of Monet and Picasso. The stimuli will be projected on the screen using a video projector (Phillips LCP5000). Gimp 2.8 software will be used to blur the images using a gaussain blur of varying percentages. The breakdown of percentage corrected to rate of blurriness is as follows: Percentage Rate of Blur Across Image 0% Base image, no blur 20% Image blurred at 50.0 x 1 rate
PIGEON DISCRIMINATION OF PAINTINGS 4 40% Image blurred at 100.0 x 1 rate 60% Image blurred at 150 x 1rate 80% Image blurred at 200 x 1 rate 100% Image blurred at 250 x 1 rate Procedure Fig. 1 represents the rate of blur applied across the entire image (as according to the GIMP 2.8 photoshop software) and the author represented percentage associated with rate of blur. Pigeons will first be trained to discriminate photographs of paintings by Picasso and Monet. During the experiment two sets of 10 different paintings from each artist will be used. The paintings (reference Watanabe et al. 1995 for list) will be the same as used in the Watanabe et al. (1995) study. The pigeons will be trained to peck an illuminated key first without the paintings. Then the pigeons were divided into two groups: a Monet group (A) and a Picasso group (B). In the Monet S+ group, responses to Monet paintings were reinforced by 5 seconds access to food and responses to Picasso were extinguished. In the Picasso S+ group, Picasso responses will be reinforced with 5 seconds of food access and Monet responses will be extinguished. Two birds per group will receive training with set A and two birds of the other group will receive training
PIGEON DISCRIMINATION OF PAINTINGS 5 with set B. This training consists of 20 randomly ordered paintings once, for 20 seconds, followed by a 5 second blackout period. The discrimination training will persist until all pigeons are at 90% successful discrimination ratio. Discrimination training, of at least two sessions, will occur between sets. These sets will be repeated, but with each set comprised of paintings of increasingly blurred sharpness. As noted by Watanabe et al. (1995), Picasso paintings have sharp edges while Monet s do not, therefore the increasingly blurred edges of the stimuli presented will eliminate this by controlling for the role of contour. Anticipated Results Fig. 2: Picasso (top row) and Monet (bottom row) levels of blurring with the first image being the base image, the middle at 60% blurred, and the last at 100% blurred. This level of blurredness was the same across the 10 images used. NOTE: there were 6 levels of blurredness for each photo, however, only the base, middle, and final are depicted above. RESULTS AND DISCUSSION It is anticipated that the pigeons behavior would be brought under stimulus control with the number of sessions required to reach criterion would vary by pigeon. In regards to the Monet and Picasso photographs, while the pigeons would likely be able to discriminate the first few
PIGEON DISCRIMINATION OF PAINTINGS 6 levels of blurredness of the paintings, similar to the results of the Watanabe et al. (1995) study, I suspect the percent correct would drop as the level of blurredness increased. By the 100% blurred photos, I anticipate the pigeons to not be able to accurately, at a level above chance, discriminate Picasso from Monet. This would suggest that sharpness could be attributed to the pigeons visual discrimination. The following is a chart that lists my expected results: The bar graphs for each pigeon clearly lower accuracy as the blurriness of individual paintings increase. While these results may occur it would be interesting to note variations that might occur. For example, if pigeons were to immediately be less capable of differentiating between painting as blurriness increases this would indicate that pigeons are entirely dependent on crisp and definite visual stimuli for accurate classification. If pigeons are able to accurately categorize images regardless of how blurry they become, this indicates the pigeons are reliant on other stimuli such as color to differentiate between stimuli. Future experiments might effectively compare human and pigeon results, which could greatly refine and add to our knowledge of stimulus perception and categorization. If we better
PIGEON DISCRIMINATION OF PAINTINGS 7 understand these cognitive processes we might be able to train animals and humans to be more perceptive or more effectively use visual stimuli to alter behavior such as in advertising, posting signs or signals such as in vehicular traffic, or assist animal trainers in interacting with their subjects more effectively of efficiently. References Bhatt, R. S., Wasserman, E. A., Reynolds, W. F. Jr., & Knauss, K. S. (1988). Conceptual behavior in pigeons: Categorization of both familiar and novel examples from four classes of natural and artificial stimuli. Journal of Experimental Psychology: Animal Behavior Processes, 14, 219-234. Wasserman, E. A., Kiedinger, R. E., & Bhatt, R. S.(1988). Conceptual behavior in pigeons: Categories, subcategories, and pseudocategories. Journal of Experimental Psychology: Animal Behavior Processes, 14, 235-246. Watanabe, S., Sakamoto, J., & Wakita, M. (1995). Pigeons' discrimination of paintings by Monet and Picasso. Journal of the Experimental Analysis of Behavior, 63, 165-174.