According to research, pigeons solve problems in a similar manner to artificial intelligence.
A recent research has discovered that pigeons’ problem-solving abilities are similar to artificial intelligence.
Pigeons are commonly disregarded as bothersome creatures, but in reality, they are incredibly intelligent beings with the ability to recognize faces, perceive vibrant colors, navigate intricate paths, deliver information, and even rescue individuals.
During the research, 24 pigeons were presented with different visual challenges. Some of these challenges were quickly learned and categorized within a few days, while others took several weeks. The study revealed that the pigeons used a similar method to AI models in order to make accurate decisions.
According to Edward Wasserman, a professor of experimental psychology at the University of Iowa and co-author of the study, the behavior of pigeons indicates that nature has developed an algorithm that is extremely efficient in acquiring difficult skills. This algorithm may not prioritize speed, but it demonstrates remarkable consistency.
Pigeons were presented with various stimuli on a screen, including lines of varying width, placement, and orientation, as well as sectioned and concentric rings. They were required to peck a button on either the right or left to determine the category to which the stimuli belonged. Successful identification resulted in a food reward in the form of a pellet, while incorrect responses yielded no reward.
According to Brandon Turner, a psychology professor at Ohio State University and lead author of the study, pigeons do not require specific guidelines. Instead, they acquire knowledge through experimentation. For instance, when presented with a visual labeled as “category A,” they would categorize anything resembling it as “category A” as well, demonstrating their ability to recognize similarities.
During the experiments, pigeons showed progress in correctly making choices from 55% to 95% of the time for simpler tasks. However, when faced with a more difficult challenge, their accuracy increased from 55% to 68%.
According to Onur Güntürkün, a professor of behavioral neuroscience at Ruhr University Bochum, studying humble animals such as pigeons can help us understand the limits of their associative minds. He was not part of the study but believes that this research demonstrates the impressive strength and cognitive abilities of associative systems.
The primary objective of an AI model is to identify patterns and make decisions. Studies have revealed that pigeons possess this same capability. Through learning from consequences, pigeons are able to effectively correct their mistakes when not provided with a food pellet. Additionally, pigeons utilize a similarity function to compare and find similarities between two objects.
Turner stated that using these two mechanisms alone, one can create a neural network or artificial intelligence that is capable of solving categorization problems. This suggests that the same mechanisms found in AI are also found in pigeons.
The researchers are seeking to work together with experts in pigeon and brain research. They aspire to apply these results to gain a better understanding of brain damage in humans.
Wasserman suggested that we could gain more understanding about the thought processes of birds. He praised their brains, noting that while small, they are highly capable of learning.
None of the pigeons were harmed during the study.