Researchers have developed a new method to interpret the West Antarctic Ice Sheet Divide core, extracting data on sea surface temperatures and tracking climate change. The technique reveals signs of an entropy increase around 10,000 years ago, suggesting human impacts on climate began before the Industrial Revolution.
The study suggests that languages develop word order rules to minimize miscommunication across a noisy channel. Researchers found a correlation between word order and case markings, with strongly case-marked languages like Japanese defaulting to SOV word order.
Princeton engineers developed a technique to clarify images using rays of light scattered by clouds, human tissue, or murky water. The method, known as stochastic resonance, can potentially improve signal technologies such as sonograms, radar systems, and night vision goggles.
A new theory proposes a method to determine fair CEO compensation using information theory and statistical thermodynamics. The ratio of CEO pay to the lowest employee salary has increased significantly in recent years, with US CEOs averaging around 344-to-1, raising questions about market efficiency.
Research from the University of Alberta found that sold-out products create a sense of urgency, increasing consumer demand for similar items. Consumers infer that if a product is sold out, it must be good and therefore desirable.
Researchers investigated two theories of forgetfulness and found that temporal confusability, not decay, is the main cause of forgetting over the short term. The study reveals that making information distinct can alleviate and even reverse the classic pattern of forgetting.
Researchers used information theory to analyze humpback whale songs and found a hierarchical structure with multiple layers of repetition. This study offers a new approach to studying animal communication and may inform research in evolutionary biology.
Researchers will investigate how brains perceive novelty and importance using a computational theory of surprise. The project combines experiments, modeling, and neurophysiological studies to develop a new understanding of how the brain codes for these concepts.
Researchers Jody O'Sullivan and Pierre Moulin have developed a theory that sets fundamental limits on the amount of information that can be reliably hidden in various data types. This breakthrough has significant implications for security applications, including detecting forgery and intercepting terrorist messages.