John Campbell, Department of Philosophy, UC Berkeley
Friday, November 20, 2020
Temporal Reasoning and Free Will
This talk is about the connection between two puzzles, about human temporal reasoning and about free will. Practically all animals have some kind of sensitivity to time, whether a mere sensitivity to time of year or day, or an ability to perform complex calculations on the results of data from interval timing. Humans seem, though, universally, to think of time as linear: to think in terms of an ongoing time in which one’s birth and death can be plotted, for example. This seems to be unique to humans. And it’s not just a capacity we have: human languages tend to make it mandatory, through the demands of tense, for example, to indicate the place of events reported in this linear time. So why do humans do this, when the other animals seem to get along fine without linear temporal reasoning?
The other puzzle is the characterization of free will. Freedom is usually taken to be a matter of a capacity for self-control, or self-regulation, of one kind or another. But there is a similar puzzle here: why do humans need and use this capacity for self-regulation, when other animals seem to manage without it? The answer I propose is that this has to do with the unique causal structure of human psychology, in which we find singular causation that is not grounded in patterns of general causation. This means that humans have unique problems with social coordination and with the organization of the individual’s own time. And our ability to think in terms of linear time lets us solve them, more or less.
Emily Cooper, School of Optometry, UC Berkeley
Friday, October 30, 2020
Perceptual Science for Augmented Reality
Recent years have seen impressive advances in near-eye display systems for augmented reality. In these systems, digital content is merged with the user’s view of the physical world. There are, however, unique perceptual challenges associated with designing a display system that can seamlessly blend the real and the virtual. By understanding the relevant principles that underlie our visual perception, I will show how we can address some of these challenges.Link to event will be provided closer to date.
Keith Holyoak, Department of Psychology, UCLA
Friday, October 23, 2020
Abstract Semantic Relations in Mind, Brain, and Machines
Abstract semantic relations (e.g., category membership, part-whole, antonymy, cause-effect) are central to human intelligence, underlying the distinctively human ability to reason by analogy. I will describe a computational project (Bayesian Analogy with Relational Transformations) that aims to extract explicit representations of abstract semantic relations from non-relational inputs automatically generated by machine learning. BART’s representations predict patterns of typicality and similarity for semantic relations, as well as similarity of neural signals triggered by semantic relations during analogical reasoning. In this approach, analogy emerges from the ability to learn and compare relations; mapping emerges later from the ability to compare patterns of relations.
Richard Futrell, Department of Language Science, UC Irvine
Friday, September 18, 2020
Efficiency-based models of natural language: Predicting word order universals Using information theory
Why is human language the way it is? I claim that human languages can be modeled as systems for efficient communication: that is, codes that maximize information transfer subject to constraints on the cognitive resources used during language production and comprehension. I use this efficiency-based framework to formulate quantitative theories of word order, aiming to explain the cross-linguistic universals of word order documented by linguists as well as the statistical distribution of word orders in massively cross-linguistic corpus studies.
I present three results. First, I show that word orders in 54 languages are shaped by dependency locality: a pressure for words linked in syntactic dependencies to be close to each other, which minimizes working memory usage during language processing. Second, I introduce a new model of information-processing difficulty in online language processing, which simultaneously captures effects of probabilistic expectations and working memory constraints, recovering dependency locality as a special case and making new predictions, in particular about adjective order. Third, I present a computational framework in which grammars can be directly optimized for efficiency. When grammars are optimized to maximize information transfer while minimizing processing difficulty, they end up reproducing 8 typological universals of word order.