Michael L. Kalish

Perception of Translational Heading From Optical Flow

Radial patterns of optical flow produced by observer translation could be used to perceive the direction of self-movement during locomotion, and a number of formal analyses of such patterns have recently appeared. However, there is comparatively little empirical research on the perception of heading from optical flow, and what data there are indicate surprisingly poor performance, with heading errors on the order of 5 degrees-10 degrees. We examined heading judgments during translation parallel, perpendicular, and at oblique angles to a random-dot plane, varying observer speed and dot density. Using a discrimination task, we found that heading accuracy improved by an order of magnitude, with 75%-correct thresholds of 0.66 degrees in the highest speed and density condition and 1.2 degrees generally. Performance remained high with displays of 63-10 dots, but it dropped significantly with only 2 dots; there was no consistent speed effect and no effect of angle of approach to the surface. The results are inconsistent with theories based on the local focus of outflow, local motion parallax, multiple fixations, differential motion parallax, and the local maximum of divergence. But they are consistent with Gibsons (1950) original global radial outflow hypothesis for perception of heading during translation.

Language evolution by iterated learning with bayesian agents.

Languages are transmitted from person to person and generation to generation via a process of iterated learning: people learn a language from other people who once learned that language themselves. We analyze the consequences of iterated learning for learning algorithms based on the principles of Bayesian inference, assuming that learners compute a posterior distribution over languages by combining a prior (representing their inductive biases) with the evidence provided by linguistic data. We show that when learners sample languages from this posterior distribution, iterated learning converges to a distribution over languages that is determined entirely by the prior. Under these conditions, iterated learning is a form of Gibbs sampling, a widely-used Markov chain Monte Carlo algorithm. The consequences of iterated learning are more complicated when learners choose the language with maximum posterior probability, being affected by both the prior of the learners and the amount of information transmitted between generations. We show that in this case, iterated learning corresponds to another statistical inference algorithm, a variant of the expectation-maximization (EM) algorithm. These results clarify the role of iterated learning in explanations of linguistic universals and provide a formal connection between constraints on language acquisition and the languages that come to be spoken, suggesting that information transmitted via iterated learning will ultimately come to mirror the minds of the learners.

On the sufficiency of the velocity field for perception of heading

All models of self-motion from optical flow assume the instantaneous velocity field as input. We tested this assumption for human observers using random-dot displays that simulated translational and circular paths of movement by manipulating the lifetime and displacement of individual dots. For translational movement, observers were equally accurate in judging direction of heading from a “velocity field” with a two-frame dot life and a “direction field” in which the magnitudes of displacement were randomized while the radial pattern of directions was preserved, but at chance with a “speed field” in which the directions were randomized, preserving only magnitude. Accuracy declined with increasing noise in vector directions, but remained below 2.6° with a 90° noise envelope. Thus, the visual system uses the radial morphology of vector directions to determine translational heading and can tolerate large amounts of noise in this pattern. For circular movement, observers were equally accurate with a 2-frame “velocity field”, 3-frame “acceleration” displays, and 2-frame and 3-frame “direction fields”, consistent with the use of the pattern of vector directions to locate the center of rotation. The results indicate that successive independent velocity fields are sufficient for perception of translational and circular heading.

Iterated learning: Intergenerational knowledge transmission reveals inductive biases

Cultural transmission of information plays a central role in shaping human knowledge. Some of the most complex knowledge that people acquire, such as languages or cultural norms, can only be learned from other people, who themselves learned from previous generations. The prevalence of this process of “iterated learning” as a mode of cultural transmission raises the question of how it affects the information being transmitted. Analyses of iterated learning utilizing the assumption that the learners are Bayesian agents predict that this process should converge to an equilibrium that reflects the inductive biases of the learners. An experiment in iterated function learning with human participants confirmed this prediction, providing insight into the consequences of intergenerational knowledge transmission and a method for discovering the inductive biases that guide human inferences.

Population of linear experts: knowledge partitioning and function learning

Knowledge partitioning is a theoretical construct holding that knowledge is not always integrated and homogeneous but may be separated into independent parcels containing mutually contradictory information. Knowledge partitioning has been observed in research on expertise, categorization, and function learning. This article presents a theory of function learning (the population of linear experts model--POLE) that assumes people partition their knowledge whenever they are presented with a complex task. The authors show that POLE is a general model of function learning that accommodates both benchmark results and recent data on knowledge partitioning. POLE also makes the counterintuitive prediction that a persons distribution of responses to repeated test stimuli should be multimodal. The authors report 3 experiments that support this prediction.

Theoretical and empirical evidence for the impact of inductive biases on cultural evolution

The question of how much the outcomes of cultural evolution are shaped by the cognitive capacities of human learners has been explored in several disciplines, including psychology, anthropology and linguistics. We address this question through a detailed investigation of transmission chains, in which each person passes information to another along a chain. We review mathematical and empirical evidence that shows that under general conditions, and across experimental paradigms, the information passed along transmission chains will be affected by the inductive biases of the people involved—the constraints on learning and memory, which influence conclusions from limited data. The mathematical analysis considers the case where each person is a rational Bayesian agent. The empirical work consists of behavioural experiments in which human participants are shown to operate in the manner predicted by the Bayesian framework. Specifically, in situations in which each persons response is used to determine the data seen by the next person, people converge on concepts consistent with their inductive biases irrespective of the information seen by the first member of the chain. We then relate the Bayesian analysis of transmission chains to models of biological evolution, clarifying how chains of individuals correspond to population-level models and how selective forces can be incorporated into our models. Taken together, these results indicate how laboratory studies of transmission chains can provide information about the dynamics of cultural evolution and illustrate that inductive biases can have a significant impact on these dynamics.

Simplified learning in complex situations: knowledge partitioning in function learning.

The authors explored the phenomenon that knowledge is not always integrated and consistent but may be partitioned into independent parcels that may contain mutually contradictory information. In 4 experiments, using a function learning paradigm, a binary context variable was paired with the continuous stimulus variable of a to-be-learned function. In the first 2 experiments, when context predicted the slope of a quadratic function, generalization was context specific. Because context did not predict function values, it is suggested that people use context to gate separate learning of simpler partial functions. The 3rd experiment showed that partitioning also occurs with a decreasing linear function, whereas the 4th study showed that partitioning is absent for a linearly increasing function. The results support the notion that people simplify complex learning tasks by acquiring independent parcels of knowledge.

The Dimensionality of Perceptual Category Learning: A State-Trace Analysis

State-trace analysis was used to investigate the effect of concurrent working memory load on perceptual category learning. Initial reanalysis of Zeithamova and Maddox (2006, Experiment 1) revealed an apparently two-dimensional state-trace plot consistent with a dual-system interpretation of category learning. However, three modified replications of the original experiment found evidence of a single resource underlying the learning of both rule-based and information integration category structures. Follow-up analyses of the Zeithamova and Maddox data, restricted to only those participants who had learned the category task and performed the concurrent working memory task adequately, revealed a one-dimensional plot consistent with a single-resource interpretation and the results of the three new experiments. The results highlight the potential of state-trace analysis in furthering our understanding of the mechanisms underlying category learning.

Working memory does not dissociate between different perceptual categorization tasks.

Working memory is crucial for many higher level cognitive functions, ranging from mental arithmetic to reasoning and problem solving. Likewise, the ability to learn and categorize novel concepts forms an indispensable part of human cognition. However, very little is known about the relationship between working memory and categorization. This article reports 2 studies that related peoples working memory capacity (WMC) to their learning performance on multiple rule-based and information-integration perceptual categorization tasks. In both studies, structural equation modeling revealed a strong relationship between WMC and category learning irrespective of the requirement to integrate information across multiple perceptual dimensions. WMC was also uniformly related to peoples ability to focus on the most task-appropriate strategy, regardless of whether or not that strategy involved information integration. Contrary to the predictions of the multiple systems view of categorization, working memory thus appears to underpin performance in both major classes of perceptual category-learning tasks.

Systems of Category Learning. Fact or Fantasy

Abstract Psychology abounds with vigorous debates about the need for one or more underlying mental processes or systems to explain empirical observations. The field of category learning provides an excellent exemplar. We present a critical examination of this field focusing on empirical, methodological, and mathematical modeling considerations. We review what is often presented as the “best evidence” for multiple systems of category learning and critique the evidence by considering three questions: (1) Are multiple-systems accounts the only viable explanations for reported effects? (2) Are the inferences sound logically and methodologically? (3) Are the mathematical models that can account for behavior sufficiently constrained, and are alternative (single-system) models applicable? We conclude that the evidence for multiple-systems accounts of category learning does not withstand such scrutiny. We end by discussing the varieties of explanation that can be offered for psychological phenomena and highlight why multiple-systems accounts often provide an illusory sense of scientific progress.