Maria Kon

Induced grouping refers to the influence of a perceived group of elements on the grouping of another set of elements that cannot be explained by other grouping principles. Vickery (2008) first highlighted this phenomenon and, despite convincing demonstrations of this principle, seems to be the only direct study. Here we report two successful large sample replications of one of Vickery’s experiments. We also explain Vickery’s results with a cortical model of visual grouping and selection. We extended a previous model, so that it performs a feature-based search of an image for a target. We show that induced grouping effects are the result of a connection strategy that links together target pairs in a visual search task combined with a selection strategy that tends to place a selection signal at locations close to the target pair features. These strategies interact because the connection strategy that links target pairs also sometimes links inducing elements, thereby influencing the selection signal location. The model extension plays a key role in explaining this phenomenon and enables the model to simulate other tasks, like visual search, where the observer uses a dynamic and feature-guided selection process.

We describe the verbs _pursue_, _chase_, and _follow_ as “tracking” verbs because they share conceptual similarities: they are all motion verbs that describe a dynamic spatial relation between two entities, as in “the cat chased the mouse”. What distinguishes them from one another? If, as some cognitive scientists argue, mental simulations underlie the way the mind processes all motion verbs — including those that describe static scenarios, such as run in “the road runs through the desert” — then those simulations may explain the differences between tracking verbs. For instance, _chase_ and _pursue_ may describe conceptually faster motion than _follow_. We tested this hypothesis in two experiments. The studies presented participants with imagery of one car chasing another along a straight road. In Experiment 1, participants estimated the distance that the pursued car would travel 3 seconds into the future by dragging a slider to an appropriate point on the road. In Experiment 2, participants estimated the distance by selecting from several distance options on a logarithmic scale. Both studies validated the hypothesis that _chase_ and _pursue_ describe faster motion, i.e., participants reliably estimated longer distances for descriptions that included those verbs. We place the results in the context of broader theories of pursuit perception and verb comprehension.

Although we are generally good at observing a busy scene and determining whether it contains one agent pursuing another, we are not immune to making errors and may identify a pursuit when there is none. Further, we may have difficulty articulating exactly what information allowed us to determine whether there was a pursuit. To gain a better measure of when people correctly or erroneously detect pursuit, we designed a novel pursuit detection task. To compare performance given different strategies, we developed a cognitive model that can perform this task. The results of our pursuit detection experiment indicate that, indeed, people typically identify pursuit events correctly, but they make infrequent yet systematic errors for particular scenes. When the model implements specific strategies, simulation results are well correlated with empirical results. Moreover, the model makes the same errors as human participants. We show how the empirical results can be accounted for in terms of decision criteria indicated by high performing model strategies.

Attention and memory are considered the bedrock of intellectual function. Further, disruptions of attention and memory are a core characteristic of many human disorders. However, neither attention nor memory are unitary constructs and little is known about the relation between them. Understanding how the human brain gives rise to such complex cognitive processes remains one of our biggest challenges, and modeling can be a useful tool in that pursuit. Simple models composed of neurons with physiologically plausible properties (such as those widely observed in the brain) can provide insight into (a) the meaning of these physiological properties, (b) how specific mechanisms may relate to behaviors such as attention and memory, (c) how lesions of the same network can result in either attention and memory correlations (similarities) or lack of correlations (differences), and finally (d) how such instantiations can impact the understanding of how attention and memory are represented and related in the brain. Most importantly, models can be helpful in making predictions and suggesting new ways to impact and the improve treatment of human disorders of attention and memory.

A classic finding reported in Beck (1966a) is that observers tend to indicate a more natural texture break between a set of T’s and tilted T’s than between a set of T’s and backward L’s. This finding has played a prominent role in discussions about the properties of texture segmentation and in the development of computational theories of texture segmentation. Due to the small sample size of the original study, we replicated the original experiment with a larger sample. Regrettably, we discovered that the description in Beck (1966a) is insufficient to allow us to reproduce the stimuli. For our replication, we created stimuli that seem consistent with the spirit of the original study. The results of the replication study partly matchBeck’s original data (r = .59), but conclusions indicated by these results deviate from some of Beck’s key conclusions.We also further explored the influential claim derived from a second experiment in Beck (1966a) that texture segmentation was not related to the perceived similarity of elements. Contrary to this claim, a re-analysis of Beck’s data indicates the counterintuitive conclusion that observers tend to indicate stronger texture segmentation between regions that are rated as being more similar. In a replication of the second experiment, this relation flipped so that the replication observers indicated stronger texture segmentation for regions with lower similarity ratings. Despite being influential, we conclude that there are substantial problems with the stimuli, analyses, and conclusions in Beck (1966a).

In a series of articles, Jacob Beck proposed that a variety of texture segmentation phenomena occurs due to emergent features that arise from “links” between elements with appropriate local properties, such as alignment, orientation, and proximity. His findings and ideas guided theoretical and computational models, and some of his demonstrations became textbook knowledge about visual perception. We build on this work in two ways. First, we provide a modern replication of a classic texture segmentation study using a much larger sample size. Overall, the replication agrees with Beck’s original findings, although there are some quantitative differences. Second, we show how to apply a quantitative model of visual cortex to Beck’s experiment and demonstrate that the model can explain many aspects of Beck’s findings. Key to the model’s success is cognitive control of connections between individual elements (akin to Beck’s “links” between elements) and a selection mechanism that makes it easy to identify how well elements within a region connect and how well different regions are disconnected. Overall, the model supports Beck’s claim that local properties can facilitate patterns of connections between stimulus elements and that some connection patterns allow an observer to easily distinguish textures.

There is growing interest in the effects of sports-related repetitive head impacts (RHIs) on athletes’ cognitive capabilities. This study examines the effect of RHIs in data collected from adolescent athletes to estimate the magnitude and longevity of RHIs on sensorimotor and cognitive performance. A non-linear regression model estimated the longevity of RHI effects by adding a half-life parameter embedded in an exponential decay function. A model estimate of this parameter allows the possibility of RHI effects to attenuate over time and introduces a mechanism to study the cumulative effect of RHIs. The posterior distribution of the half-life parameter associated with short-distance headers (<30 m) is centered around 6 days, whereas the posterior distribution of the half-life parameter associated with long-distance headers extends beyond a month. Additionally, the magnitude of the effect of each short header is around 3 times smaller than that of a long header. The results indicate that, on both tasks, response time (RT) changes after long headers are bigger in magnitude and last longer compared to the effects of short headers. Most important, we demonstrate that deleterious effects of long headers extend beyond 1 month. Although estimates are based on data from a relatively short-duration study with a relatively small sample size, the proposed model provides a mechanism to estimate long-term behavioral slowing from RHIs, which may be helpful to reduce the risk of additional injury. Finally, differences in the longevity of the effects of short and long RHIs may help to explain the large variance found between biomechanical input and clinical outcome in studies of concussion tolerance.

Although perceptual grouping has been widely studied, its mechanisms remain poorly understood. We propose a neural model of grouping that, through top-down control of its circuits, implements a grouping strategy involving both a connection strategy (which elements to connect) and a selection strategy (that defines spatiotemporal properties of a selection signal to segment target elements and facilitate identification). We apply the model to a letter discrimination task that investigated relationships among uniform connectedness and the grouping principles of proximity and shape similarity. Participants reported whether small circles formed a global letter E or H, and these circles could be connected by a line or be embedded in a matrix of squares. In the model, a good grouping strategy for this task consists of a connection strategy that connects circles but not squares for all conditions and a selection strategy that uses a selection signal of varying size, depending on whether squares were present. Consistent with empirical results, which were verified in two replication studies, model performance is worse with distractor squares, and line connectors improve performance only in the condition with squares. Rather than relying on abstract grouping principles, we show how the empirical results can be explained in terms of observers implementing a task-dependent grouping strategy that promotes overall performance.

A fundamental characteristic of human visual perception is the ability to group together disparate elements in a scene and treat them as a single unit. The mechanisms by which humans create such groupings remain unknown, but grouping seems to play an important role in a wide variety of visual phenomena. I propose a neural model of grouping; through top-down control of its circuits, the model implements a grouping strategy that involves both a connection strategy (which elements to connect) and a selection strategy (spatiotemporal properties of a selection signal that segments target elements to facilitate identification). With computer simulations I explain how the circuits work and show how they can account for a wide variety of Gestalt principles of perceptual grouping. Additionally, I extend the model so that it can simulate visual search tasks. I show that when the model uses particular grouping strategies, simulated results closely match empirical results from replication experiments of three visual search tasks. In these experiments, perceptual grouping was induced by proximity and shape similarity (Palmer & Beck, 2007), by the spacing of irrelevant distractors and size similarity (Vickery, 2008), or by the proximity of dots and the proximity and shape similarity of line figures (Trick & Enns, 1997). Thus, I show that the model accounts for a variety of grouping effects and indicates which grouping strategies were likely used to promote performance in three visual search tasks.

In his constructive and well-informed commentary, Andrei Khrennikov acknowledges a privileged status of classical probability theory with respect to statistical analysis. He also sees advantages offered by the Contextuality-by-Default theory, notably, that it “demystifies quantum mechanics by highlighting the role of contextuality,” and that it can detect and measure contextuality in inconsistently connected systems. He argues, however, that classical probability theory may have difficulties in describing empirical phenomena if they are described entirely in terms of observable events. We disagree: contexts in which random variables are recorded are as observable as the variables’ values. Khrennikov also argues that the Contextuality-by-Default theory suffers the problem of non-uniqueness of couplings. We disagree that this is a problem: couplings are all possible ways of imposing counterfactual joint distributions on random variables that de facto are not jointly distributed. The uniqueness of modeling experiments by means of quantum formalisms brought up by Khrennikov is achieved for the price of additional, substantive assumptions. This is consistent with our view of quantum theory as a special-purpose generator of classical probabilities. Khrennikov raises the issue of “mental signaling,” by which he means inconsistent connectedness in behavioral systems. Our position is that it is as inherent to behavioral systems as their stochasticity.

One can often encounter claims that classical (Kolmogorovian) probability theory cannot handle, or even is contradicted by, certain empirical findings or substantive theories. This note joins several previous attempts to explain that these claims are unjustified, illustrating this on the issues of (non)existence of joint distributions, probabilities of ordered events, and additivity of probabilities. The specific focus of this note is on showing that the mistakes underlying these claims can be precluded by labeling all random variables involved contextually. Moreover, contextual labeling also enables a valuable additional way of analyzing probabilistic aspects of empirical situations: determining whether the random variables involved form a contextual system, in the sense generalized from quantum mechanics. Thus, to the extent the Wang-Busemeyer QQ equality for the question order effect holds, the system describing them is noncontextual. The double-slit experiment and its behavioral analogues also turn out to form a noncontextual system, having the same probabilistic format (cyclic system of rank 4) as the one describing spins of two entangled electrons.

A fundamental characteristic of human visual perception is the ability to group together disparate elements in a scene and treat them as a single unit. The mechanisms by which humans create such groupings remain unknown, but grouping seems to play an important role in a wide variety of visual phenomena, and a good understanding of these mechanisms might provide guidance for how to improve machine vision algorithms. Here, we build on a proposal that some groupings are the result of connections in cortical area V2 that join disparate elements, thereby allowing them to be selected and segmented together. In previous instantiations of this proposal, connection formation was based on the anatomy (e.g., extent) of receptive fields, which made connection formation obligatory when the stimulus conditions stimulate the corresponding receptive fields. We now propose dynamic circuits that provide greater flexibility in the formation of connections and that allow for top-down control of perceptual grouping. With computer simulations we explain how the circuits work and show how they can account for a wide variety of Gestalt principles of perceptual grouping, texture segmentation tasks, amodal illusory contours, and ratings of perceived groupings. We propose that human observers use such top-down control to implement task-dependent connection strategies that encourage particular groupings of stimulus elements in order to promote performance on various visual tasks.

Where does the study of the flow of time belong: physics, the cognitive sciences, philosophy, or somewhere else? Physicists and philosophers have set themselves up into two camps: those who believe there is genuine flow or becoming in the world and those who believe there is just a block of events. What had not been considered is whether the subjective feeling of flow of time is the same the world over, whether it could be tampered with by brain injury, or whether it is present at all developmental stages. We lay out the problem, explain terms, and provide synopses of relevant ideas.

Recent philosophical work on temporal experience offers generic models that are often assumed to apply to all sensory modalities. I show that the models serve as broad frameworks in which different aspects of cognitive science can be slotted and, thus, are beneficial to furthering research programs in embodied music cognition. Here I discuss a particular feature of temporal experience that plays a key role in such philosophical work: a distinction between the experience of succession and the mere succession of experiences. I question the presupposition that there is such an evident, clear distinction and suggest that, instead, how the distinction is drawn is context-dependent. After suggesting a way to modify the philosophical models of temporal experience to accommodate this contextdependency, I illustrate that these models can fruitfully incorporate features of research projects in embodied musical cognition. To do so I supplement a modified retentionalist model with aspects of recent work that links bodily movement with musical perception (Godøy, 2006; 2010a; Jensenius, Wanderley, Godøy, and Leman, 2010). The resulting model is shown to facilitate novel hypotheses, refine the notion of context-dependency and point towards means of extending the philosophical model and an existent research program.

One of Julian Barbour’s main aims is to solve the problem of time that appears in quantum geometrodynamics (QG). QG involves the application of canonical quantization procedure to the Hamiltonian formulation of General Relativity. The problem of time arises because the quantization of the Hamiltonian constraint results in an equation that has no explicit time parameter. Thus, it appears that the resulting equation, as apparently timeless, cannot describe evolution of quantum states. Barbour attempts to resolve the problem by allegedly eliminating time from his interpretation of QG. In order to evaluate the efficacy of his solution, it is necessary to ascertain in what sense time has been eliminated from his theory. I proceed to do so by developing a form of conceptual analysis that is applicable to the concept of time in physical theories and applying this analysis to Barbour’s account.

This paper has two aims. First, to bring together the models of temporal phenomenology on offer and to present these using a consistent set of distinctions and terminologies. Second, to examine the methodologies currently practiced in the development of these models. To that end we present an abstract characterisation in which we catalogue all extant models. We then argue that neither of the two extreme methodologies currently discussed is suitable to the task of developing a model of temporal phenomenology. An entirely top-down methodology is largely insensitive to empirical findings, while one that is entirely bottom-up is insensitive to both introspective evidence and more general philosophical considerations. We recommend a methodology that falls between these two extremes