University at Buffalo (UB)
Wednesday, January 31, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Jean-Pierre Koenig, Ph.D.
Department of Linguistics
University at Buffalo
"What's "in" a Word?"
Recent work on the lexicon has stressed how much the syntactic context in which a word occurs is determined by its meaning and that words pattern into word classes on the basis of their meaning. This research suggests that if you know the meaning of a word, you can predict to a large measure the kind of sentences it occurs in. Taking this result as a point of departure, the first part of this talk will discuss how we can use the fact that lexical meaning determines syntactic context of occurrence to infer the structure of the semantic information encoded in words. In particular, my own research suggests that the semantic information included in verbs has two parts, a relational component (which covers notions like possesion, causality, having a mental representation...) and a modal component (which covers notions like negation, necessity, and time).
In the second part of this talk, I will investigate an unanswered question of the recent research on the relationship between syntax and semantics, namely how we can determine the semantic information included in lexical entries. I will hypothesize that two criteria affect the inclusion of participant information in lexical entries: (1) Whether participant information is obligatory and (2) Whether participant information is specific to a restricted set of verbs. I will present the result of several experiments which draw on a comprehensive survey of the English verbal lexicon.
For a printable version of this file click here
Wednesday, February 7, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Kenneth Forbus, Ph.D.
Department of Computer Science
Northwestern University, Evanston, IL
"Qualitative Physics as a Language
for Cognitive Modeling"
Most research in qualitative reasoning has been driven by applications in engineering, education, and other areas. However, I believe that perhaps the most important role for qualitative physics is providing representations and reasoning techniques for cognitive modeling. This talk will examine ideas from qualitative physics in this light, including speculations on how they can be used for modeling developmental results, as a component in natural language semantics, and as a bridge between perceptual and conceptual representations.
For a printable version of this file click here
Wednesday, February 14, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Carl Alphonce, Ph.D.
Dept. of Computer Science and Engineering, UB
"Computational Implementations of Theories"
Theories of human cognition and human cognitive performance are typically complex, reflective no doubt of the complexity of the underlying system and processes which they seek to model. It can be difficult at times to grasp the consequences of such a theory by manual inspection. Implementing the theory can prove helpful. I will explore selected issues related to the computational implementation of theories, drawing examples especially from natural language syntax and sentence processing.
For a printable version of this file click here
Wednesday, February 21, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
David Smith, Ph.D.
Department of Psychology
Department of Music
Center for Cognitive Science
University at Buffalo
"Journey to the Center of the Category"
Knowlton and Squire showed that amnesics perform relatively normally when categorizing dot patterns derived from an underlying prototype. But they are impaired in performing an old/new recognition task with similar materials. Knowlton and Squire concluded that categorization performance relies on an implicit memory system--intact in amnesics--that represents category-level information in the form of prototypes. They concluded that recognition performance relies on an explicit memory system--impaired in amnesia--that contains declarative memories about specific exemplars. This dissociation in amnesia between categorization and recognition seems to challenge a unitary exemplar theory that assumes a single exemplar-based processing system. Responding to this challenge, Nosofsky and Zaki (1998) derived new formal models of categorization and recognition that were intended to explain the amnesia data using only exemplar-based processing.
Our theoretical analysis of Knowlton and Squire's data, and of Nosofsky and Zaki's reinterpretation of them, suggests these conclusions. 1) Comparing to-be-categorized items to a category center or prototype produces strong prototype advantages and steep typicality gradients, whereas comparing to-be-categorized items to the training exemplars that surround the prototype produces weak prototype advantages and flat typicality gradients. 2) Participants (including amnesics) show the former pattern, suggesting their use of prototypes. 3) Exemplar models account poorly for these categorization data, but prototype models account well for them. 4) The recognition data suggest that controls use an exemplar-memorization process more powerfully than amnesics. By pairing categorization based in prototypes with recognition based in memorized exemplars, we support the idea of multiple systems or processes underlying categorization and recognition, we extend other recent accounts of cognitive performance that intermix prototypes and exemplars, and we reinforce traditional interpretations of the categorization-recognition dissociation in amnesia.
For a printable version of this file click here
Wednesday, February 28, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
For a printable version of this file click here
Wednesday, March 14, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Donald Stuss, Ph.D.
Rotman Research Institute
Baycrest Centre for Geriatric Care
Toronto, Canada
"Attentional Functioning:
The Roles of the Frontal Lobes"
The functions of the frontal lobes have been difficult to differentiate. Terms such as supervisory system and executive control have been used. Recently, the fractionation of frontal lobe functions has been demonstrated. In this presentation, a model of the functions of the frontal lobes in attention will be presented, based on a modified version of the Supervisory Attentional System of Shallice. Various attentional tasks assessing hypothesized frontal lobe attentional functions were presented to patients with lesions in various regions of the frontal lobes as well as to patients with non-frontal lesions. These tasks include the Stroop, feature integration decision reaction time tasks, and a "select-what, respond-where" paradigm. The results indicate that the anterior attentional system is a complex interaction of distinct attentional processes, related to different regions of the frontal lobes, and integrated with posterior brain regions.
For a printable version of this file click here
Wednesday, March 21, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Amanda Woodward, Ph.D.
Department of Psychology
Center for Early Childhood Research
University of Chicago
How Infants Make Sense of Intentional Action
The ability to make sense of the actions of other people is critical to human functioning, and the origins of this ability have inspired much speculation and debate. Until recently, there has been little empirical evidence available to inform our understandings this ability in infants. My collaborators and I have begun to address this gap by asking whether infants make sense of human behavior in ways that are continuous with later concepts of intentional action. Adults have a strong propensity to construe behavior as goal-directed. Mature reasoners represent action sequences in terms of the actor's probable goals, weighting goal-related information more heavily in memory than aspects of the events that were not relevant to the actor's goals. With this in mind, we began by assessing infants' encoding of a simple goal-directed action. Babies saw a person reach through a distinctive path in order to grasp one of two toys. This event was repeated until infants had habituated to it. Then, the positions of the toys were reversed, and infant saw test events in which either the path of motion or the goal object of the actor's reach had changed. Six- and 9-month-old infants showed a greater novelty response to the latter events than to the former. Infants at both ages who saw an inanimate object reach toward and grasp or touch the toy did not show this pattern. That is, infants selectively encoded the relation between a human actor and the object she grasped, and do not do this for inanimate graspers. Thus, there is one way in which infant reasoning is continuous with mature reasoning.
Subsequent findings revealed an interesting set of limitations to infants' ability to interpret actions as goal-directed: (1) Infants do not encode all events in which a person touches an object as goal-directed; (2) Infants' encoding of action changes as their knowledge about specific actions changes; and, (3) Infants' encoding of other people's actions is related to their own experience as actors. These findings indicate that infants' initial conceptions of goal-directed action are grounded in their knowledge about specific acts, rather than being the product of innate abstract conceptions of intentional action. In a final study, we explored one route for infants' moving beyond these early, specific notions of goal-directed action. Mature reasoners are not limited to seeing a canonical set of actions as goal-directed. Rather, we can freely interpret action in context, drawing on our knowledge about actors and situations. We found that in one very simple context, 12-month-old infants were similarly able to use the behavioral and physical context of a novel action to interpret it as goal-directed.
For a printable version of this file click here
Wednesday, March 28, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Daeyeol Lee, Ph.D.
Dept. of Brain and Cognitive Sciences
Center for Visual Science
University of Rochester
"Neural Mechanisms for Learning
Sequential Movements"
People often achieve their behavioral goals by various movement sequences. In most cases, these movement sequences are learned and through training, their speed and accuracy improve. I will present some of the results from our behavioral studies showing that different dimensions of movement sequences (e.g., temporal vs. spatial) are not learned independently, but rather acquired as an integrated unit. I will then describe the results from our single-cell recording studies in non-human primates. We recorded the activity of neurons in the supplemetary motor area and the primary motor cortex using a multi-electrode recording system, while the animals were performing a serial reaction time task. The results indicate that majority of neurons in both areas display changes in their activity during the course of sequence learning, suggesting that information about movement sequence is distributed in multiple brain areas.
For a printable version of this file click here
Wednesday, March 21, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
David Mark, Ph.D.
Department of Geography, UB
and
Barry Smith, Ph.D.
Department of Philosophy, UB
"Geographic Objects and Their Categories"
This colloquium will report the results of a series of experiments designed to establish how non-expert subjects conceptualize geospatial phenomena. Subjects were asked to give examples of geographic categories in response to a series of differently phrased elicitations. The results yield an ontology of geographic categories--a catalogue of the prime geospatial concepts and categories shared in common by human subjects independently of their exposure to scientific geography. When combined with nouns such as feature and object, the adjective geographic elicited almost exclusively elements of the physical environment of geographic scale or size, such as mountain, lake, and river. The phrase things that could be portrayed on a map , on the other hand, produced many geographic scale artifacts (roads, cities, etc.) and fiat objects (states, countries, etc.), as well as some physical feature types. These data reveal considerable mismatch as between the meanings assigned to the terms 'geography' and 'geographic' by scientific geographers and by ordinary subjects, so that scientific geographers are not in fact studying geographic phenomena as such phenomena are conceptualized by naive subjects. The data suggest, rather, a special role in determining the subject-matter of scientific geography for the concept of what can be portrayed on a map. This work has implications for work on usability and interoperability in geographic information science, and it throws light also on subtle and hitherto unexplored ways in which ontological terms such as 'object', 'entity', and 'feature' interact with geographic concepts.
For a printable version of this file click here
Wednesday, April 18, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Ellen Prince, Ph.D.
Department of Linguistics
University of Pennsylvania
On Identifying the Topic and
Why We Might Want To Do So
Topics' are usually defined in structural terms, typically the leftmost NP in a clause (Halliday 1967; Gundel 1974, 1885...; Reinhart 1981; Foley and Van Valin 1984...). However, topics so defined are clearly of no use in research attempting to correlate linguistic form and cognitive function, given that the cognitive notion 'topic' is in fact defined by linguistic form. In contrast, Centering Theory, a computationally tractable means of modeling local attention in discourse provides us with an algorithm for identifying what I believe we intuitively think of as 'topic', without defining it on the basis of its form. In this presentation, I shall show how this works and shall cite some (perhaps surprising) research findings on the relationship between topic thus defined and the syntactic forms known as Subject-Prodrop, 'Topicalization', and Left-Dislocation.
Wednesday, August 29, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Francisco Gil-White, Ph.D.
Department of Psychology
University of Pennsylvania
"Cognitively Speaking,
What is an Ethnic Category?"
If, despite the fact that ethnic essences do not exist, ethnic actors nevertheless represent ethnies as essentialized natural groups, we must understand why. This talk will present and defend the argument that humans process ethnic groups (and a few other related social categories) as if they were species - with the associated essentialism - because their surface similarities to species make them inputs to the living kinds mental module that initially evolved to process species-level categories. The similarities in processing between species and ethnic categories will be explored (the main responsible similarities are category-based endogamy, and descent-based membership), and an evolutionary argument for this pyschological borrowing from 'living kinds' reasoning to the social domain will be defended. In a nutshell, I will argue that thinking about ethnies as if they were species was adaptive in the ancestral environment because it solved problems of inference and coordination in the domain of interactional norms.
Co-sponsored by the Departments of Anthropology and Philosophy
For a printable version of this file click here
Wednesday, September 12, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Ezra Zubrow, Ph.D.
Department of Anthropology
University at Buffalo
"The Origin of Music"
This paper reports on the recent results of a joint project between the University of Buffalo and University of Cambridge. There are four results:
For a printable version of this file click here
Wednesday, September 19, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Leonard Talmy, Ph.D.
Department of Linguistics
Center for Cognitive Science
University at Buffalo
"The Representation of Spatial Structure in Spoken and Signed Language"
Linguistic research to date has determined many of the factors that structure the spatial schemas found across spoken languages. It is now feasible to integrate these factors and to determine the comprehensive system they constitute for spatial structuring in spoken language. This system is characterized by several features: It has a relatively closed universally available inventory of fundamental spatial elements that are combined to form whole schemas. It has a relatively closed set of categories that these elements appear in. And it has a relatively closed small number of particular elements in each category, hence, of spatial distinctions that each category can ever mark.
An examination of signed language shows that its structural representation of space systematically differs from that in spoken language in the direction of what appear to be the structural characteristics of scene parsing in visual perception. Such differences include the following: Signed language can mark finer spatial distinctions with its inventory of more structural elements, more categories, and more elements per category. It represents many more of these distinctions in any particular expression. It also represents these distinctions independently in the expression, not bundled together into "pre-packaged" schemas. And its spatial representations are largely iconic with visible spatial characteristics. The findings suggest that instead of some discrete whole-language module, spoken language and signed language are both based on some more limited core linguistic system that then connects with different further subsystems for the full functioning of the two different language modalities.
For a printable version of this file click here
Wednesday, September 26, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Kristin Tjaden, Ph.D.
Department of Communicative Disorders and Sciences
University at Buffalo
"Acoustic-Perceptual Distinctiveness and Coarticulatory patterns in Dysarthria"
The dysarthrias are a group of communicative disorders resulting from impairment to central and/or peripheral nervous system structures important for the motor execution of speech. From a functional viewpoint, speech intelligibility and naturalness may be impaired owing to reduced speed, strength, range, accuracy, and timing of speech movements in the respiratory-laryngeal, velopharyngeal, and oral articulatory mechanisms. Although a great deal of progress has been made in characterizing the speech production deficits associated with the various dysarthrias, vocal tract activity in dysarthria remains poorly understood. The current presentation focuses on oral articulatory impairments associated with Parkinson's disease, Amyotrophic Lateral Sclerosis (ALS), and Multiple Sclerosis, as inferred from the acoustic speech signal. In addition to describing how these neurologic diseases affect the articulatory-acoustic working space for individuals' habitual or normal speech mode, changes in the acoustic working space associated with speech rate and vocal intensity manipulations will be discussed. The relationship between the size of the acoustic working space and auditory-perceptual impressions of speech also will be discussed as well as coarticulatory differences for individuals with dysarthria and neurologically healthy speakers. Finally, the theoretical implications of a relationship between coarticulatory patterns in dysarthria and the size of the articulatory-acoustic working space will be considered..
For a printable version of this file click here
Wednesday, October 3, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Harry Heft, Ph.D.
Department of Psychology
Denison University
From "Thing and Medium" to Ecological Psychology:
A Tale of Two Research Programs
In the middle decades of the 20th century, two psychologists -- James Gibson working in perception, Roger Barker working in social development - separately proposed an ecological psychology that was a radical break from standard approaches in their respective areas. Although each employed the label 'ecological psychology' to describe their contributions, their proposed programs were distinctly dissimilar from each other, both with respect to their problem focus and also with regard to their level of analysis. In keeping with these notable differences, theory and research in each program proceeded independently. And yet, in spite of their notable differences, each program embraced as one of its foundational ideas Fritz Heider's (1926) highly original analysis of 'thing and medium.' This presentation will identify some of the distinctive and significant contributions of each ecological program, examine their common ties to Heider's seminal work, and offer an integrated view of an ecological psychology that functions at both the level of individual-environment interaction and collective social processes.
For a printable version of this file click here
Wednesday, October 10, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Robert Van Gulick, Ph.D.
Department of Philosophy
Syracuse University
"Maps, Gaps, and Traps: Metaphors for Understanding Consciousness"
The metaphors we use to talk about the problem of consciousness can both illuminate and restrict our understanding. They can reveal otherwise hidden aspects, or blind us to things outside their perspective. The much invoked metaphor of the "explanatory gap" - first coined by Joe Levine - provides a good case study (others might include Chalmer's Hard Problem/ Easy Problem distinction, or Nagel's equation of being a conscious x with "there being something that it's like to be an x".) The gap metaphor is both powerful but ambiguous in ways that can lead us astray if we do not can take care. I will explore the metaphor and its many meanings, in hope to dispel confusion and foster greater understanding of the mind/matter basis of consciousness.
For a printable version of this file click here
Wednesday, October 17, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Tamar Gendler Ph.D.
Department of Philosophy
Syracuse University
"Philosophical Implications of Research
on Childhood Pretense"
Recent empirical research concerning children's games of pretense seems to show that they are marked by the presence of two central features, which I call quarantining' and 'fertility'. Quarantining is manifest to the extent that causes within the pretense-episode are taken to have effects only within the pretense-episode (so, for example, the child does not expect the table really to be wet if the child "spills the tea"). Fertility is manifest to the extent that features of the imaginary situation that have not been explicitly stipulated are derivable via features of their real-world analogues (so, for example, the child does expect the table to be wet in the pretence if she up-ends the teapot above its surface). At the same time, from the same early age, it seems that both quarantining and fertility are constrained in crucial ways. Quarantining gives way to 'contagion' in cases of affect-laden imagination (so, for example, a child who imagines a bear on the staircase may be reluctant to go upstairs alone). And fertility gives way to 'unproductivity' as a result of the fundamental incompleteness of the imaginary (so, for example, there may be no fact of the matter (in the pretence) just how much "tea" there is left in the teapot). Exactly when and how these constraints relate to the principles to which they are exceptions raises complicated and interesting questions.
In my talk, I will present these distinctions in light of recent empirical research, and discuss some of their implications for a number of issues of current philosophical discussion in epistemology, aesthetics and the philosophy of mind.
For a printable version of this file click here
Wednesday, October 24, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Suzanne MacDonald, Ph.D.
Department of Psychology
York University, Canada
"Studying Memory and Cognition
in Zoo Primates:
Rewards and Challenges"
Exploring memory and cognition in nonhuman primates has traditionally been done in laboratory situations. While labs offer great experimental control, the range of species that can be studied is limited. Zoo animals offer a unique opportunity to examine cognitive processes in a wide variety of species.
In this talk, I will focus on my research with prosimians, New and Old World monkeys, and Great Apes. I'll discuss my findings on spatial memory and foraging strategies, as well as my current work on abstract concept discrimination, communication, and social cognition.
For a printable version of this file click here
Wednesday, October 31, 2001
2:00 pm - 3:30 pm
280 Park Hall
North Campus
K. Nicholas Leibovic, Ph.D.
Department of Biophysical Sciences, UB
"Brain and Vision"
For a printable version of this file click here
CANCELLED! Watch for news on a new speaker!
Wednesday, November 7, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Andre Kukla, Ph.D.
Department of Psychology and Philosophy
University of Toronto, Scarborough, Canada
"Epistemic Boundedness"
A psychological theory represents the mind as epistemically bounded if it is a consequence of the theory that our cognitive organization imposes epistemically significant constraints on the beliefs that we can entertain. Three arguments for epistemic boundedness are evaluated. Two of them-Colin McGinn's argument and the argument from mediocrity-are found to be defective. The third-Jerry Fodor's-underwrite no more than the relatively weak conclusion that we are dialectically justified in presuming that the mind is epistemically bounded when conversing with our cognitive-scientific colleagues.
For a printable version of this file click here
Wednesday, November 14, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Keith Oatley, Ph.D.
Department of Psychology
University of Toronto, Canada
"Emotions and the
Psychology of Fiction"
Fiction is typically highly structured. The goal of its writers can be thought of as offering materials to a reader (or audience member) so that they can construct and run cognitive simulations. Though such simulations run on minds rather than on computers, the analogy is close. Like computer simulations, fictions have aspects that correspond to (a) real-world models and (b) instructions as to how to compute over these models. In literary theory these aspects have been called fabula (the story world, or event structure) and siujhet (plot, or discourse structure). I will discuss two further aspects. One is the suggestion structure, which is based on priming and other such devices; it sets up resonances with the reader and prompts her or his own emotions and memories. The other is the realization: the enactment, or inner performance, of the fictional piece in the mind of reader as she or he runs the simulation in a way that�if the fiction is successful�involves the emotions. I shall present empirical evidence for some of these claims. This evidence includes demonstrations that people do indeed experience emotions when they read short stories, and that these emotions shape their understandings of, and reasoning about, the stories.
For a printable version of this file click here
Wednesday, November 28, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Shaun Gallagher, Ph.D.
Department of of Philosophy
and Cognitive Science
Canisius College
"Expressive Movement in a Deafferented Subject"
Explanations of neonate imitation of facial gestures have been framed in terms of motor ability and intermodal perception (Meltzoff and Moore, 1977; 1993; 1995). Meltzoff and his colleagues have offered explanations that rely on concepts like an innate body schema, perception and action coupling, and the reproduction of movement based on matching proprioception to visual stimulus (Chaminade , Decety, and Meltzoff, in press; Decety, et al., 2001; Gallagher and Meltzoff, 1996). I want to ask whether imitation of facial gestures is given a full account in these terms, or whether there might be some other mechanism that needs to be considered. To provide a framework for this question, I examine a case where gesture is clearly dissociated from certain aspects of normal motor ability, that is, where gesture, as a form of expressive movement, is irreducible to instrumental or locomotive movement. The case is that of IW, a man who lives without the sense of touch and proprioception below the neck. IW has profound problems with both locomotive and instrumental movement. Without proprioception he is not capable of controlling his movement without conscious use of vision and cognitive effort. When he wants to pick up a glass from the table, for example, he must think through his movement, consciously calculating distance, trajectory, grip, pressure, etc. Despite these problems with movement, IW, with and without vision, is capable of conversational gestures that are in most regards normal. I will report on experiments that show in precise terms that this is the case. I will also offer a theoretical account (in contrast to motor theories of gesture) to explain why gestures are not reducible to instrumental or locomotive movement. If gesture is a form of expressive movement that is not reducible to instrumental or locomotive movement, and if imitation of gestures is also a form of expressive movement, then I want to suggest that neonate imitation of facial gesture is not fully accounted for by innate body schemas, or other terms that focus on movement alone.
For a printable version of this file click here
Wednesday, December 5, 2001
2:00 pm - 4:00 pm
280 Park Hall
North Campus
Allison Sekuler, Ph.D.
Department of of Psychology, McMaster University
Canada
"Visual Completion: A Case Study in Grouping and Perceptual Organization"
One of the most important goals of vision is to recognize objects so that we can interact appropriately with them. Our everyday experiences suggest that the visual system is finely tuned to achieve this goal: recognition seems to occur instantly and effortlessly. However, research in my lab and others reveals that perceptual organization and recognition are much more complex processes than our phenomenology would lead us to believe. This lecture focuses on one aspect of my research: The perception of partly occluded objects. Because the information reaching our eyes is often incomplete, occlusion represents a huge obstacle to our perception of the world. I will describe evidence that the visual system completes contours and makes use of those completed contours. I will also discuss the effects of spatio-temporal context on visual completion, and I will describe our approach to combining behavioural methods with neuroimaging. Our results suggest that completion acts as a grouping mechanism enabling observers to use the relevant parts of the stimulus more efficiently for shape discrimination and object recognition. In this sense, visual completion is conceptualized not as end in and of itself, but as means to an end.