CNBC Brain Bag
Title: Ethics Seminar
Presenter: Dr. Strick
Location: MI 3rd Floor Social Room
Additional Information: Please RSVP to Joost Wagenaar (jbw14@pitt.edu) by Friday, March 14th, and indicate if you require a vegetarian meal.
Abstract: 

Cocaine abusers demonstrate faulty decision making as manifested by compulsive patterns of drug consumption despite negative social and physiological consequences. The orbital frontal cortex (OFC), which contributes to a variety of behavioral states and functions, including the processing of reward, emotion and decision making (Bechara et al., 2000; London et al., 2000), is thought to be affected by chronic drug use. Evidence supporting this hypothesis comes from clinical studies of drug users that show cognitive deficits similar to patients with brain injury to the OFC (Bechara et al., 2001). Though it seems clear that the OFC is involved in cognitive dysfunction in addicts, it is still unknown whether or not preexisting traits might also contribute to the cognitive function seen in addicts following cocaine exposure. It is important to identify and clarify the neural substrates that might be dysfunctional in cocaine addicts, resulting in cognitive deficits before and after cocaine exposure. The aim of the current study is to delineate the role of preexisting traits versus neurobiological consequences of cocaine exposure as the primary cause of cognitive dysfunction in chronic cocaine self-administering rhesus macaque monkeys. More specifically, we are probing OFC dysfunction by employing a visual 3 object reversal learning task to monkeys prior to and after chronic cocaine self-administration (SA). The reversal learning task allows for an examination of stimulus reward learning (acquisition of a novel object discrimination) as well as a conditioned reinforcement and inhibitory modulation of reward related behavior (reversal learning).  Because it is difficult to know the preexisting traits in humans prior to cocaine exposure, we use a rhesus monkey model that allows us to assess cognition before and after chronic cocaine SA. We hypothesize that chronic cocaine SA results in a regionally specific impairment associated with the OFC in monkeys.

 CNBC Brain Bag
Title: Spectral Contrast Effects on Lexical Tone Perception
Presenter: Jingyuan Huang

Location: MI 3rd Floor Social Room
Additional Information: Please RSVP to Joost Wagenaar (jbw14@pitt.edu) by Friday, March 20, and indicate if you require a vegetarian meal.
Abstract:
 Within tone languages that use pitch variations to contrast meaning, large variability exists in the pitches produced by different speakers.
 Context-dependent perception may help to resolve this perceptual challenge.  However, whether speakers rely on context in contour tone perception is
 unclear; previous studies have produced inconsistent results. The present study aimed to provide an unambiguous test of the effect of context on
 contour lexical tone perception and to explore its underlying mechanisms.  In three experiments, Mandarin listeners¹ perception of Mandarin first and
 second (high-level and mid-rising) tones was investigated with preceding speech and non-speech contexts. Results indicate that (1) the mean
 fundamental frequency (f0) of a preceding sentence affects perception of contour lexical tones and the effect is spectrally contrastive. Following
 a sentence with a higher-frequency mean f0, a following word is more likely to be perceived as a lower frequency lexical tone and vice versa; (2)
 non-speech precursors modeling the mean spectrum of f0 also elicit this effect, suggesting general perceptual rather than articulatory-based or
 speaker-identity-driven mechanisms.

CNBC Alumni Lecture
Title: The Anterior Temporal-Lobes and Semantic Memory: Putting Everything Together
Presenter: Tinothy Rogers
Location: WPIC Auditorium
Abstract: Essentially all current theories about the neural basis of semantic knowledge agree that much of the content of our semantic memory is represented in brain regions that overlap with, or even correspond to, the regions responsible for perceiving and acting. The more global neuroanatomical organization of the semantic system remains,however, something of a mystery. I have argued that anterior temporal lobe (ATL) regions play a critical role in mapping between different sensory, motor, and linguistic representations distributed widely in cortex. Consequently these regions come to serve as an amodal and domain-general =93semantic hub=94 that is important for representing all kinds of concepts and for supporting performance on semantic tasks irrespective of the particular modality of testing. This view originated with studies of patients with semantic dementia (SD), a disorder in which progressive impairment to semantic knowledge across all modalities of reception and expression is accompanied by progressive gray-matter loss and hypometabolism localized within anterior temporal regions. The theory is challenged, however, by findings from functional neuroimaging studies of healthy individuals, which routinely report semantic activation in posterior temporal, temporo-parietal, and prefrontal regions, but only rarely in ATL regions. It is also challenged by behavioral studies of patients with other forms of pathology in ATL regions, who rarely show the same profound patterns of semantic impairment observed in SD. In this talk I will try to reconcile these apparent contradictions with reference to new data from patient studies, functional neuroimaging, and computer modeling.

CNBC Friday Seminar
Title: Enhancing Peer Review at NIH
Presenter: Cheryl A. Kitt, PhD
Location: MI 115
Abstract:

·NIH and peer review at CSR
·The drivers for change in peer review
·CSR’s efforts to enhance peer review
·The NIH Director’s enhancing peer review initiatives
·Peer review’s strategic, national value 

CNBC Seminar
Title: Uncovering the Fundamental Principles of Visual Cortex
Presenter: Michael Tarr, PhD
Location: Mellon Institute Third Floor Social Room, Bellefield Street Entrance
Abstract: Visual object recognition is one of the most poorly understood mental faculties. Twenty years of research has led me to the conclusion that theories of object recognition are underspecified with respect to both the functional roles ascribed to different neural structures in the inferior-temporal (IT) cortex and the range of visual recognition behaviors exhibited by humans. The inadequacies of current theories stem from a theoretical status quo combined with methodological limitations inherent in both psychophysics and neurophysiology. To better understand object recognition we must discard both standard feed-forward, hierarchical models that bear little resemblance to the facts as we know them and the behavioral studies that test such models. We must also develop new neuroimaging methods to complement neurophysiology, which under-samples object representation space and typically relies on ad hoc/atheoretic strategies for determining which features/objects yield maximal neural responses.

I will motivate these arguments with examples drawn from my earlier behavioral and neuroimaging work. I will also discuss our development of new tools for mapping feature and object selectivity across human visual cortex using functional Magnetic Resonance Imaging (fMRI). The effort is motivated by neurophysiological studies with similar objectives. However, the informativeness and generality of visual physiology has been limited by the low number of samples (~10^3 recordings) relative to the size of the neural representational space (~10^9 neurons). fMRI, which measures brain responses in voxels (~10^6 neurons), enables the study of neural codes at a macro level, yet at a resolution fine enough to capture meaningful functional differences between brain regions. To explore the feature selectivity of localized regions of IT, visual stimulation will be driven by real-time fMRI, in which accruing neural contrasts between conditions are computed instantaneously. This mapping approach will be enhanced by employing at least two principled strategies for moving through feature space: an a priori method that relies on an algorithm for automatically segmenting objects into features (which has been validated against human segmentations); and, an a posteriori method that relies on “mutual information” to identify features that carry more or less task-relevant information.

VASC Seminar Series
Title: Material Recognition By Humans and Machines
Presenter: Lavanya Sharan
Location: Newell Simon Hall 3305
Abstract:
We can easily tell if a spoon is made of stainless steel or
plastic, if a shirt is clean or if food is fresh. These judgments of
material appearance are ubiquitous. We use our material perception abilities
to decide where to step on an icy sidewalk, which items to pick in the fresh
produce aisle, and if a rash requires a trip to the doctor. In spite of the
importance of these judgments, little is known about material recognition in
the fields of human vision or computer vision.

We have studied human material judgments on real world photographs by asking
observers questions like "Is that object made of paper or plastic?" or "Are
those flowers fake or real?". We find that observers can recognize materials
very well, even when images are presented very fast (40 millisecond/image).
This performance was robust to low-level image degradations like removal of
color, blurring and inversion of contrast polarity, suggesting that
low-level information is not crucial for observers.

What do these results imply for machine vision systems? We evaluated the
performance of simple classifiers based on low-level image features (e.g.
jet-like features, SIFT) at the same material categorization task that
humans did. We find that low-level features are not sufficient for
categorizing materials on our data set suggesting a parallel with the
results from human experiments. We conclude that there is rich territory to
be explored both by computer vision and human vision researchers for this
problem.
 

Pitt Departments of Statistics and Electrical Engineering Seminar
Title: Stabilized trans-cerebrocerebellar "long-loop" responses enable basic motor control without internal models of body dynamics
Presenter: Steve G. Massaquoi, MD
Location: 6014 BST3
Abstract: A computational model of cerebrocerebellar control demonstrates that many concerns regarding the limitations of feedback control are not compelling. The Recurrent Integrator Proportional-Integral-Derivative (RIPID) model proposes that recurrent corollary discharge from and to the cerebral cortex via the cerebellum enhances the efficacy of long-loop responses, thereby obviating internal models of body dynamics.

When applied to arm control, the model can be shown to reproduce basic features of postural stability and movement performance as well as to reproduce important features of internal signals recorded in the cerebral cortex and cerebellum of behaving primates.   Simulated lesions of the system reproduce important features of the clinical syndromes associated with such lesions.  It can then be shown that upright posture of the body can be stabilized against disturbances using a RIPID-based mechanism that switches between two sets of 3x3 matrices representing scheduled cerebellar control.  Finally, when coupled with simple pattern generators, the above mechanism can  generate realistic simulated walking.   This effective control occurs in the presence of neural transmission delays and phase lags present  in the nervous system.  The findings support the possibility that the central nervous system need not compute detailed models of body dynamics and that therefore central motor control mechanisms may be simpler than often suspected.