Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, 2003. / Includes bibliographical references. / The cognitive and neural processes underlying memory formation may differ depending on the vividness, or detail, of information encoded. In Chapter 1, a divided attention paradigm was used to examine memory formation when resources are (a) devoted primarily to encoding and (b) directed away from encoding, and toward a secondary task. In condition (b) the memories formed often lacked vividness. The formation of these less detailed memories recruited right inferior prefrontal cortex (PFC) and left parahippocampal gyrus. The left inferior PFC and left anterior hippocampus were additionally recruited in condition (a) when vivid memories could be formed. Investigations of memories' vividness have typically included only neutral information. The studies in Chapter 2 revealed that emotional information is vividly remembered more frequently than information lacking emotional import. This enhancement occurred for words with valence only (i.e., negative words that did not elicit physiological arousal) as well as for arousing ("taboo") words, but was stronger for the arousing words. In Chapter 3 a divided attention paradigm was employed to investigate the contributions of automatic and controlled processing to the recollective enhancement for the emotional words. Automatic processes (unaffected by task manipulation) drove the enhancement for arousing words, whereas controlled processes (disrupted by task manipulation) supported the enhancement for words with valence only. Thus, dissociable cognitive processes contributed to the enhancement for the two types of emotional words. In Chapter 4, fMRI was used to examine whether distinct encoding processes underlie enhanced memory for words with valence only versus words with arousal. / (cont.) Successful encoding of words with valence only was via a PFC-hippocampal network associated with controlled encoding processes (e.g., elaboration and rehearsal), whereas successful encoding of arousing words was mediated by an amygdalar-hippocampal network that may be important for automatic processing of emotional content. In conclusion, distinct neural processes appear to support the ability to form vivid memories as compared to less detailed ones. The specific cognitive and neural processes depended on the emotional nature of the stimuli. Vividly-remembered neutral words, and words with valence only, relied on similar encoding processes. In contrast, dissociable processes mediated successful encoding of vividly-remembered arousing words. / by Elizabeth A. Kensinger. / Ph.D.

Thesis: S.M. in Neuroscience, Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (page 16). / The ventral pre-arcuate area (VPA) in the primate prefrontal cortex has recently been found to play an important role in feature-based selection of visual targets in the context of a naturalistic free-gaze visual search task. While VPA neuronal activation was found to be necessary for behavioral performance as well as target selection in the FEF, its role in the broader context of the visual system remains to be addressed. To this end, we have interrogated the role of the VPA in mediating the effects of feature attention in the macaque visual area V4 by recording in V4 with and without muscimol inactivation in the VPA. We report here that neuronal activation in the VPA is necessary for firing rate increases related to target selection in V4. KEYWORDS: feature attention, visual search, muscimol, neurophysiology / by Azriel Ghadooshahy. / S.M. in Neuroscience

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, 2009. / "September 2009." Cataloged from PDF version of thesis. / Includes bibliographical references (p. 191-201). / What do words and phrases mean? How do we infer their meaning in a given context? How do we know which sets of words have sensible meanings when combined, as opposed to being nonsense? As language learners and speakers, we can solve these problems starting at a young age, but as scientists, our understanding of these processes is limited. This thesis seeks to address these questions using a computational approach. Bayesian modeling provides a method of combining categories and logical constraints with probabilistic inference, yielding word and phrase meanings that involve graded category memberships and are governed by probabilistically inferred structures. The Bayesian approach also allows an investigation to separately identify the prior beliefs a language user brings to a particular situation involving meaning-based inference (e.g., learning a word meaning or identifying which objects an adjective applies to within a given context), and to identify what the language user can infer from the context. This approach therefore provides the foundation also for investigations of how different prior beliefs affect what a language user infers in a given situation, and how prior beliefs can develop over time. Using a computational approach, I address the following questions: (1) How do people generalize about a word's meaning from limited evidence? (2) How do people understand and use phrases, particularly when some of the words in those phrases depend on context for interpretation? (3) How do people know and learn which combinations of predicates and noun phrases can sensibly be combined and which are nonsensical? / (cont.) I show how each of these topics involves the probabilistic induction of categories, and I examine the constraints on inference in each domain. I also explore which of these constraints may themselves be learned. / by Lauren A. Schmidt. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, September 2011. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / "September, 2011." Cataloged from student submitted PDF version of thesis. / Includes bibliographical references. / This dissertation documents how healthy aging and Parkinson's disease (PD) affect brain anatomy and physiology and how these neural changes relate to measures of cognition and perception. While healthy aging and PD are both accompanied by a wide-range of cognitive impairments, the neural underpinnings of cognitive decline in each is likely mediated by deterioration of different systems. The four chapters of this dissertation address specific aspects of how healthy aging and PD affect the neural circuits that support sensory processes and high-level cognition. The experiments in Chapters 2 and 3 examine the effects of healthy aging on the integrity of neural circuits that modulate cognitive control processes. In Chapter 2, we test the hypothesis that the patterns of age-related change differ between white matter and gray matter regions, and that changes in the integrity of anterior regions correlate most strongly with performance on cognitive control tasks. In Chapter 3, we build upon the structural findings by examining the hypothesis that age-related changes in white matter integrity are associated with disrupted oscillatory dynamics observed during a visual search task. Chapter 4 investigates healthy age-related changes in somatosensory mu rhythms and evoked responses and uses a computational model of primary somatosensory cortex to predict the underlying cellular and neurophysiolgical bases of these alterations. In contrast to the widespread cortical changes seen in healthy OA, the cardinal motor symptoms of PD are largely explained by degeneration of the dopaminergic substantia nigra, pars compacta (SNc). Cognitive sequelae of PD, however, likely result from disruptions in multiple neurotransmitter systems, including nondopaminergic nuclei, but research on these aspects of the disease has been hindered by a lack of sensitive MRI biomarkers for the affected structures. Chapter 5 presents new multispectral MRI tools that visualize the SNc and the cholinergic basal forebrain (BF). We applied these methods to test the hypothesis that degenerative processes in PD affect the SNc before the BF. This experiment lays important groundwork for future studies that will examine the relative contribution of the SNc and BF to cognitive impairments in PD. / by David A. Ziegler. / Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, 2015. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 112-126). / Human language is notable for its expressivity; syntax is powerful and allows for potentially unlimited new sentences. But even simple transitive sentences like "I broke the lamp" provide a sophisticated tool for communication, capture the basic building blocks of syntax and semantics that are widely agreed to be part of our linguistic capacity like agent or subject. With this relatively simple machinery, we are able to move a cognitive representation of an event from one person's head to another. How is this possible? In this dissertation, I examine both adult and child language to understand this capacity. Paper 1 examines the link between non-linguistic cognition and preschoolers' expectations about the meaning of novel verbs. We find that even though transitive verbs can refer to many event types, 3- and 4-year-olds are more likely to associate them with scenes with spatiotemporal features indicating causation. Papers 2 and 3 ask a second question: how do people organize language to facilitate communication? Paper 2 probes how adults order the basic elements (Subject, Verb, Object) in a task that appears to be independent of native language constraints, and tests whether the content of the message leads gesturers to reorganize their utterances. Paper 3 asks whether adults and children are aware that the relative informativity of arguments depends on context, and whether they can successfully make decisions in a novel communication task. By limiting the expression of transitive sentences to just two words (e.g. MONKEY EAT), we discover which elements people consider to be most informative. Both adults and children flexibly adjust their expectations about informative sentences according to which arguments are the most ambiguous in context. Together, these case studies help us understand how human language accomplishes its communicative goals, both in terms of the cognitive representations recruited for processing complex events in language, and the strategies used for expressing them. Whatever the formal nature of the representations involved in syntax and semantics, they must ultimately allow us to form predicates over nonlinguistic representations of the world, and they must support the kinds of pragmatic inferences that we know people can make. / by Melissa Kline. / Ph. D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, 2016. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 201-215). / The ability to differentiate between rewarding and threatening stimuli, and engaging the appropriate behavioral response is critical for survival. The Basolateral Amygdala (BLA) is an almond shaped structure in the brain where rewarding and fearful associations are encoded by different populations of neurons. However, identifying features of these populations have remained an enigma. My thesis work shows that populations of BLA neurons that differ in their long range anatomical connectivity play opposing roles in the acquisition of positive and negative associations, and dissects a mechanism by which these associations are encoded in the BLA. We show that BLA neurons projecting to the nucleus accumbens (NAc) and BLA neurons projecting to the centromedial nucleus of the amygdala (CeM) undergo opposing changes at their input synapses after rewarding and fearful associations. We then establish the in vivo ramifications of these opposing changes in synaptic strength in response to rewarding and fearful associations by assaying neural activity from BLA neurons and identifying the NAc and CeM projectors. Finally, in order to compare and contrast the role of BLA neural populations in encoding positive and negative associations, we propose a model that parametrizes neural responses to positive and negative cues from large scale electrophysiological recordings. My thesis work identifies functional roles of specific circuit components based on their long range anatomical connectivity, identifies differentially expressed receptors within these circuit components and provides a mechanistic explanation, on synaptic, cellular, circuit and molecular levels, for how positive and negative associations can be formed within, and diverge from the BLA. / by Praneeth Namburi. / Ph. D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 120-151). / This thesis aims to address a central question in cognitive science: how we reason about our own and others' cognition. Representing the self and others as distinct individuals is a fundamental epistemological feature of being human; the richness of these representations underlies our ability to tackle our own objectives and to understand the goals of others. Yet there is much debate about the metacognitive abilities of young children, in particular the extent to which children's estimations of their own and others' knowledge are accurate, whether children's beliefs about their own and others' cognition are influenced by the evidence they observe, and if these beliefs inform effective self-directed learning. I investigate these questions, examining metacognition and its relationship to learning in 3- to 8-year-olds. Chapter 1 provides an overview of metacognition regarding the self and others. Chapter 2 considers whether young children expect others will learn rationally from evidence. We find that by age 4.5 years, children have a nuanced understanding of how evidence and prior beliefs interact to yield new knowledge. Chapter 3 investigates how children's exploration is influenced by representations of task difficulty, as indexed by the discriminability of alternative hypotheses. We show that there is a precise quantitative relationship between uncertainty and information seeking. Chapter 4 considers how preschoolers use social comparison information to calibrate their self-directed learning, demonstrating that when a task is within children's zone of proximal development, observing evidence that peers perform better increases one's own persistence. Chapter 5 asks how 3- to 5-year-olds integrate representations of their own and others' abilities when allocating roles across contexts. This work demonstrates that children consider who is best suited for a task based on relative ability. Across all four chapters, the results of these studies demonstrate that children have a sophisticated understanding of their own and others' knowledge and skills. In addition, children use information about others to effectively direct their own learning and problem solving. I end by arguing that young children have a theory of individuals' characteristics, of which reasoning about the self is a special case. Taken together, these studies illustrate the importance of considering how reasoning about the self and about others are integrated and are fundamental to our human intelligence. / "Sources of funding, the BCS Halis Fellowship, the NSF Graduate Research Fellowship, the Center for Brains, Minds, and Machines (CBMM), funded by NSF STC award CCF- 1231216, and NSF Career Award (#0744213) awarded to Laura Schulz"--Page vi / by Rachel W. Magid. / Ph. D.

Thesis (Ph. D. in Cognitive Science)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 75-78). / In this thesis, I investigate how scenes are represented by the human visual system and how observers use visual information to reorient themselves within a space. Scenes, like objects, are three-dimensional spaces that are experienced through twodimensional views and must be recognized from many different angles. Just as people show a preference for canonical views of objects, which best show the object's surfaces and shape, people also show a preference for canonical views of scenes, which show as much of the surrounding scene layout as possible. Unlike objects, scenes are spaces which envelope the observer and thus a large portion of scene processing must take place in peripheral vision. People are able to perform many scene perception tasks, such as determining whether a scene contains an animal, quickly and easily in peripheral vision. This is somewhat surprising because many perceptual tasks with simpler stimuli, such as spotting a randomly-rotated T among randomly-rotated Ls, are not easily performed in the periphery and seem to require focal attention. However, a statistical summary model of peripheral vision, which assumes that the visual system sees a crowded, texture-like representation of the world in the periphery, predicts human performance on scene perception tasks, as well as predicting performance on peripheral tasks with letter stimuli. This peripheral visual representation of a scene may actually be critical for an observer to understand the spatial geometry of their environment. People's ability to reorient by the shape of an environment is impaired when they explore the space with central vision alone, but not when they explore the space with only peripheral vision. This result suggests that peripheral vision is well-designed for navigation: the representation in peripheral vision is compressed, but this compression preserves the scene layout information that is needed for understanding the three-dimensional geometry of a space. / by Krista Anne Ehinger. / Ph.D.in Cognitive Science

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, 1998. / Includes bibliographical references (leaves 195-204). / Estimating motion in scenes containing multiple moving objects remains a difficult problem in computer vision yet is solved effortlessly by humans. In this thesis we present a computational investigation of this astonishing performance in human vision. The method we use throughout is to formulate a small number of assumptions and see the extent to which the optimal interpretation given these assumptions corresponds to the human percept. For scenes containing a single motion we show that a wide range of previously published results are predicted by a Bayesian model that finds the most probable velocity field assuming that (1) images may be noisy and (2) velocity fields are likely to be slow and smooth. The predictions agree qualitatively, and are often in remarkable agreement quantitatively. For scenes containing multiple motions we introduce the notion of "smoothness in layers". The scene is assumed to be composed of a small number of surfaces or layers, and the motion of each layer is assumed to be slow and smooth. We again formalize these assumptions in a Bayesian framework and use the statistical technique of mixture estimation to find the predicted a surprisingly wide range of previously published results that are predicted with these simple assumptions. We discuss the shortcomings of these assumptions and show how additional assumptions can be incorporated into the same framework. Taken together, the first two parts of the thesis suggest that a seemingly complex set of illusions in human motion perception may arise from a single computational strategy that is optimal under reasonable assumptions. / (cont.) The third part of the thesis presents a computer vision algorithm that is based on the same assumptions. We compare the approach to recent developments in motion segmentation and illustrate its performance on real and synthetic image sequences. / by Yair Weiss. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Brain and Cognitive Sciences, 2003. / Includes bibliographical references (p. 311-318). / This thesis examines article choice and parameter-setting in second language (L2) acquisition. It argues, on the basis of L2-English elicitation and production data, that L2- learners have access to UG-based semantic distinctions governing article choice, but do not know which distinction is appropriate for English. A Fluctuation Hypothesis (FH) is proposed, according to which L2-learners fluctuate between different parameter settings until the input leads them to set the parameter to the target value. The thesis proposes that articles cross-linguistically may encode definiteness or specificity. The definition of specificity that is adopted is based on Fodor and Sag's (1982) view of specificity as speaker intent to refer. The behavior of referential this, a specificity marker in colloquial English, is examined, and it is proposed that the definition of specificity incorporates the concept of noteworthy property. An Article Choice Parameter is next proposed, which governs whether articles in a given language are distinguished on the basis of definiteness or on the basis of specificity. While English has the Definiteness setting of this parameter, it is suggested, on the basis of data from Mosel and Hovdhaugen (1992), that Samoan has the Specificity setting. It is hypothesized, in accordance with the FH, that L2-learners fluctuate between the two settings of the Article Choice Parameter. This hypothesis leads to the prediction that L2- English errors of article use should come in two types: overuse of the with specific indefinites and overuse of a with non-specific definites. These predictions are examined in a series of studies with adult speakers of Russian and Korean, two languages with no / (cont.) articles. The empirical data confirm the predictions, and show that L2-English article choice is not random but reflects access to the two settings of the Article Choice Parameter. The same patterns of results are found for L-Russian and L-Korean speakers, and it is shown that the results are not attributable to LI-transfer. On the basis of these findings, it is concluded that L2-learners have direct UG-access to semantic distinctions underlying article choice. The data also provide evidence for the existence of a specificity distinction which cross-cuts the definiteness distinction. / bu Tania Ruth Ionin. / Ph.D.