|
|
 |
||||||
 |
April 28th, 2010 Effects of Intensive Behavioral Training on Verbal and Spatial Working Memory Capacity Jeff Phillips I will present behavioral data from a working memory training paradigm which has been linked to improved performance on the Stroop task and a test of reading comprehension (Chein & Morrison, in press). Chein & Morrison's finding is one of several recent reports of learning with far transfer, raising the question of what distinguishes these training paradigms from a multitude of negative findings in the field. I will propose that recent behavioral training successes depend on the use of 1) a high-intensity training paradigm (analogous to interval-style physical training) and 2) adaptive task demands, which discourage the development of automaticity and performance "plateaus." Research Writing Feedback: Commenting Strategies and their Impact Melissa Patchan Three common sources of feedback are used in Writing Across the Curriculum and Writing in the Discipline courses: content instructors, writing instructors, or peers. There are benefits and disadvantages for each type of source. It is also unclear how the commenting styles differ between the sources, and what kind of impact these differences would have on students writing. In an introduction to cognitive science course, undergraduate students received feedback on their writing from either a cognitive psychology graduate student, a rhetoric graduate student, or a peer. All three sources used the same rubric, which included both content dimensions and writing dimensions. We analyzed the comments provided to the students to see how the commenting styles differed. More specifically, we were interested in whether the content 'experts' focused more on the substance of the paper and whether the writing 'experts' focused more on the prose of the paper. In addition, we wanted to know where peers fit in--were they more like content 'experts' or writing 'experts' or did they have their own unique style of commenting?
April 21st, 2010 Does cross-language similarity affect second language processing? Alba Tuninetti Cross-language similarity has become an increasingly important factor in second language processing models. This study aims to investigate English second language learners’ sensitivity to syntactic violations in a self-paced reading paradigm. Previously unstudied with this methodology, this investigation will be able to shed light on the validity of models of second language processing, such as the Unified Competition Model (UCM) (MacWhinney, 2005) and the Shallow Structure Hypothesis (SSH) (Clahsen & Felser, 2006). This will help us adopt a more exact model of second language processing that will aid tremendously in discovering how bilinguals and second language learners process and differentiate between languages. I will present my pilot data with monolingual participants and outline my expectations for second language learners. Ultimately, the proposed research will become my Master’s proposal. Joel Chan Innovative designs have often been linked to inspiring analogies. However, all analogies are not created equal: sometimes they can help designers come up with innovative designs, but variations in key properties can result in negative design outcomes (e.g., fixation). Understanding what makes analogies inspiring has important implications for understanding analogical reasoning and creative thinking generally, and also for developing tools that support innovation via analogy. I report current work that investigates the effects of variations in analogical distance, familiarity, and representation modality of given analogs on ideation processes in engineering design. In our experiments, engineering students generated solution concepts for an engineering design problem with or without provided analogies (analogy groups vs. control group). Analogies were fully crossed by distance (near vs. far), familiarity (familiar vs. unfamiliar), and modality (pictures vs. text). Formal coding and analyses are still in progress, but preliminary results suggest that familiarity of analogs influence the number of ideas generated, regardless of analogical distance or modality, with familiar examples suppressing idea output relative to no or unfamiliar examples. Possible interpretations and next steps are discussed.
April 14th, 2010 Eliciting the Error-Related Negativity (ERN) from a Speeded Judgment Spelling Task Lindsay Harris The ERP waveform component known as the error-related negativity (ERN) is a negative voltage shift commencing 50 to 100 milliseconds after an incorrect key press during a forced-choice judgment task. In the last 20 years the ERN has been studied by researchers interested in the neural substrates underlying everything from human judgment-making to psychobiological disorders such as schizophrenia. The majority of these studies have used a Flanker task to induce the ERN; rarely has the ERN been elicited by tasks requiring higher-level cognitive processes, such as language judgment tasks (for examples see Balass et al., in preparation; Horowitz-Kraus & Breznitz, 2008). I am preparing a study that will use as stimuli misspelled English words in an attempt to trigger the ERN; I will present pilot data and outline the upcoming stages of the project. Ultimately, I am interested in detecting a correlation between the ERN and individual differences in reading ability. Examining English-German Translation Ambiguity using Primed Translation Recognition Chelsea Eddington Second language learners may encounter difficulties when trying to translate words that have more than one translation in the second language. This so-called “translation ambiguity” can be due to meaning ambiguity, in which a word has two meanings that correspond to two distinct translations or it can be due to synonymy, in which a a word has one meaning but two or more translations can represent that meaning. Little research has examined how translation-ambiguous words are processed when presented in context. I present my plans for my master’s project, which will examine how context and frequency of meaning influence processing of translation ambiguous words. Native English speakers who are intermediate to advanced level German students will participate in a primed translation recognition task. During the task, participants will be presented with a related or unrelated prime then, a target-translation pair will be presented and they will be asked to decide if the word pairs are correct or incorrect translation equivalents. The targets will be comprised of translation unambiguous words, meaning, and synonym type ambiguous words. I will evaluate processing time through response times on the target translation pairs. Translating cognitive science into classroom instruction: The case of analogical comparison. J. Elizabeth Richey How can the principles of analogical comparison discovered in the laboratory be translated into classroom pedagogy? This talk will examine efforts to isolate the mechanisms underlying analogical learning gains and implement them in existing middle school science curricula. Part of a massive experiment led by the 21st Century Research and Development Center on Cognition and Science Education, this work suggests directions for future efforts to integrate cognitive science research and classroom instruction.
March 31st, 2010 Kai Hwang The human cerebral cortex is organized into distributed systems with different functional specializations. Cortical hubs are brain regions that display high connectivity by coordinating functions across distributed neural networks. Previous studies have identified brain regions in the frontal and parietal cortices as potential cortical hubs. However little is known about the development of cortical hubs and their connections to functional networks. We used resting state fMRI to map cortical hubs that display high intrinsic functional connectivity, and characterized the developmental pattern of frontal-parietal networks connected by cortical hubs from late childhood to early adulthood (ages 11 - 18). Results showed that hubs in the parietal cortex developed early in development, but hubs in the medial and lateral prefrontal cortex emerged later in mid-adolescence. The development of frontal-parietal intrinsic functional networks paralleled the development of frontal hubs, emerging later in mid-adolescence. These findings have implications regarding the known behavioral improvements in cognitive control functions during adolescence. Developmental changes in brain function underlying reward-based cognitive control Aarthi Padmanabhan Adolescence is a period characterized by a number of changes including increased risk taking and impulsivity that result from immaturities in motivational and cognitive systems. Research has suggested that cognitive control of behavior is modulated by rewards and that adolescents have a heightened sensitivity to rewards relative to adults. However, the full developmental trajectory of reward-based cognitive control is not well delineated. To assess age effects on neural circuitry underlying reward-based cognitive control, we used an event-related functional magnetic resonance imaging (fMRI) task that places monetary incentive contingencies on performance in the antisaccade (AS) task, an established task to index voluntary response inhibition. 30 healthy participants (10 children: 8-13yo,10 adolescents:14-17yo,10 adults:18-25yo), were scanned on a 3T scanner. Results suggest that core circuitry underlying response inhibition and reward processing is established in childhood. However, neural responses to reward contingencies in reward-related regions demonstrate a non-linear modulation with age, indicating that adolescents may be especially sensitive to and more influenced by rewards.
March 24th, 2010 Kyung Hwa Lee An individual may experience a similar emotion (i.e., concordant) or may experience a different emotion (i.e., discordant) to one presented in emotional stimuli. Despite this fact, most imaging studies have examined neural mechanisms of explicit emotional processing of self (i.e., one’s own emotion) and others (i.e., emotions of other people presented in stimuli) without distinguishing between concordant and discordant conditions. Relatively less is known about neural mechanisms underlying emotional processing of self and others in discordant conditions. Past research suggested that one’s own concordant and discordant emotions to other people were determined by characteristics of others (Hareli & Weiner, 2002; van Dijk et al., 2006). The present research attempted to manipulate people’s information (e.g., characteristics) to make participants to perceive people as either good (i.e., concordant) or bad (i.e., discordant), using scripts and video clips. Behavioral pilot study revealed that participants reported more negative emotional responses to the bad person’s video clip compared to the good person’s video clip. Such prior experience made participants to feel concordant emotions to the good person and discordant emotions to the bad person in negative events. Neuroimaging study has been conducted to elucidate neural mechanisms associated with emotional processing of self and others in both concordant and discordant conditions. Preliminary ROI analyses showed that some brain regions such as rostral anterior cingulate cortex (rACC) and insula were more specifically involved in evaluating one’s own emotion (self) and brain activity in the regions were modulated by the discordant condition. Plural Sets Can Saturate Reciprocal Thematic Roles Nikole Patson Patson and Ferreira (2009) showed that the standard garden path effect in sentences with structures like (1) (i.e., increased processing time on the disambiguating region, cried) is reduced or eliminated when the subject of the initial clause is a complex reference object and the initial verb is reciprocal (1a), as compared to minimally different sentences with either another kind of plural subject (e.g. a plural definite description like the lovers or the two lovers) or an optionally transitive verb (1b). They argued that this reduction of the garden path effect occurred because the two individuals in the complex reference object saturated the thematic roles of the reciprocal verb, leaving no unfilled expectation for an object/patient NP. The current study probed deeper into what kinds of plurals can saturate a reciprocal verb’s theta roles. Patson and Ferreira’s findings suggested that only plurals that have ‘visible’ individuals, or individuals accessible via pointers, can fill a reciprocal verb’s roles. But there remain many unanswered questions about what kinds of properties the individuals must have, and the readings that must be available for the reciprocal. For example, simply having two pointers to different sets might allow those sets to fill the open thematic roles. However, given that Patson and Ferreira only tested conjoined singular NPs, it could also be necessary that the two pointers each pick out individuals. Their evidence also does not rule out the possibility that the verb may need to be interpreted in a strongly reciprocal (Dalrymple, Kanazawa, Kim, McHombo, & Peters, 1998) way in order for its theta roles to be satisfied. The current self-paced reading study investigated whether complex reference objects made up of plural sets saturate the thematic roles of reciprocal verbs. The experiment had a 2x2 design, manipulating verb type (reciprocal (1a, 2a) vs. optionally transitive (1b, 2b) and subject type (conjoined singular NPs (1a,b) vs. conjoined plural NPs (2a,b)). Results indicated a main effect of verb type such that reading times on the disambiguating region were longer in the optionally transitive verb conditions (M= 552 ms) than in the reciprocal verb conditions (M= 482 ms, F1(1,47) = 4.78, MSe= 50430.18, p < .05; F2(1,27) = 5.72, MSe= 5888.82, p < .05). There was no interaction, indicating that conjoined plural noun phrases also block the garden path when presented with a reciprocal verb. These findings suggest that all that is required to saturate the thematic roles of reciprocal verbs is a subject that establishes two pointers to different sets. The fact that these sets were plural definite descriptions and did not force a strongly reciprocal interpretation suggests that even weakly reciprocal interpretations of a verb are enough to saturate the thematic roles of the verb, thus eliminating the expectation for an upcoming argument. 1a. While the man and the woman kissed the baby cried in the crib. 1b. While the man and the woman cleaned the baby cried in the crib. 2a. While the men and the women kissed the baby cried in the crib. 2b. While the men and the women cleaned the baby cried in the crib.
March 17th, 2010 Developing Robust Lexical Representations in English as a Second Language Susan Dunlap Learning to read and spell in English as a second language is a challenge for the adult learner. I will report results of an intervention designed to test two types of spelling instruction—a form-focus condition, and a form-meaning integration condition—and their effect on developing ESL students’ receptive and productive lexical representations (Perfetti & Hart, 2001). On the spelling recognition (i.e., lexical decision) task, students showed moderate improvement on all items regardless of training. On the spelling production (i.e., audio dictation) task, students showed remarkable improvement for items in both instructional conditions. After just two one-hour training sessions, students in each training condition gained an average of 14 percentage points on productive spelling of our target words (significantly above zero), compared with no statistically significant gain for control items. These gains are especially remarkable because the audio dictation task is more difficult than the lexical decision task. These findings suggest that even minimal, focused training on spelling can help ESL students improve their orthographic representations. This could mean that explicit, targeted instruction helps students improve on production tasks, but that incidental exposure is adequate for receptive tasks such as word recognition or lexical decision. Collaborative Facilitation through Error-Detection: A Classroom Experiment Soniya Gadgil Prior work has shown that individuals working in groups often perform worse than individuals working alone, a finding commonly referred to as collaborative inhibition. In the current work we examine whether engaging in error correction processes can mitigate or eliminate the collaborative inhibition effect and perhaps even facilitate collaborative facilitation. Participants engaged in a writing error-detection and revision task while working either with a partner or individually. On the error-detection task, dyads found more structural flaws in the text, whereas individuals found more surface flaws. Moreover, when comparing dyads nominal groups the dyads did not show the collaborative inhibition effect. A similar pattern of results was found on the revision task. The results are discussed in terms of the underlying cognitive and social processes that support successful collaboration.
March 3rd, 2010 Separated displays improve learning and efficiency Jooyoung Jang This classroom experiment explored whether a seemingly simple factors such as the physical organization of instructional material significantly influences the problem solving performance and overall learning in the domain of statistics. Participants learned to perform statistical analyses with printed instructions; half received instructions in a separated side-by-side format and half in a stacked format. They were asked to analyze data, and then took a test on statistical concepts and skills. Results showed that students given instructions in the separated format performed the data analysis task much more quickly and showed evidence of improved learning on the test. Second language morphosyntactic instruction and cross-language similarity Leida Tolentino Second language (L2) research and theory suggest that the level of explicitness of instruction and the nature of the L2 input affect attained adult proficiency. However, it is unclear to what extent instruction method interacts with similarity between the first language (L1) and L2. In this brown bag I will talk about my dissertation plans, which include an evaluation of the effectiveness of different methods of L2 instruction as a function of morphosyntactic similarity between L1 and L2. Following a pre-test, adult native English speakers will be trained on Swedish vocabulary and will be subsequently assigned to one of three instruction groups that differentially incorporate input salience and metalinguistic explanation of grammatical features varying in degree of L1-L2 similarity: a control group (non-salient, non-contrastive L2 input); a salience group (contrastive and color-highlighted input); and a rule & salience group (salient, contrastive input and metalinguistic explanation). Learning will be assessed through immediate and delayed post-tests using both behavioral (e.g., grammaticality judgments; translation) and event-related potentials (ERPs) measures. I will also present pilot data from a relevant summer internship project at Rosetta Stone, Inc.
February 24th, 2010 Neural signaling of the inferred predicted value of never-sampled stimuli Karin Cox In situations that require learning to choose between higher- and lower-valued stimuli (S+ and S-), neural representations of each option’s value are often modeled as arising from direct trial-by-trial sampling of the relevant contingencies. However, in a recent study of probabilistic reversal learning, Hampton et al. (2006) showed that reward-predictive activity in the ventromedial prefrontal cortex (VMPFC) was best explained by a model that incorporated the knowledge that the S+ and S- values were anticorrelated, and thereby was able to update the estimated values of both the stimuli that were and were not sampled on each task trial. Since reversal learning involves repeated experience of the two possible sets of contingencies, it is unclear whether the updating of VMPFC responses to the unsampled stimuli was due to (1) retrieval of the currently appropriate set of concrete stimulus-outcome relationships, or (2) an abstract reasoning process that would also be capable of applying the anticorrelation principle to novel stimuli. To test whether updating of VMPFC responses to unsampled stimuli can be driven by inferences derived via abstract rules, I have proposed a neuroimaging experiment based on a stimulus-replacement task originally designed by Lockhart et al. (1963). This task consists of a series of unique deterministic discrimination problems in which the stimulus that is chosen on each initial problem trial is replaced with a novel stimulus of the same reward status on the second trial. Since optimal second-trial performance requires an inference regarding the value of the initially-unsampled option, this task will permit the measurement of neural activity evoked by stimuli for which the contingencies have never been directly experienced, but instead must be deduced using abstract rules. Learning translation-ambiguous words Tamar Degani Relatively little is known about the role of ambiguity in adult second-language learning. In Experiment 1, native English speakers learned Dutch–English translation pairs that either mapped in a one-to-one fashion (unambiguous items) in that a Dutch word uniquely corresponded to one English word, or mapped in a one-to-many fashion (ambiguous items), with two Dutch translations corresponding to a single English word. These two Dutch translations could function as exact synonyms, corresponding to a single meaning, or could correspond to different meanings of an ambiguous English word (e.g., wisselgeld denotes the monetary meaning of the word change, and verandering denotes alteration). Several immediate and delayed tests revealed that such translation ambiguity creates a challenge for learners. Furthermore, words with multiple translations corresponding to the same meaning are more difficult to learn than words with multiple translations corresponding to multiple meanings, suggesting that a one-to-many mapping underlies this ambiguity disadvantage. Experiment 2 replicates these findings, and further examines how different teaching approaches influence this ambiguity disadvantage.
February 3rd, 2010 Elizabeth Votruba-Drzal Socioeconomic disparities in reading and math achievement are some of the most well documented findings in research across disciplines. Few longitudinal and nationally representative studies have tracked the course of these disparities during elementary school and the independent contributions of separate dimensions of SES are not well understood. Using data from the Early Childhood Longitudinal Study Kindergarten (ECLS-K) (N = 9,970), this study characterizes the size and persistence of achievement disparities in reading and math related to parental education, income and occupational prestige from kindergarten through fifth grade. Furthermore, it examines learning-related parenting practices as pathways through which SES influences the development of academic skills. Moderate disparities in average levels of reading and math skills related to parental education and income were found. From kindergarten to fifth grade, the magnitude of achievement disparities related to parental education and income doubled. Cognitive stimulation in the home environment was a significant mediator of socioeconomic disparities in average levels of reading and math achievement and of growth in math skills during elementary school. Policy implications of these findings are discussed.
January 27th, 2010 From synapse to BOLD: linking circuit plasticity to the macroscopic responses of sensorimotor cortex Timothy Verstynen Simple network models, such as line attractors, can perform statistically-principled computations, like maximum likelihood estimation, on noisy inputs. We and others have shown that applying Hebbian learning to these networks allows them to use recent input histories to approximate a form of adaptive Bayesian estimation. This learning process captures observed experience-dependent changes in movement statistics for both reaching and saccades. Here we use fMRI to begin to test this model of adaptive retuning in sensorimotor circuits. To this end, we have developed a novel model-based approach to studying experience-dependent changes in cortical activity, as reflected in the blood-oxygenation level dependent (BOLD) signal. Our approach assumes that changes in the BOLD signal are dominated by changes in local synaptic inputs. This allows us to quantitatively predict the responses of both individual voxels and entire regions of interest under different testing conditions. According to the model, repeated (or highly similar) inputs re-tune the lateral connections of local units, biasing future computations towards the repeated input position. This result in an increase of the simulated BOLD response to inputs presented around the repeated position. Indeed we find target-specific changes in BOLD that correlate across subjects with the magnitude of experience-dependent behavioral changes in a novel saccade task. Moreover, this effect was only observed in the frontal and parietal regions thought to mediate saccade planning. The general nature of this approach allows us to test the prediction of network level models in a variety of different behavioral contexts, making fMRI a useful tool for studying network dynamics beyond simple region identification and inter-region correlation dynamics.
January 13th, 2010 Elisabeth Ploran A previous study (Ploran et al., 2007) has shown that an extended recognition paradigm in which objects are slowly revealed can create accumulating BOLD signals prior to recognition. These effects were observed in select parietal, temporal, and frontal areas as recorded through functional magnetic resonance imaging (fMRI) in humans. However, the quantity of perceptual evidence throughout the trial increased in the original study, limiting interpretation of the data. The current study further examined the accumulation response during perceptual decision-making by maintaining a constant amount of perceptual evidence and sometimes altering the consistency of the evidence. Greyscale objects were presented underneath a checkerboard-like mask with randomly spaced transparencies that changed in sync with each new whole brain image. This produced high quality but consistent limited quantity evidence throughout the trial. In addition, on certain trials the object underneath the mask alternated between two similarly shaped, but not identical objects. This manipulation tested whether accumulation occurs due to attentional needs even without consistent evidence. Despite a consistent limited quantity of evidence, accumulation was still found in bilateral inferior frontal gyrus, bilateral fusiform, and left intraparietal sulcus. Furthermore, changing the picture underneath the mask as an alteration of available evidence decreased the overall magnitude of accumulation, but did not affect sensory or execution processes. These findings support the possibility that accumulating BOLD signals reflect signal integration over time.
December 9th, 2009 Kristine Wilckens It is well established that older adults tend to perform worse on tests of memory compared with younger adults. However, there is a lack of consensus regarding the mechanisms underlying these differences in memory performance. One hypothesis posits that the deficit is specific to encoding and retrieving of relationships between memory components (Dew, Bayen, & Giovanello, 2007). However, because age-related impairments are often related to difficulty accessing memories that require controlled processing, an alternate hypothesis posits that this age-related memory deficit may result from a more general decline in cognitive control function, which certain types of memory processes may rely upon. Behavioral evidence also suggests that older adults may tend to rely on a less efficient ‘late correction’ strategy rather than an ‘early selection’ strategy at retrieval (Jacoby et al 2005). Using a cued memory task in which subjects switch or repeat the same retrieval task trial by trial (Morcom & Rugg, 2001), I provide behavioral and ERP evidence to support this model of age-related differences in memory retrieval.
December 2nd, 2009 Transfer and Motivation: Cognitive Psychology Can Assess Achievement Goal Outcomes Daniel Belenky Achievement goals are the broad motivational reasons people have for engaging in achievement situations, like academics. They can be defined by one main contrast; a focus on developing skill ("mastery goals") or a focus on demonstrating skill ("performance goals"). These goals have been well studied by social and educational psychologists in college classrooms, and their effects on strategy use, affect, and grades have been mapped out. However, one counterintuitive finding has been shown consistently; while performance goals correlate with final grades, mastery goals do not. I will present an experiment that uses multiple types of assessment to show how different achievement goals and instructional activities influence the development of different kinds of knowledge. Implications for assessment in college classrooms will be discussed.
November 11th, 2009 Uh-oh! I Made a Mistake: ERP correlates of Error Detection in a Semantic Categorization Task Michal Balass The error-related negativity (ERN) is a negative deflection in the ERP record elicited 0 to 160ms post-response related to conscious error detection. The ERN is regarded as the brain’s “Uh-Oh!” signal when a conscious error has been made. Studies have shown evidence for the ERN in verbal perception tasks (Ganushchak & Schiller, 2008), and also indicated its usefulness for showing differences in error detection related to reading skill (Horowitz-Kraus & Breznitz, 2008). However, there are not many studies published to date showing an ERN response to higher level processing; for example, semantic processing. In this study, we used a time-pressured semantic categorization task in an attempt to elicit errors to simple category decisions (e.g., is Africa a country?). In this talk, I will present evidence showing greater ERN negativity at approximately 100ms post-response to incorrect responses than for correct responses. Further, I will discuss correlational analyses showing that ERN negativity may be related to individual differences measures of reading skill.
November 4th, 2009 Adolescence without the Fun: Examining Reward Function in Youth with Depression Erika E. Forbes In the search for neural substrates of depression, some recent work has suggested that the disruption of reward processing occurs as part of the essential pathophysiology of the disorder. Depression has long been conceptualized as a disorder of dysregulated positive affect and unusual reward function, and affective neuroscience findings have begun to support this perspective. Furthermore, taking a developmental perspective is critical because the onset of depression typically occurs during adolescence, and altered reward function during adolescence is a predictor of future depression. This presentation will describe neuroimaging, behavioral, and subjective studies on reward function in adolescent depression. These studies were conducted using monetary reward functional MRI paradigms, performance on reward-related decision paradigms, and experience sampling of positive affect. Together, the results of this set of studies indicate that adolescents with depression exhibit less striatal reactivity to reward, are less behaviorally sensitive to reward, and experience lower intensity of reward-related affect. Furthermore, preliminary findings suggest that reward-related brain function predicts response to treatment in adolescents with depression. The presentation will emphasize the importance of placing adolescent depression in a developmental context, examining brain-behavior associations, and using multimethod approaches to capturing affective processes.
October 28th, 2009 Examining Attention Allocation in Gaze-Contingent Visual Search Tasks Using Diffusion Models Polina Vanyukov In two eye-movement experiments (presented at the brown bag last fall) we examined whether attention is allocated serially or in parallel during three visual search tasks using 2-8 character arrays subtending two degrees of visual angle: (1) single-feature detection; (2) conjunctive-feature detection; and (3) ordered-feature detection. In the first experiment, participants’ eye movements were recorded; in the second, a gaze-contingent paradigm was used to evince covert attentional shifts. In this follow-up work, random-walk (diffusion) models of parallel and serial search (Thornton & Gilden, 2007) were fit to the data from the gaze-contingent study to determine which type of search best accounts for the patterns of reaction times and error rates in these tasks. The results from the computational modeling suggest that single-feature and conjunction-feature detection tasks are better described by a parallel model of visual search. Ordered-feature detection task, however, was well approximated by both serial and parallel (limited capacity) models of visual search. Implications for attention allocation during reading are discussed.
October 21st, 2009 Individual Differences in Patterns of Reading Jessica Nelson A large database of reading test scores for over 2000 participants was analyzed using a factor analysis to determine the main dimensions along which readers vary in skill. A smaller subset of these participants had their eye movements monitored while they read paragraphs collected mostly from online news stories or informational articles. Individual differences in patterns of eye movements during reading, including regression and fixation statistics, will be related to the dimensions of reading skill obtained from the factor analysis of reading scores.
October 14th, 2009 Looking for Failure and Success: Factors that Determine Dual-Tasking Performance Nicole M. Hill Although humans frequently engage in dual-tasking, generally performance is suboptimal. When component tasks are initially performed simultaneously, accuracy decreases relative to single-task competency producing a dual-task deficit. Single-task training does not readily transfer to dual-tasking; this the case even when component tasks are both automatically processed (Schneider & Hill, 2006). Although dual-task practice is a necessary it is not sufficient for operating without a deficit. Schneider and Fisk (1982) trained participants to perform a dual-task comprised of an attention demanding task and an automatic task. The dual-task deficit was attenuated after substantial practice but only when the attention demanding task was prioritized. Hill and Schneider attempted to train participants on a dual-task which was also comprised of an attention demanding task and an automatic task. Despite instructing participants to prioritize the attention demanding task, the majority of the participant’s performance was sub par relative to their single-task detection rate. In contrast, failure only rarely occurs when both tasks are automatic, suggesting that successfully timesharing an attention demanding task requires more than practice and appropriate task prioritization. After replicating Schneider and Fisk’s result a series of follow up studies were conducted to determine which factor(s) determine dual-task learning. Preliminary results of these studies will be presented.
October 7th, 2009 Professional Development: Panel on the Job Search Process The panel will offer their insights, advice and memories of the job search process. Come prepared with questions!
September 23rd, 2009 Elements of a Job Application in Psychology We will discuss the various types of academic jobs available in psychology. We will also talk about the types of application materials that are expected for each type of job (e.g., research statement, teaching statement). A brief overview of each component will be given with the goal of helping students to prepare these materials when necessary.
September 9, 2009 Mapping Brain Circuits with High Definition Fiber Tracking (HDFT) for Cognitive Neuroscience and Medical Imaging In the last 60 days we have made a dramatic advance in fiber tracking we call High Definition Fiber Tracking (HDFT). The technology provides in-vivo imaging allowing tracking of 300,000 fiber tracts throughout the brain with clear identification of the fiber contacts and millimeter resolution of connectivity and quantification of relative connectivity. The HDFT images are likely the highest fidelity images ever collected of the in-vivo human brain and on a par with micro dissection. HDFT allows tracing of circuits such as language processing, executive system, and affect and following individuals through development or disease process. We are applying HDFT in neurosurgerical planning and diagnosis of traumatic brain injury. Pittsburgh has a leading edge of this technology that can be applied to many areas of study including a proposal to map the Human Connectome. This talk introduces the technology and illustrates how it can be applied to basic issues such as mapping the connectivity of the Cognitive Control Net (CCN) and the Human Connectome. A goal of the talk is to seek collaborative opportunities to exploit the technology in Pittsburgh. |
|||||
 |
||||||
* If you would like to schedule a talk or have questions about the Brown Bag series, please send an e-mail to Elizabeth Richey, this year's student coordinator.
