|
|
 |
|||||
 |
April 18, 2012
Prior Probability as a Source of Bias in Perceptual Decision-Making Categorical decision-making is frequently modeled as a multi-stage process in which choices are determined by comparing the degree of accumulated evidence with a criterion level. It is well established that the criterion level can be influenced by expectations about event probabilities, speed/accuracy considerations, and personality traits, as well as other factors. As modeled by Signal Detection Theory, the criterion boundary for a given choice may be adjusted to optimize behavior when there is reason to think that one outcome is more likely than another. This outcome has been traditionally manipulated at the block level, with trials in that block all adhering to a given probabilistic outcome. In the present study, we manipulated prior probabilities on a trial-by-trial basis using cues. Subjects performed a face vs. house discrimination task and received a cue before each stimulus indicating the probability that the stimulus would be a face or house. For each of face and house, the probabilities ranged from high to neutral to low. We hypothesized that trial-level cueing would affect criterion levels in a manner consistent with traditional block-level manipulations. That is, following a high probability face cue, subjects would lower their criterion level to make a face choice. In line with our hypotheses prior probability was observed to lower criterion and increase accuracy for the biased response category. Future research will investigate the neural basis of evidence accumulation under varying levels of criterion.
Linguistic- and Event-Based Constraints in Sentence Processing: A Visual
World Study People use many different kinds of information when they understand language. However, it is currently unclear whether linguistic knowledge can be separated from world knowledge during sentence comprehension. Additionally, few studies have attempted to dissociate the different contributions of world knowledge and linguistic knowledge towards prediction of upcoming words. We present a visual world study designed to identify the different effects linguistic- and event-based constraints may have on prediction.
Cognitive Control and Cortical Gamma-band Synchrony Cognitive control encompasses a broad array of processes and networks that allow us to direct our actions in favor of achieving a desired goal. Recent work with EEG imaging has allowed researchers to begin characterizing the neural mechanisms allowing for such top-down guided behavior. Neural oscillations, specifically in the gamma frequency band (30-80 Hz), have been implicated in several perceptual, attentional, and executive functions. Here we present results of an EEG study demonstrating that frontal gamma-band synchrony modulates with preparatory control demands; moreover, we show that in a cohort of medication na•ve first episode patients with schizophrenia, a condition characterized by prefrontal and cognitive dysfunction, this effect is absent. However contrary to our hypothesis, the modulation manifested as a decrease from baseline raising the possibility of strategic adjustments from a heightened basal level of control. In further approaching this possibility, we present preliminary evidence of long-term control adjustments favoring a stable representation at the detriment of overall task performance.
The use of physical and virtual manipulatives in learning algebra
through robotics Studies have shown that robotics can act as an effective context for teaching mathematics (Nagchaudrhuri et al., 2002; Petre & Price, 2004). However, it is currently unclear whether the robot itself can help students to understand mathematical concepts. We asked students to complete a robot programming task using a physical robot or a virtual simulation of the same robot. We compared students' approaches to the task, as well as their mathematical learning gains after completing the task. Preliminary results show that students who interacted with the physical robot were more likely to pay attention to important features of the robot and to recognize mathematical strategies than students who interacted with the virtual simulation. These results suggest that the physical manipulative may provide unique benefits to students' mathematical learning.
Aerobic Fitness Moderates an Age-Related Decline in BDNF Higher aerobic fitness levels are associated with elevated cognitive performance in older adults. Unfortunately, the molecular pathways by which aerobic fitness is associated with enhanced cognitive and brain health in humans are poorly understood. In rodents, brain-derived neurotrophic factor (BDNF) is upregulated after exercise and is critical in learning, memory and the proliferation of new neurons in the hippocampus. In humans, serum BDNF levels decline in late adulthood, but greater amounts of serum BDNF have been associated with larger hippocampal volumes and better spatial memory performance. In this study we examined whether more aerobically fit older adults would have higher serum levels of BDNF and whether higher aerobic fitness levels would moderate an age-related decline in serum levels of BDNF. Fasting blood was collected on 145 adults between 59 and 79 years old (Mean age=67) and BDNF was determined by an enzyme-linked immunosorbent assay. Aerobic fitness levels were determined by a graded maximal exercise test (VO2max) on a motor-driven treadmill. Serum BDNF was not related to BDNF genotype, years of education, or sex. We employed a linear regression model with Age, Fitness, and an Age x Fitness interaction as predictors and serum BDNF as the dependent variable. Consistent with our hypothesis, serum BDNF declined with age (p<.05), but higher aerobic fitness levels offset the age-related decline (Age x Fitness interaction; p<.05). We conclude that higher aerobic fitness levels moderate an age-related decline in serum BDNF.
Effects of Omega-3 and Physical Activity on Cognitive Performance Many individual differences have been appearing in the cognitive performance literature, specifically with the aging population. The aim of the present study is to look at a sample of middle aged adults and identify what lifestyle factors predict cognitive performance. This moderator model focuses on diet, specifically omega03 fatty acid plasma levels, and physical activity and their roles on performance in the executive Trails task.
Translation Ambiguity between English and Mandarin Chinese
Adults prefer one-to-one mappings in translation, for which a word in one language translates into one word in another language (e.g., the Mandarin Chinese word Òsong1 shu3Ó only translates into the English word ÒsquirrelÓ), over one-to-many mappings in translation, for which a word in one language translates into more than one word in another language (e.g., the Mandarin Chinese word Òlao3 shu3Ó translates into the English words ÒmouseÓ and ÒratÓ). Even though adults prefer one-to-one mappings in translation, one-to-many mappings in translation are relatively common (Eddington, Degani, & Tokowicz, 2011; Prior, MacWhinney, & Kroll, 2007; Tokowicz, Kroll, de Groot, & van Hell, 2002). Here, I will show preliminary data from the translation norms that I am collecting for English and Mandarin Chinese, as well as describe plans for future research based on these translation norms.
The effect of meaning based strategies and the generation effect on
German vocabulary learning
The process of learning a new language can be fairly challenging especially for adult language learners. Vocabulary is thought to be one of the most essential aspects of learning a new language. Without an adequate vocabulary the second language (L2) learnersÕ production and comprehension abilities will be greatly limited. This study focuses on how different meaning-based vocabulary strategies influence second language vocabulary learning. Participants learned 48 German vocabulary words in four different training strategies. Participants either: read a definition for each German-English vocabulary word, read a sentence replacing the English word with the German word (e.g., ÒI spread butter on my Brot/bread.Ó) or generated their own definitions and sentences for the German vocabulary words. This was a 2 (generation vs. read) by 2 (sentence vs. definition) within subjects design. After participants trained on the vocabulary words they completed a filler task, and then were tested. Participants were first asked to recall all English-German words they could remember, and then were asked to translate the vocabulary words from German to English. Lastly participants were asked to make semantic relation decisions as to whether English-German word pairs were 1. Semantically related, 2. Translations, or 3. Unrelated. After one week participants came back and were tested again. Participants were more likely to recall words they learned by generating a sentence, definition or reading a sentence. The delayed (1 week) English German translation task revealed that participants were more likely to correctly translate words with which they had read or generated a sentence. Additionally, participants were more accurate in deciding English-German word pairs were semantically related when they had read or generated a sentence with the vocabulary word during training. Overall generating or reading a sentence using the German vocabulary was beneficial to German vocabulary learning. This suggests that using L2 vocabulary in a meaningful context can help aid in recall and aid in processing the L2 vocabulary in a meaning way.
The contributions of embodied action, mental imagery, and
iconicity to L2 lexical learning.
Through a series of 5 studies, this work aims to investigate the role of embodiment in second language (L2) lexical learning. In studies 1-4, adults unfamiliar with Hungarian learned word meanings via iconic gesture, imagery, meaningless motion, or a control condition in either an active or passive paradigm. The results revealed that active mental imagery formation facilitated L2 word learning to the greatest extent. In study 5, hearing adults unfamiliar with American Sign Language (ASL) learned signs varying in iconicity via enactment, mental imagery, meaningless motion, or a control condition. The results revealed that enactment, but not iconicity, enhanced sign acquisition. Overall, the results of this work suggest that mental imagery plays a pivotal role in L2 word learning, and that enactment plays a pivotal role in L2 sign learning. These findings will be discussed with reference to theories of L2 lexical acquisition (Ellis & Beaton, 1993) and embodied language processing (Barsalou, 1999).
Previous research in younger adults has demonstrated that Met carriers of the Brain-Derived Neurotrophic Factor (BDNF) Val66Met polymorphism have reduced volume and function and poorer episodic memory compared to their Val homozygous counterparts. In older adults, the effect of the BDN hippocampal F polymorphism is less consistent with several studies finding no effects of the polymorphism and others reporting both positive and negative effects of the Met allele on executive function and episodic memory. Few studies have examined the influence of the BDNF polymorphism on brain volume in older adults. To address this, we examined the influence of the BDNF polymorphism on the volume of the hippocampus, caudate nucleus, and putamen in a well-characterized sample of older adults (N = 155). An automated segmentation algorithm on high-resolution MR anatomical images was used to determine hippocampus, putamen, and caudate nucleus volumes. DNA was obtained from fasting blood samples and genotype was determined using standard protocols (15 met carriers and 140 val homozygotes). We found that the BDNF polymorphism was unrelated to hippocampal, caudate nucleus, or putamen volumes (all pÕs > 0.18) after controlling for variance associated with sex. The polymorphism was also unrelated to years of education, aerobic fitness levels, or body mass index. Despite this null result, linear regression analyses confirmed main effects of age and aerobic fitness on hippocampal volume (all p<.005). Overall, our results indicate that the BDNF polymorphism may have a negligible role in explaining individual variation in hippocampal, caudate nucleus, and putamen volumes in older adults.
Sleep behavior and its relation to cognitive performance in older adults Inadequate sleep is very common among the adult population. Only a small percentage of adults report getting the recommended eight or more hours of sleep per night. This is a concern given the marked impairments in cognition consistently shown with sleep deprivation. However, total sleep deprivation (when subjects are deprived for a full night) is not necessary to reveal dramatic cognitive impairments. More ecologically valid studies of chronic sleep restriction involving less than 7 hours of sleep per night for multiple nights have revealed a range of cognitive deficits including deficits on tasks of attention and working memory, similar to those found with total sleep deprivation. This type of chronic inadequate sleep is similar to sleep behavior of healthy older adults and could potentially result in cognitive impairments that may be difficult to reverse with a few good nights of sleep. Given that older adults also typically show impairments in cognition, it may be that poor sleep mediates these age-related cognitive impairments. The present study investigated the relationship between sleep behavior and task-switching performance in young and older adults. Stimulus onset asynchrony (SOA) varied trial-by-trial to examine how time to prepare influenced performance in good and poor sleepers. Preliminary results demonstrated that older adults with good sleep behavior maximized time to prepare compared with poor sleepers. Overall performance and pattern of results displayed by older adults with good sleep behavior was similar to young adults. These results suggest that older adults with good sleep behavior may be more likely to maintain cognitive abilities.
The Dynamics
of Interregional Synchrony During Spontaneous and Stimulus-Related
Activity in the Healthy and Autistic Brain. Assessing the correspondence between spontaneous and stimulus-related neural activity can reveal intrinsic properties of the brain. Recent studies have demonstrated that many large-scale functional networks have a similar spatial structure during spontaneous and stimulus-related states. However, it is unknown whether the temporal dynamics of network activity are also similar across these states. The first set of studies I will present demonstrate that, in the human brain, interhemispheric coupling of auditory cortices is preferentially synchronized in the alpha frequency band (~7-12 Hz) and interhemispheric coupling of somatosensory regions is preferentially synchronized in the high beta frequency band (~20-30 Hz). Critically, these preferred synchronization frequencies were the same for both spontaneous and stimulus-related activity. This similarity suggests that frequency-specific oscillatory dynamics are intrinsic to the interactions between the nodes of these brain networks. The second set of studies I will present examines whether these interregional dynamics are abnormal in individuals with autism spectrum disorders (ASD). To address this issue we measured neural synchrony during auditory click stimulation and during rest. Specifically, we examined synchrony between left hemisphere primary auditory cortex (Laud) and RAud in 16 high functioning individuals with ASD and 18 typically developing (TD) age and IQ matched controls. No group differences were found in the stimulus-specific synchrony between LAud and RAud. In marked contrast, the TD subjects had significantly greater spontaneous LAud-RAud synchrony in the alpha frequency range than ASD subjects. These results demonstrate that individuals with ASD have reduced spontaneous, but not stimulus-related neural synchrony.
It has been known for many years that words preceded by related words are processed more quickly and accurately than words preceded by unrelated words. This priming effect has been demonstrated through the use of behavioral designs in which accuracy and reaction times are measured, as well as through the use of electrophysiological studies measuring event-related potentials (ERPs). In the latter, ERP components thought to reflect lexical access/integration have been found to be modulated by the degree to which word pairs are associated. In this talk, I will describe findings from a semantic relatedness task in which measures of intralexical association, semantic similarity, and individual differences were collected and analyzed with respect to behavioral and electrophysiological responses. We found that, consistent with previous work, the ERPs were modulated by the degree of intralexical association. In addition, we found that ERPs were modulated by semantic similarity as indexed by latent semantic analysis (LSA). Finally these effects were correlated with offline measures of linguistic knowledge and performance. As the relatedness judgment task was part of a larger study, initial findings from a sentence comprehension task will be briefly described, as well.
The behavioral changes and computational dynamics of long-term complex motor learning Typical studies of motor learning focus on changes that occur within the span of 1 or 2 training sessions. However, many everyday skills are acquired over much longer time-scales, on the order of days, weeks or months. To date very little is known about how the behavioral dynamics, and underlying cortical networks, change as skills become ingrained to the point of being automatic. Our goal in this study was to determine how motor responses and learning dynamics change across long-term training on a complex sensorimotor task. Participants (N = 25) performed an adaptive, bimanual version of the serial reaction time (SRT) task (32-item long sequence) across 10 days of training. We found that extended training the SRT task shows continued performance changes up to the last day of training. Over time, a configural “chunking” of responses appears to take place, suggesting that separate key presses are being processed as a unified motor command over time. Analysis of learning dynamics, using linear dynamical systems modeling, shows that training increases reliance on previous response states as well as a greater influence of errors on future states after long-term training. These preliminary findings suggest that a major component of long-term, complex motor skill learning involves combining multiple motor plans into a singular, unified motor command.
How Motivation and Instruction Interact to Produce Successful Learning Successful theories in cognitive psychology have produced a small set of principles for the design of instructional activities. For example, prompts that guide students to self-explain (Chi, DeLeeuw, Chiu, & LaVancher, 1994), materials which asks students to compare and contrast (Richland & McDonough, 2009), and asking students to invent novel mathematical procedures before being shown a canonical method (Schwartz & Martin, 2004), have all been found to lead to better learning outcomes, such as the ability to transfer knowledge from instruction to novel scenarios. However, cognitive psychology theories of how students learn rarely draw on student motivation as a factor to consider in the success or failure of student learning. Conversely, theories chronicling the effect of student motivation on academic outcomes (e.g., Elliot & McGregor, 2001), such as grades, do not explore the variations in cognitive processes that may be recruited in different motivational states. In the current talk, I will present a framework for considering how a student's motivation may interact with instructional activities to influence successful learning outcomes. Specifically, I will demonstrate how different types of instructional activity produce different levels of engagement, with concomitant changes in learning; how existing motivational dispositions influence learning; and how different motivational states interact with the type of instruction to lead to successful learning outcomes. Possibilities for integrating student motivation into theories of how knowledge transfers from instruction will be discussed.
Predicting Performance with a Task-Based Behavioral Measure of Achievement Goals In recent years, educational researchers have given significant attention to the role of achievement goals in predicting behaviors and learning outcomes in academic environments. Most of the data collected in classroom settings has relied on self-report questionnaires (Linnenbrink-Garcia, Tyson, & Patall, 2008). Although questionnaires have provided a great deal of insight into the predictive value of achievement goals for performance and learning outcomes (e.g. tests, grades, SATs), the creation of a behavioral measure could further our understanding of how achievement goals relate to student behavior, classroom goal adoption, and academic achievement. The purpose of the current work was to test a behavioral measure of achievement goals embedded within a typical classroom activity in a non-intrusive format. I present the relationships between our behavioral data and students’ performance on an embedded task, their broader content knowledge, and their self-reported achievement goals. Finally, I explore this measure’s implications for deepening our understanding of the performance and learning outcomes associated with different achievement goals.
Capturing content and context effects on children's motivation towards science learning Under what conditions are children motivated and captivated by science? Given that children’s motivation affects a range of outcomes, including persistence and achievement (Sideridis & Kaplan, 2011; Pintrich & De Groot, 1990; Hidi,Renninger, & Krapp, 2004), understanding the complex factors underlying student motivation necessitates careful consideration. Through our pilot interviews, we identified a number of key dimensions affecting science motivation in 5th and 6th grade students. This information was used in the construction of a survey that incorporated these dimensions. The survey was administered to 250 5th and 6th graders to explore the various conditions under which children are motivated towards science learning, as well as the differences that arise between these factors among individuals. Implications for measurement, design work, and student engagement will be discussed.
January 18, 2012 Cross-language similarity in L2 processing: an eye-tracking study How do second-language learners process second language (L2) syntax? Using online (eye-tracking) and offline (grammaticality judgment) measures, we examine how L2 learners of English process grammaticality in syntactic structures that are formed similarly or differently in L1 and L2. Native Arabic and native Mandarin Chinese speakers who were studying English as an L2 read grammatical and ungrammatical English sentences. Their reading behavior was compared to that of monolingual English speakers. L2 learners showed native-like processing of the syntactic violations compared to the native English speakers at different measures. These results provide important considerations for models of second language acquisition and bilingualism, such as the Unified Competition Model (MacWhinney, 2005) and the Shallow Structure Hypothesis (Clahsen & Felser, 2006).
January 11, 2012 Construction of the Chinese Orthography Database: Application for Chinese second language acquisition This study aims to develop a database to provide psycholinguists with a large-scale orthographic representation system for Mandarin Chinese, as well as to examine orthographic awareness of Chinese as a factor in Chinese second language (CSL) learning. The construction of the Chinese Orthography Database is based on the composition of Chinese constituents: configurations and radicals. Four hundred and thirty-nine radicals are used, with eleven patterns of configurations, to define 6097 frequent characters. The characters come from the union of two sets of frequent characters defined by the Big-5 encoding method and the Chinese Knowledge and Information Processing group. The radical combinations enable several orthographic indices of Chinese characters to be computable; the indices include (a) configuration frequency, (b) radical frequency by type/token, (c) radical position-based frequency, and (d) neighborhood sizes of radicals. With these data-driven indices, researchers can not only control experimental stimuli precisely but construct experimental materials such as pseudo-characters or non-characters effortlessly. To further examine the validity of the database, 61 Thai speakers participated in this study that investigated the psychological functioning of Chinese orthography for CSL learners. The subjects were asked to compare two artificial characters and choose the one more like a real character. Experiment 1 assessed their knowledge of Chinese configuration types, while Experiment 2 examined their awareness of radical position-based frequency. The results revealed that the CSL learners are sensitive to the frequency of configuration and radical position. Overall, the use of the Chinese Orthography Database established a system for the experimental implementation of the study of CSL reading.
November 30, 2011 Exploring Sketching as a Window on Student Understanding Sketching was used in an experiment to assess what students learn when they read science texts. Higher and lower ability students were asked to read science texts that contained complex diagrams. Students used sketching software to copy diagrams from the texts. Sketching was reliably associated with student learning. Sketches revealed aspects of student learning that were similar to those uncovered using text-based assessments. Results also suggest that sketching can reveal unique aspects of student learning not shown in text-based assessments.
December 7, 2011 The psychological path that leads to breakthrough innovations in engineering design and other creative domains is complex and difficult to unpack. However, cognitive science has been able to illuminate some key cognitive processes involved in this path. One key process that has been identified is analogy, which involves systematically mapping and transferring elements and relations from some base knowledge domain to a target domain. In particular, analogies that draw from domains distant (e.g., surface dissimilar) from one’s problem domain have often been viewed in the literature as a significant and effective means of generating highly novel and useful ideas. In this talk, I present findings in support of a more cautious characterization of the impact of analogy on creative thought. In a detailed analysis of the temporal interplay between analogy use and ideation in the naturalistic brainstorming conversations of a real-world professional design team, self-generated analogies (including distant ones) were not systematically associated with the generation of highly novel ideas; on the contrary, ideas tended to be more similar to their precedents after analogy use in comparison to baseline situations (i.e., without analogy use). This surprising finding joins some other work (e.g., Kevin Dunbar’s in vivo studies of analogy and scientific discovery) in questioning the oft-touted direct path between distant analogies and creative ideation, and suggests important theoretical and practical implications regarding the use of analogy in creative thought.
November 30, 2011 Exploring Sketching as a Window on Student Understanding Sketching was used in an experiment to assess what students learn when they read science texts. Higher and lower ability students were asked to read science texts that contained complex diagrams. Students used sketching software to copy diagrams from the texts. Sketching was reliably associated with student learning. Sketches revealed aspects of student learning that were similar to those uncovered using text-based assessments. Results also suggest that sketching can reveal unique aspects of student learning not shown in text-based assessments.
November 9, 2011 A Mismatch Between Confidence and Performance?:
Evidence from an ESL Spelling Intervention Study
Adult learners of a second language must learn new mappings among orthographic, phonological, and semantic representations (Perfetti & Hart, 2001, Verhoeven, 2000). This can be especially difficult in English as a second language (ESL), due to its inconsistent grapheme-to-phoneme and phoneme-to-grapheme mappings (Frost, Katz, & Bentin, 1987). Over the course of one semester, non-native English speakers in the English Language Institute at the University of Pittsburgh participated in an in-class English spelling intervention experiment designed to improve their orthographic representations. As students went through the training paradigm, they had opportunities to check the accuracy of their spelling productions and to re-attempt words they checked. I will report data on how first language background (i.e., Arabic, Chinese, Korean, Spanish) and second language proficiency (i.e., intermediate to advanced) related to study behaviors (listening, checking, retrying, and reviewing) during the spelling intervention training tasks. Specifically, I will describe patterns of checking and retrying words relative to the accuracy of spelling during training and at testing. I will show that there is sometimes a mismatch between a student’s confidence in their grasp of knowledge components and his or her actual performance on tests of production.
November 9, 2011 Spatiotemporal brain dynamics of inhibitory control Kai Hwang, Avniel Ghuman, Beatriz Luna In two experiments, we demonstrate that error-related negativities (ERNs) recorded during spelling decisions can expose individual differences in lexical knowledge. The first experiment found that the error-related negativity (ERN) can be elicited during spelling decisions, and that its magnitude was correlated with independent measures of subjectsÕ spelling knowledge. In the second experiment, we manipulated the phonology of misspelled stimuli and observed that ERN magnitudes were larger when misspelled words altered the phonology of their correctly spelled counterparts than when they preserved it. Thus, when an error is made in a decision about spelling, the brain processes indexed by the ERN reflect both phonological and orthographic input to the decision process. In both experiments, ERN effect sizes were correlated with assessments of lexical knowledge and reading, including offline spelling ability and vocabulary knowledge. These results affirm the interdependent nature of orthographic, semantic, and phonological knowledge components and demonstrate the value of ERNs as indicators of lexical knowledge.
November 2, 2011 Biological basis of variability in brain function over adolescence Adolescence is a period of development characterized by neurobiological changes that significantly influence behavior and brain function. Adolescence is of particular interest due to the alarming statistics indicating that despite a peak in physical health, mortality rates increase two to three-fold during this time compared to childhood, due largely to a peak inrisk-taking behaviors resulting from increased impulsivity, sensation seeking and perhaps an increased sensitivity to incentives. To this end, I will present work from my master's thesis studying the influence of rewards on cognitive control and the underlying neural circuitry from childhood to adulthood. Furthermore, there exists large variability in these behaviors, some of which is likely explained by environmental factors (i.e. social). However,unexplained variability exists that may be more inherent and mediated bybiological factors (i.e hormonal, neurochemical, genetic). Recent advances in human molecular genetics and functional neuroimaging have provideda unique and exciting opportunity to non-invasively study the influence of biological factors on brain function in humans. While genes donÕt code for specific behaviors, they do determine the structure and function of proteins that are essential to the neuronal processes that underlie behavior. I will review existing literature relating to the neurobiologicalbasis of adolescent development in the context of dopamine neurotransmission. Dopamine has been of specific interest because of extensive research linking its role in cognitive, motivational, and affective behaviors, its role in the pathogenesis of major psychopathology, and its protracted development over adolescence. I will then propose using imaging genetics to explore the influence of the catechol-o-methyltransferase gene on brain function in adolescence as measured by fMRI and present some preliminary findings from a resting state functional connectivity study.
October 26, 2011 Error-Related Negativities During Spelling Judgments Expose Orthographic Knowledge In two experiments, we demonstrate that error-related negativities (ERNs) recorded during spelling decisions can expose individual differences in lexical knowledge. The first experiment found that the error-related negativity (ERN) can be elicited during spelling decisions, and that its magnitude was correlated with independent measures of subjectsÕ spelling knowledge. In the second experiment, we manipulated the phonology of misspelled stimuli and observed that ERN magnitudes were larger when misspelled words altered the phonology of their correctly spelled counterparts than when they preserved it. Thus, when an error is made in a decision about spelling, the brain processes indexed by the ERN reflect both phonological and orthographic input to the decision process. In both experiments, ERN effect sizes were correlated with assessments of lexical knowledge and reading, including offline spelling ability and vocabulary knowledge. These results affirm the interdependent nature of orthographic, semantic, and phonological knowledge components and demonstrate the value of ERNs as indicators of lexical knowledge.
October 12, 2011 Assessing the Magnitude and Specificity of Transfer Effects after Ten Days of Intensive Cognitive Training Jeff Phillips, Timothy Verstynen, Emily Braun, Brett Workman, Lucas Horan, Christian Schunn, & Walter Schneider Intensive cognitive training regimens have generated widespread interest, due to their potential applications to education and clinical rehabilitation. In this experiment, we examine the efficacy and specificity of training in two paradigms: complex working memory (CWM) and serial reaction time (SRT) tasks. Fifty participants (25 per group) were trained for 10 days in 30-40 minute sessions. Before and after training, they completed a battery of near- and far-transfer assessments, including working memory, reaction time, Stroop, logical inferences, and non-verbal reasoning tasks. CWM trainees improved from a starting verbal span of 4.9 (1.3) to 8.8 (2.5) items, and from a spatial span of 3.1 (0.7) to 6.4 (1.3) items. Likewise, SRT trainees demonstrated robust sequence learning, relative to randomized responses (mean z=3.4). In transfer assessments, the CWM group improved on digit span and spatial working memory tasks, while the SRT group improved on a simple reaction time task. These results suggest that short periods of intensive cognitive training can affect performance on closely-related cognitive tasks.
September 28, 2011 Effect of Frequency on Eye Movements in Reading and Visual Search Studies have demonstrated that the effect of word frequency on eye movements is robust when participants are reading for comprehension or proof-reading but not when they are scanning text for a target word (Kaakinen & Hyöna, 2010, JEP: Learning, Memory & Cognition, 36, 1561-6; Rayner & Fischer, 1996, Perception & Psychophysics, 58(5), 734-47; Rayner & Raney, 1996, Psychonomic Bulletin & Review, 3(2), 245-8). As an extension to these studies, we compared eye-movement behavior on fully-crossed high- and low-frequency adjective-noun pairs embedded in paragraphs when participants were reading vs. engaged in one of three visual search tasks: scanning for a target word with an asterisk (e.g., “h*rse”), a word containing the letter "q" (e.g., “quilt”), or a word rhyming with "blue" (e.g., “shoe”). The results demonstrate that eye-movement measures are affected by frequency in the tasks requiring in-depth processing, such as reading and rhyme-judgment, but not in shallow-processing tasks like asterisk-detection. In addition, we report task-dependent analyses investigating spillover (i.e., the effect of the adjective frequency on the processing of the noun) and any parafovea-on-foveal effects (i.e., the effect of the noun frequency on the processing of the preceding adjective). Results suggest that the degree to which attention is allocated in a serial versus parallel manner is modulated by task demands
| ||||
 |
|||||
* If you would like to schedule a talk or have questions about the Brown Bag series, please send an e-mail to Chelsea Eddington, this year's student coordinator.
