12:00 - 1:30 pm
Back Bay A&B
12:00 - 1:30 pm
Back Bay C&D
12:00 - 1:30 pm
Data Blitz Session 1
Saturday, March 24, 12:00 - 1:30 pm, Back Bay A&B
Speakers: Samantha Cohen, Aaron Kucyi, Karen Campbell, Jonathan Greenberg, Megan Boudewyn, Amy Belfi, Tzipi Horowitz-Kraus, Joe Bathelt, Richard Betzel, Pieter Verbeke, Emily Kubicek, Guannan Shen, Daniela Palombo, Noah C. Yeagley, Ying Cai
Talk 1: Age and sex modulate the variability of neural responses to engaging videos
Samantha Cohen1,2, Agustin Petroni1, Nicolas Langer1,3, Simon Henin1, Tamara Vanderwal5, Michael P. Milham3,6, Lucas C. Parra1; 1The City College of New York, 2The Graduate Center of the City University of New York, 2Child Mind Institute, 2University of Zurich, 2Yale Child Study Center, 2Nathan Kline Institute for Psychiatric Research
Neural development is generally marked by an increase in the efficiency and diversity of neural processes. In a large sample (N = 114) of human children and adults with ages ranging from 5 - 44 years, we investigated the neural responses to naturalistic video stimuli. Videos from both real-life classroom settings and Hollywood feature films were used to probe different aspects of attention and engagement. For all stimuli, older ages were marked by more variable neural responses. Variability was assessed by the inter-subject correlation of evoked electroencephalographic (EEG) responses. Young males also had less variable responses than young females. These results were replicated in an independent cohort (N = 303). When interpreted in the context of neural maturation, we conclude that neural function becomes more variable with maturity, at least during the passive viewing of real-world stimuli.
Talk 2: Frequency-dependent temporal fluctuations of functional connectivity within intrinsic networks in human cortex
Aaron Kucyi1, Josef Parvizi1; 1Stanford University
Intrinsic brain networks are comprised of remote regions that have correlated spontaneous activity seen in functional imaging, known as intrinsic functional connectivity (FC), that is remarkably persistent across distinct consciousness states. Given the limited temporal resolution of imaging methods, little is known about fluctuations of FC on the scale of tens of seconds (known as “dynamic FC”), and controversy remains over the potential behavioral and neural relevance of dynamic FC. Here we aimed to identify electrophysiological correlates of intrinsic BOLD FC in canonical cortical networks and to clarify how electrophysiological FC within those networks fluctuates across frequencies on short time scales. We studied neurosurgical patients with intracranial electrodes directly implanted simultaneously within individually-localized nodes of the default, dorsal attention and frontoparietal control networks. Within these networks in both wakeful rest and sleep states, we found that electrophysiological functional connectivity of both high-frequency broadband (HFB, or high gamma; 70-170 Hz) and alpha (8-12 Hz) power amplitudes were reproducibly correlated spatially with functional connectivity in separately recorded resting-state fMRI within the same subjects. Although there were modest negative correlations between HFB and alpha local activity within each network, spatial connectivity patterns showed similarities between these frequency ranges. In contrast, within-network HFB and alpha coupling often diverged from one another temporally across short windows (on the order of seconds). These results suggest that temporal fluctuations of functional connectivity within widely studied human brain networks are shaped by multiple, dissociable neurophysiological processes that potentially have distinct behavioral relevance.
Talk 3: I did it my way: Explaining age-related declines in inter-subject synchronization during naturalistic viewing
Karen Campbell1, Cam-CAN2, Linda Geerligs3; 1Brock University, 2Cambridge Centre for Ageing and Neuroscience, University of Cambridge and MRC Cognition and Brain Sciences Unit, 2Donders Institute for Brain, Cognition and Behaviour, Radboud University
When people watch the same naturalistic stimuli (movies), significant inter-subject synchronization (or correlation) between their fMRI timecourses can be observed, not only in primary sensory areas, but also in frontal and parietal regions. We have previously shown that this inter-subject synchronization decreases with age (Campbell et al.,2015-Neurobiol Aging). In the current study, we replicate this effect in a larger sample (N=585) from the Cambridge Centre for Ageing and Neuroscience (www.cam-can.com), and aim to determine its etiology. We show that while synchrony within primary auditory and visual regions is preserved with age, it declines within a number of higher-order regions, mainly the default mode (DMN) and frontoparietal control networks (FPCN). Cluster analyses confirmed that aging is associated with an increase in idiosyncratic (or individualistic) responding to the movie, rather than the emergence of distinct subgroups who respond to the movie in a similar way. Using a sliding window analysis, we show that synchrony within the FPCN and DMN increases throughout the course of the movie and that older adults’ idiosyncratic responding increases over time. Finally, we show that (controlling for age) greater synchrony in the FPCN relates to greater functional connectivity (FC) within and between the FPCN and DAN, while synchrony in the DMN relates to FC within and between the DMN and inferior temporal cortex. Together, these findings suggest that age-related differences in FC may contribute to older adults’ idiosyncratic responding, and this idiosyncratic responding appears to increase over time, possibly reflecting older adults’ diverging interpretations as the narrative progresses.
Talk 4: Reduced interference in working memory following mindfulness training is associated with increases in hippocampal volume
Jonathan Greenberg1,2, Victoria L Romero3, Seth Elkin-Frankston3, Matthew A Bezdek4, Eric H Schumacher4, Sara W Lazar1,2; 1Department of Psychiatry, Massachusetts General Hospital, 2Harvard Medical School, 2Charles River Analytics, 2Georgia Institute of Technology
Proactive interference occurs when previously relevant information interferes with retaining newer material. Overcoming proactive interference has been linked to the hippocampus and deemed critical for cognitive functioning. However, little is known about how this ability can be improved or about the neural correlates of such improvement. Mindfulness training emphasizes focusing on the present moment and minimizing distraction from competing thoughts and memories. It improves working memory and increases hippocampal density. The current study examined whether mindfulness training reduces proactive interference in working memory and whether such improvements are associated with changes in hippocampal volume. 75 participants were randomized to a four-week web-based mindfulness training program or a similarly structured creative writing active control program. The mindfulness group exhibited lower proactive interference error rates compared to the active control group following training, and these improvements significantly associated with volume increases in the left hippocampus. These results provide the first evidence suggesting that mindfulness training can protect against proactive interference, and that these benefits are related to hippocampal volumetric increases. Clinical implications regarding the application of mindfulness training in conditions characterized by impairments to working memory and reduced hippocampal volume such as aging, depression, PTSD, and childhood adversity are discussed.
Talk 5: Tracking Attention to Spoken Language using EEG Alpha Oscillations
Megan Boudewyn1, Cameron Carter1; 1University of California, Davis
STUDY GOALS: Attention is critical to the construction of mental representations of language context during comprehension. We investigated the consequences of momentary lapses in attention during listening comprehension on neural activity and behavior. METHODS: Healthy participants (N=44) listened to two full-length stories while EEG was recorded, and afterwards completed multiple choice comprehension questions. Listening was periodically interrupted by attention probes, in which participants were asked whether their attention immediately preceding the probe’s appearance was focused on the story. RESULTS SUMMARY: The results showed that (1) participants spent a substantial amount of time off-task, endorsing attention lapses on over 30% of probes; (2) for probes on which an attention lapse was endorsed, later accuracy on comprehension questions querying pre-probe information was decreased; (3) just prior to the endorsement of an attention lapse, a greater proportion of above-threshold oscillations in the alpha-band (8-12 Hz) was detected compared to just prior to the endorsement of on-task or split-attention listening; and (4) when participants made “I have no idea” responses to comprehension questions, their EEG record revealed a greater proportion of above-threshold alpha oscillations during the original presentation of the information queried by the comprehension questions, compared to correct responses or incorrect guesses. CONCLUSIONS: These results connect changes in neural activity in the alpha band to episodes of mind-wandering during listening comprehension, and in turn to decreased comprehension accuracy. This demonstrates how alpha can be used to track attentional engagement during language comprehension, and illustrates the dependence of successful language comprehension on attention.
Talk 6: The left anterior temporal lobe is a bidirectional convergence region mediating the relation between names and semantic knowledge for unique entities
Amy Belfi1, Brett Schneider2, Jonah Heskje3, Joel Bruss3, Daniel Tranel3; 1Missouri University of Science & Technology, 2University of Wisconsin-Madison, 2University of Iowa
Prior research has implicated the left temporal pole (LTP) as a critical region for naming semantically unique items, including famous faces, landmarks, and musical melodies. Such studies typically use a confrontation naming paradigm, where a participant is presented with a stimulus and asked to retrieve its name. Here, we propose that the LTP functions as a two-way convergence region for proper naming. Under this hypothesis, damage to the LTP should result in impairments both in name retrieval when presented with a concept (as in prior work) and in concept retrieval when given a name. We tested this hypothesis using a “recognition-from-name” paradigm. Participants included individuals with LTP damage, and brain-damaged and healthy comparison participants. Participants were presented with names of famous individuals (e.g., Marilyn Monroe), landmarks (e.g., Leaning Tower of Pisa), or melodies (e.g., Rudolph the Red-Nosed Reindeer) and were asked to provide conceptual knowledge about each. Individuals with LTP damage were significantly impaired at conceptual knowledge retrieval when given names of famous people and landmarks, but not melodies. This supports the theory that the LTP is a bidirectional convergence region for proper naming, but suggests that melody retrieval may rely on additional processes other than those supported by the LTP.
Talk 7: Alterations in neural circuits supporting executive functions in children with reading difficulties
Tzipi Horowitz-Kraus1,2, Rola Farah1; 1Educational Neuroimaging Center, Faculty of Education in Science and Technology, Technion, Haifa, Israel, 2Reading and Literacy Discovery Center, General Pediatrics, Cincinnati Children's Hospital Medical Center, Ohio, USA
Reading difficulties have been linked to challenges in engaging executive functions (EF) while reading, such as working memory, attention, and speed of processing. Our study was designed to directly define the differences in neural processing during the Stroop task, which is a classical EF task, in children with reading difficulties compared with typical readers. Reading and EF abilities were assessed in 8-12 year-old children with reading difficulties and typical readers. Functional connectivity and global efficiency of several cognitive-control networks during a functional MRI Stroop task were defined in both groups. Children with reading difficulties showed lower reading and EF abilities and demonstrated greater functional connectivity between the EF network and visual, language, and cognitive-control regions during the Stroop task compared to typical readers. Our results suggest that children with reading difficulties utilize neural circuits supporting EF more so than do typical readers in order to perform an EF task. The connection between reading and EF suggests that insufficient triggering of EF in childhood will harm reading acquisition later in life. It also suggests that an EF-based training in childhood has the potential to improve future reading abilities in children.
Talk 8: The cingulum as an important measure of individual difference in brain development
Joe Bathelt1, Mengya Zhang1, the CALM team1, Duncan Astle1; 1MRC Cognition & Brain Sciences Unit, University of Cambridge
We investigated the relationship between individual differences in white matter microstructure and cognitive abilities in children. Individual differences in white matter were investigated in a representative database of typical development (NKI Rockland Sample, n=74, Age: 13.93 + 3.164SD) by extracting FA values for 10 major white matter tracts (JHU white matter atlas) and grouping individuals by similarity using a data-driven clustering approach. The algorithm indicated the presence of two groups that were distinguished primarily by FA of the left and right anterior cingulum (p<0.001). The range of FA values within the cingulum were used to group children in an independent sample with large variation in cognitive abilities (Centre for Attention, Learning, and Memory; n=165, Age: 9.81 -1.191SD). Comparison of cognitive scores between these groups indicated significant differences in fluid IQ, vocabulary, verbal and visuospatial short-term and working memory, and long-term memory (p<0.05). To investigate the association between cognitive scores and specific connections, the cingulum tract was reconstructed and connections were mapped using a connectomics approach. The results indicated significant differences between the clustering-defined groups in connections of the cuneus, parahippocampal, enthorhinal, and superior frontal cortex (p<0.05). A specific association between variation in fluid IQ and strength of the connection between the left precuneus and left superior frontal cortex was found (beta=0.286, p=0.005). These results indicate that cingulum-mediated connections are closely associated with inter-individual variation in cognitive ability in development.
Talk 9: The specificity and robustness of long-distance connections in weighted inter-areal structural brain networks
Richard Betzel1, Danielle Bassett1; 1University of Pennsylvania
The functional repertoire of a brain area depends upon the configuration of its incoming and outgoing connections. The complete set of these connections defines a connectome, which can be represented as a network and analyzed using tools from network science. Among the most salient features of brain networks is their cost efficient spatial embedding, which results in disproportionately many short-range connections. Nonetheless, brain networks exhibit a small number of costly long-distance connections, which are thought to confer functionality to neural systems by enabling efficient signaling and transfer of information by reducing the average number of processing steps separating any two brain areas. However, in real-world networks connection weights are log-normally distributed and weaken monotonically as a function of distance. This observation implies that long-distance connections play a minor role in the network's shortest path structure, suggesting a diminished capacity for promoting efficient inter-areal communication. What then is the functional role of long-distance connections? Using a network analytic framework, we provide evidence suggesting that long-distance connections confer connectional specificity to brain areas and in the process define its functionality. Using four network datasets (human, macaque, drosophila, and mouse), we show that long-distance connections are also highly reinforced and redundant, suggesting that long-distance connections are resilient to most perturbations. However, using dynamical simulations, we show that in their absence the complexity of spontaneous neural activity is dramatically reduced. These findings help clarify the functional roles of costly long-distance architectural features and inform future studies of inter-areal network structure and function.
Talk 10: Fast synchronization and slow synaptic learning as a solution to the stability-plasticity dilemma
Pieter Verbeke1, Tom Verguts1; 1Ghent University
The human ability to adapt to a constantly changing environment is remarkable. This relies on the ability to learn quickly about associations between perceptual, motor, and goal representations. Nevertheless, fast learning in neural networks typically leads to forgetting of older information; this is unfortunate because one would like to retain environmental regularities without disruption from novel information. Thus, there exists a tradeoff between being sufficiently adaptive to novel information (plasticity) while retaining valuable earlier regularities (stability). We propose that the brain deals with this issue by relying on two separate, yet interacting learning mechanisms in the same neural structures. The first, fast learning mechanism implements binding-by-synchronization (Fries, 2015) (sync learning). Here, perceptual, motor, and goal representations are bound together by synchronization of neural firing. The second, slow learning, mechanism corresponds to classical synaptic learning by (reward-modulated) Hebbian learning. To implement this hypothesis, we adapted the Verguts (2017) model and tested it on a reversal learning task. Simulations demonstrated that a model using only synaptic learning could not deal with sudden changes in task rules. A model using only sync learning could flexibly deal with task rule changes, but overwrote earlier learned rules. Combining sync learning with synaptic learning however, allowed the model to deal with task rule changes without overwriting earlier information. Thus, the resulting model combined (fast) plasticity using sync learning with (slow) stability using synaptic learning to address the stability-plasticity dilemma. In addition to solving this computational problem, we compare the model to neurophysiological and –anatomical data.
Talk 11: Deaf signers’ sensorimotor system activity during perception of one and two handed signs
Emily Kubicek1, Lorna C. Quandt1; 1Gallaudet University
When a person observes someone else performing an action, the observer’s sensorimotor cortex activates as if the observer is the one performing the action, a phenomenon known as action simulation. While this process has been well-established for basic (e.g. grasping) and complex (e.g. dancing) actions, it remains unknown if the framework of action simulation is applicable to visual languages such as American Sign Language (ASL). We performed an EEG experiment to test whether deaf signers’ sensorimotor systems are differentially sensitive to signs that are produced with one hand (“1H”) or two hands (“2H”). We predicted greater alpha event-related desynchronization (ERD; previously correlated with action simulation) during the perception of 2H ASL signs compared to 1H ASL signs, due to greater demands on sensorimotor processing systems required for producing two-handed actions. The two groups of signs were matched for frequency, iconicity, and flexion. We recorded EEG from deaf participants fluent in ASL as they observed videos of ASL signs, half 1H and half 2H. Event-related spectral perturbations (ERSPs) in the alpha range were computed for the two conditions at central electrode sites overlying the sensorimotor cortex and paired comparisons showed significantly more alpha ERD when participants observed 2H signs as compared to 1H signs (ps <.05, FDR corrected). This finding suggests action simulation processes contribute to deaf fluent signers’ observations of ASL, and that these processes are sensitive to linguistic/motoric parameters of sign language. This work provides the first investigation of alpha oscillations during sign language perception.
Talk 12: Exploring Categorical and Functional Boundaries of Tactile Perception Using Somatosensory Mismatch Responses
Guannan Shen1, Peter J. Marshall1; 1Department of Psychology, Temple University
The current study examines categorical and functional boundaries in somatosensory perception, and explores two potential influences on tactile spatial discrimination: discrete body part representations and motor experiences. The mismatch negativity (MMN), an index of early automatic sensory discrimination, has been shown to be sensitive to categorical boundaries and language experience in the auditory modality. Here, we recorded somatosensory MMN (sMMN) using a tactile oddball paradigm, and compared sMMN elicited by within- and across-boundary oddball pairs. We first leveraged a categorical segmentation of tactile space at the wrist, and presented two pairs of tactile oddball stimuli with equal spatial distances, either across the wrist or within the forearm. Results showed that sMMN amplitude elicited by stimuli across the body part boundary was significantly greater than for sites within the forearm, suggesting a categorical effect on tactile perception at the early pre-attentive stage of somatosensory processing. We then examined sMMN elicited by an oddball contrast of the 3rd finger / thumb, which was significantly larger than for the 3rd / 5th fingers. We suggest this indicates a functional boundary effect on tactile perception of finger stimulation as a result of motor experiences (e.g., grasping, picking up objects). These findings demonstrate that the sMMN is a useful neurophysiological index of pre-attentive processing of somatosensory spatial discrimination, sensitive to both categorical and motor-related functional boundary effects. This paradigm can be potentially used in studying the development of somatosensory representations and body maps in early infancy.
Talk 13: Hippocampal contributions to reward learning
Daniela Palombo1,2, Mieke Verfaellie1,2; 1VA Boston Healthcare System Jamaica Plain, 2Boston University School of Medicine, Department of Psychiatry
Recent evidence suggests that the human hippocampus—known primarily for its involvement in episodic memory—also plays a role in a host of motivationally relevant behaviors, including value-based decision making. However, less is known about the role of the hippocampus in value-based learning. Such learning is typically associated with a striatal system, yet a small number of studies also show hippocampal engagement. It is not clear, however, whether this engagement is necessary for such learning. In the present study, we used both fMRI and neuropsychological methods to clarify hippocampal contributions to one form of value-based learning, namely reward learning. In experiment 1, healthy participants were scanned while learning reward-based contingencies (whether distinct players in a ‘game’ would win money or not) in the context of a probabilistic learning task. Here we observed recruitment of the hippocampus during learning. In experiment 2, we administered this task to amnesic patients with medial temporal lobe damage and healthy controls. Amnesic patients, including those with damage circumscribed to the hippocampus, failed to acquire reward contingencies, thus confirming that hippocampal engagement is necessary for task performance. Although future research is needed to determine the boundary conditions of hippocampal involvement in value-based learning, these findings elucidate the role of the hippocampus in learning and may clarify how the hippocampus contributes to goal-directed behaviors more broadly.
Talk 14: Expertise Matters in Evaluating Students’ Organization of Neuroscience Concepts
Noah C. Yeagley1, Jennifer L. Stevenson1, Joel P. Bish1; 1Ursinus College
Evaluating learning is a pivotal part of the academic process, but most assessments are explicit measures, such as tests and quizzes. One form of implicit evaluation is the Structural Assessment of Knowledge (SAK) which examines the organization of knowledge structures or networks. The current study investigates undergraduate students’ learning of structure-function relationships (n=38) and neuronal physiology (n=40) in introductory and advanced neuroscience courses. Students made pairwise rankings of 15 neuronal physiology (e.g., action potential, axon hillock) or 15 structure-function relationship concepts (e.g., Broca’s area, language production) in terms of their similarity before and after learning. Using Pathfinder software, students’ networks were compared to two types of expert networks: their individual professor and a group of four other neuroscience professors at the college. The type of expert (individual or group of professors) interacted with type of student (introductory or advanced) for both neuronal physiology (F(1,38)=19.16, p<.001) and structure-function relationships (F(1,36)=23.15, p<.001) concepts; however, the details of the interactions were reversed. For both types of concepts, students were equally similar to individual and group professors. However, for neuronal physiology, students were more similar to the group of professors than their individual professor (t(15)=-5.64, p<.001) and for structure-function relationships, students were more similar to their individual professor than the group of professors (t(16)=5.82, p<.001). These results suggest multiple factors such as type of student, type of expert, and type of knowledge are important when considering the expert for comparison in SAK. This may be more complex in interdisciplinary fields like neuroscience.
Talk 15: The influence of storage capacity versus control in visual working memory capacity limitations
Ying Cai1,2, Andrew D Sheldon3, Bradley R Postle2,4; 1National Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, 2Department of Psychiatry, University of Wisconsin–Madison, 2Medical Scientist Training Program and Neuroscience Training Program, University of Wisconsin–Madison, 2Department of Psychology, University of Wisconsin–Madison
Many studies of visual working memory (VWM) capacity confound the constructs of "storage capacity," "context binding," and “inter-item interference.” During fMRI of a delayed-estimation task (n=16), we dissociated the experimental factor of load -- one bar that could vary in orientation ("1O") vs. three differently oriented bars ("3O") – from that of category homogeneity -- 3O vs. one bar, one patch that could vary in chrominance, and one concentric-circle stimulus that could vary in luminance contrast ("1O1C1L"). Fitting behavioral data to a 3-factor mixture model revealed no difference in memory precision between 1O and 1O1C1L conditions, but significantly worse precision for 3O. Subjects with higher VWM capacity, estimated offline with color change-detection, had a smaller decline in precision of 3O relative to 1O1C1L. Probability of target responses was highest for 1O, followed by 1O1C1L, followed by 3O. Delay-period BOLD signal in parietal and frontal areas was comparable in 1O and 1O1C1L conditions, and higher for 3O. Orientation reconstruction with multivariate inverted encoding modeling (IEM) of this delay-period signal was only successful for 1O trials, a pattern most consistent with a context binding function. IEM of encoding-related signal in occipital cortex, in contrast, produced successful orientation reconstruction for 1O and 1O1C1L, but not for 3O. Furthermore, dividing subjects into homogeneity-sensitive versus homogeneity-insensitive groups (based on the behavioral measures) indicated that the 1O1C1L-3O reconstruction difference was most pronounced in the homogeneity-sensitive group, a pattern most consistent with an inter-item interference. Fronto-parietal VWM activity reflects control, an important determinant of VWM capacity.
Data Blitz Session 2
Saturday, March 24, 12:00 - 1:30 pm, Back Bay C&D
Speakers: Katarzyna Siuda-Krzywicka, Renee M. Visser, Patrick Ledwidge, Xiaoqing Hu, Rola Farah, Jie Zheng, Darya Zabelina, Örjan de Manzano, Louisa Kulke, Huang Zheng, Felipe Pegado, Gunes Sevinc, Pin-Hao Andy Chen, Amandine E. Grenier, Julie Sato
Talk 1: Color categorization without color naming: neuropsychological evidence
Katarzyna Siuda-Krzywicka1, Christoph Witzel2, Emma Chabani1, Myriam Taga3, Laurent Cohen1,4, Paolo Bartolomeo1; 11. Inserm U 1127, CNRS UMR 7225, Sorbonne Universités, UPMC Univ Paris 06 UMR S 1127, Institut du Cerveau et de la Moelle épinière, ICM, Hôpital de la Pitié-Salpêtrière, Paris, France, 2Justus-Liebig-Universität Gießen, 2University of East London, 2Hôpital de la Pitié Salpêtrière Paris, France
Colors vary continuously, however we group them into distinct categories associated with specific color names (green, blue, etc.). It is an open question whether color categories are a product of linguistic labeling and thus, if categorizing colors requires transferring color information between visual and language areas. To answer this question, we investigated color categorization in a brain-damaged patient with a rare, selective impairment of color naming. The patient’s lesion affected the left mesial occipitotemporal areas and the splenium of corpus callosum, abolishing direct communication between the bilateral visual cortex and language areas in the left hemisphere. We designed a new task measuring color categorization without explicit color naming. On each trial, we presented two color pairs, one from the same category (e.g. light-green, dark-green), the other from different categories (e.g. orange and yellow). The patient and a group of healthy controls (n=50) had to indicate pairs containing same-category colors. In a control experiment, the patient and healthy controls (n=12) named the colors from the color-categorization task. The patient performed normally on color categorization. Yet, his color naming was severely impaired only in chromatic, but not achromatic colors (black, white, grey). This chromaticity-effect was considerably weaker in color-categorization. Our results challenge the hypothesis that color categorization, like naming, involves language system. Our findings also indicate that compared with color categorization, color naming strongly segregates chromatic and achromatic information. Given that color names refer to color categories, our findings suggest that the processes of non-linguistic categorization are fundamental for color naming.
Talk 2: Persistence of hippocampal activation patterns in post-encoding rest predicts subsequent voluntary, but not involuntary recall of distressing film clips
Renee M. Visser1, Richard N. Henson1, Emily A. Holmes1,2; 1Medical Research Council Cognition and Brain Sciences Unit, University of Cambridge, UK, 2Karolinska Institutet, Stockholm, Sweden
A key question in emotional memory research is how a traumatic event may result in intrusive, involuntary memories of that event – the core clinical feature of psychological disorders such as post-traumatic stress. Previous behavioural work shows that the strength of voluntary recall of traumatic events is not necessarily related to the frequency or vividness of the involuntary recall of those events, a dissociation that is not readily explained by mainstream episodic memory theories. Here, we investigated whether the two types of recall may be associated with distinct neural profiles at the time of encoding and shortly after. For this, 32 healthy participants underwent functional Magnetic Resonance Imaging while viewing clips of distressing events (the so-called ‘trauma film paradigm’), with periods of resting state in between consecutive clips. Next, we applied ‘multi-voxel correlation structure’ to assess the degree to which voxel-by-voxel connectivity profiles related to the encoding of a specific clip persisted during the immediately-following post-encoding rest period. In the hippocampus, higher similarity between post-encoding rest and encoding profiles, compared to immediately-preceding pre-encoding rest periods, predicted the voluntary recall (both verbal and visual recognition) of the distressing events, a week later. This neural profile was not however related to the frequency or variety of intrusive memories that participants recorded in a daily diary, which they kept for a week following film viewing. These findings corroborate behavioural observations and tentatively suggest that voluntary and involuntary memories of the same event may to some extent rely on separate neural systems.
Talk 3: A Late Slow Frontal Positivity ERP reflects the resolution of contextual ambiguity during narrative discourse comprehension
Patrick Ledwidge1, Adam Ramsey1, Jeremy Foust1; 1Baldwin Wallace University
The goal of this study was to identify the event-related potential (ERP) correlate(s) of contextual ambiguity resolution during narrative discourse comprehension. Thirty-seven neurotypical college-students read narrative discourses during 256-channel ERP recording. The contexts of the discourses were initially ambiguous if read without a descriptive title (Untitled Discourse group: n = 19), but clear if preceded by a title (Titled Discourse group: n = 18). For the Untitled Discourse group, however, the contexts became more clear after reading the last word of sentence 2 (Critical Word 2) and sentence 3 (Critical Word 3), as determined offline by a separate sample. ERPs of interest were the N400 and P600 recorded to the last word of each sentence (Critical Words 1-3). For the Untitled Discourse group, N400 amplitudes (272-452 ms) became less negative from Critical Words 1-3, suggesting greater ease of lexical-semantic retrieval as the discourses unfolded. P600 amplitudes (456-700 ms) increased from Critical Words 1-3 for the Titled Discourse group. Unexpectedly, amplitudes of a Slow Frontal Positivity (SFP) ERP (684-1000 ms) increased from Critical Words 1-3 for the Untitled Discourse group only. The SFP did not occur for the Titled Discourse group. P600 amplitudes for the Untitled Discourse group were larger to Critical Word 3 than Critical Word 2. Our results suggest that the SFP reflects the resolution or revision of contextual ambiguity during narrative discourse comprehension. Alternatively, the P600 is proposed to reflect discourse context-updating when an existing context is available or after the resolution/revision of contextual ambiguity.
Talk 4: Biasing Memory Replay During Sleep: A Quantitative Synthesis of Targeted Memory Reactivation Effects
Xiaoqing Hu1, Larry Y. Cheng2, Ken A. Paller2; 1The University of Hong Kong, 2Northwestern University
Many correlative findings suggest that memory reactivation during sleep contributes to memory stabilization. Compelling support for this idea has also been provided by studies using sensory input during sleep to directly manipulate neural processing (i.e., Targeted Memory Reactivation, TMR). Here we present a comprehensive meta-analysis of currently available TMR datasets (46 articles with 1,666 participants) to estimate overall effect size and identify possible boundary conditions with moderator analyses. Moderators included 1) stimulation state, sleep vs. wake; 2) stimulation sleep stage, N2 vs. N3 vs. N2/N3 vs. REM; 3) stimulation modality, auditory vs. olfactory vs. tactile; and 4) memory type, conditioning vs. declarative vs. skill memory. Our results revealed that overall, TMR is highly effective in modulating memory (Hedge’s g=0.328, 95% Confidence Interval [0.206, 0.447], Z=5.372, p<.001). Moreover, this effect is specific to sleep as TMR did not produce a significant effect during wakefulness, although this conclusion is tempered by the fact that only a small number of types of wakefulness were studied. For other moderators, TMR during N3, but not N2 or REM, produced robust effect sizes. As for stimulation modality, both auditory and olfactory stimulation produced significant effects while the only study using tactile stimulation failed to influence skill memory. Lastly, while all types of learning examined were influenced by TMR, conditioning was associated with the largest effect size, followed by skill memory and declarative memory. In sum, TMR during sleep, and particularly during slow-wave sleep, is effective for promoting memory reactivation and enhancing multiple types of memory.
Talk 5: Functional and structural connectivity of cognitive control networks during narrative comprehension from birth to 9 years
Rola Farah1, Tzipi Horowitz-Kraus1,2; 1Technion- Israel Institute of Technology, 2Cincinnati Childrens Hospital Medical Center
Background: The cingulo-opercular (CO) and fronto-parietal (FP) networks are prominent components of the brain’s cognitive control system. Evidence suggests that over the course of development, within network connectivity of brain regions supporting cognitive control functions increases, along with an increase in between networks connectivity, due to increased maturation along development. Characterizing the developmental changes in functional and structural connectivity of neural networks supporting cognitive control from birth may provide unique insights into the role of cognitive control in narrative comprehension during infancy. Methods: 88 children divided into three age groups: A)0-3 years-old, B)3-6 years-old and C)6-9 years-old participated in the study. Participants participated in a DTI session and a functional MRI session while listening to stories inside the scanner. Measures of graph theory, within and between the CO and FP networks functional connectivity and Fractional anisotropy (FA) in the Arcuate Fasciculus (AF) were assessed. Results: Developmental increase in within-and between -network connectivity of the CO and FP networks from birth to 9 years-old was observed, mainly between groups A and C. Global efficiency of the CO and FP networks also increased, and greater FA in the AF was observed in group C, compared to A and B in hubs connecting language and cognitive control networks. Conclusions: Increase in functional and structural connections within and between networks related to cognitive control during narrative comprehension in children was observed. Future studies investigating failures in language acquisition should explore the role of those networks in the process as well.
Talk 6: Orbitofrontal cortex integrates amygdala-hippocampal information and guides schema-based emotional categorization
Jie Zheng1, Jack J. Lin1,2; 1University of California, Irvine, 2Comprehensive Epilepsy Program, Irvine, CA
Repeated emotional experiences provide a cognitive map or mental “schema” to facilitate acquisition of new emotional information, shape individuals’ personalities, and guide interpersonal communications. The orbitofrontal cortex (OFC), hippocampus (HPC), and amygdala (AMY) are critical for schema formation, decision-making and emotional processing. However, despite anatomical links among these structures, the network oscillatory dynamics that support schema-based emotional processing in humans are unclear. We recorded intracranial electroencephalography from 7 pre-surgical epilepsy patients implanted with depth electrodes to characterize the OFC-HPC-AMY dynamics. During the task, participants were instructed to either categorize the face with the appropriate linguistic label of the emotion (e.g. sad, happy, afraid), which will elicit participants’ prior emotional knowledge (schema condition) or match the face with the one having the same emotional expression (non-schema condition). We observed faster reaction time (t-test, P = 0.013) and stronger low frequency synchrony (permutation test, P < 0.01) within the tripartite network for the schema compared to the non-schema condition. Moreover, the dimensionality of the network was reduced for the schema condition, with the OFC served as the hub modulating HPC and AMY high frequency activities at distinct theta phases for different emotional categories. For one patient with bilateral OFC lesions, impaired behavioral performance was observed only in the schema condition, with no dimensional reduction in the medial temporal lobe network. These findings provide causal evidence that the OFC serves to integrate new emotional information into existing schema by reducing the dimensionality of the space humans have to search to learn new information.
Talk 7: Everything you can imagine is real: Component processes and brain systems of imagination.
Darya Zabelina1, Jessica Andrews-Hanna2; 1University of Arkansas, 2University of Arizona
The ability to form abstract mental representations – the ability to imagine – plays an important role in a wide range of behaviors, including learning, empathy, psychopathology, and vocational training. Despite this, empirical research on imagination has historically been limited. Drawing from our previous work, this study uses neuroimaging techniques (fMRI) to examine neural basis of imagination. Participants either viewed images of faces and houses, or were asked to imagine entirely new faces and houses, not the ones they have previously seen. Results indicate that recruitment of imagination relies on the interaction between frontoparietal and default mode (DM) networks, while sustaining imagination relies on the activation in the regions of the DM network alone. Additionally we examined individual differences in the complexity of imagination, and found that people who report more complex imagination show increased activation in the frontoparietal control network regions (particularly in the pre-SMA and lateral PFC) while sustaining an imagined scene compared to people who report that their imagination is not very complex. Results are discussed in the context of existing literature on related processes, namely creative idea generation, and poetry evaluation, which are reported to similarly rely on dynamic interaction between the frontoparietal and the DM networks. Implications and future directions are discussed.
Talk 8: Neuroanatomical differences between monozygotic twins discordant for musical practice
Örjan de Manzano1, Fredrik Ullén1; 1Karolinska Institutet
Using primarily neuroimaging techniques and musical expertise as a model, scientists have studied the neural correlates of skill acquisition and built a convincing case for that musical training can cause brain regions to grow and/or become better developed. However, in arguably all previous studies, training differences were confounded with genetic predispositions. This presents an issue because firstly, brain anatomy is highly heritable in many regions. Secondly, we know that different individuals may require different amounts of practice to reach a certain skill level, and that despite similar amounts of practice there can still be individual differences in skill. Genetic factors could influence brain development in a way that affects self-selection, skill acquisition and achievement within a certain domain, including music. In order to resolve this issue, we have studied a sample of monozygotic (genetically identical) twins who differ greatly in musical training. Using neuroimaging, we found that that even when eliminating genetic influences as a causal factor, large differences in musical practice can nonetheless be associated with significant differences in brain structure. The playing twins had increased cortical thickness, white-matter fractional anisotropy and cerebellar volume in regions which jointly constitute the core of the brain’s auditory-motor network.
Talk 9: Combining eye-tracking and EEG to measure attention to salient and emotional stimuli
Louisa Kulke1,2,3, Janette Atkinson3,4, Oliver Braddick4, Annekathrin Schacht1,2; 1University of Göttingen, 2Leibniz-ScienceCampus Primate Cognition, 2University College London, 2University of Oxford
In everyday life, salient or emotionally relevant stimuli often catch our attention and elicit saccades. However, most previous neuroscientific studies instructed subjects to covertly shift attention, thereby suppressing natural saccades. This set of studies used a novel method combining eye-tracking and EEG to measure overt shifts of attention. Study 1 compared neural mechanisms of covert and overt attention shifts. Twenty-four participants performed an attention shift task in which they either manually responded to peripheral targets while maintaining fixation (covert) or made a saccade towards them (overt). EEG and eye-tracking were combined to simultaneously measure neural responses and saccades. Event-related potentials were similar for overt and covert shifts of attention; however, an early fronto-central component differed between condition, potentially reflecting saccade suppression during covert attention shifts. This suggests that natural attention shifts may differ from covert shifts often recorded in labs. Study 2 manipulated disengagement of attention by comparing overt attention shifts to a salient peripheral target, both with and without a competing central target. Results from 41 participants show that saccade latencies towards targets are significantly shorter when no competing stimulus is present. The same latency pattern occurred for early occipital responses with shorter latencies for responses requiring no disengagement. Study 3 additionally introduced emotional faces as stimuli to investigate effects of valence on attention shift latencies. The results provide insights to the mechanisms of attention shifts to salient and emotional stimuli in a natural context, providing information on the neural mechanisms underlying the attentional draw towards these stimuli.
Talk 10: Reduced Persistence of Spontaneous Brain Activity in Schizophrenia
Huang Zheng1,2, Jianbo Gao1,2; 1School of Computer, Electronics and Information, Guangxi University, China, 2Institute of Complexity Science and Big Data Technology, Guangxi University, China
Schizophrenia is a complex disorder comprised of widespread affective, cognitive, and behavioral disturbances and disruptions. We studied the dynamic properties of the blood oxygenation level-dependent (BOLD) time series of patients diagnosed with schizophrenia at resting state, by computing a key measure from complexity science, the Hurst exponent (H) using Adaptive Fractal Analysis (AFA). When 0<H<1/2, a BOLD time series is considered to have anti-persistent correlation. When H=1/2, a BOLD time series is memoryless or only has short-term memory. When 1/2<H<1, a BOLD time series is considered to have persistent long range correlation. 44 participants meeting the DSM-IV criteria for schizophrenia and 77 healthy controls underwent fMRI scanning at 3.0 T. Mean H for all the 116 brain regions extracted from Anatomical Automatic Labeling (AAL) template as well as the whole brain were generated and analyzed for each subject based on voxel by voxel time series. Statistically significant reduced persistences in brain activity were detected in schizophrenia patients compared to controls both at the whole brain level and the brain region level. To be exact, 103 of 116 brain regions showed significantly reduced persistence, suggesting overwhelmingly less persistent brain activity in patients with schizophrenia. The result provides evidence that brain activity of schizophrenia is associated with weaker persistence and more randomness in signal patterns compared to healthy controls. This property may constitute a potential biomarker of schizophrenia from Resting-State fMRI.
Talk 11: Hierarchical Neural Representations Behind Naturalistic ‘Social Norm’ Perception In Autism and Controls
Felipe Pegado1, Hans Op de Beeck1; 1KU Leuven
Humans show a unique capacity to integrate information from multiple sensory and higher-order cognitive systems. Here, instead of studying one level of representation at a time, we designed a fMRI paradigm aiming to capture multiple brain representations simultaneously, from low-level processing all the way up to higher-order social representations. Our goal was to address a controversial topic: In which levels of the brain hierarchy high-functioning autism (HFA) differs from controls ? By using a naturalistic audio-visual ‘social norm’ mentalizing task (infering how ‘most people’ would judge the appropriateness of vocal responses), and multivoxel pattern analysis (MVPA), we could investigate multiple levels of representations at once. Our results show striking similar neural patterns in both groups at low and high level visual and auditory processing, with two exceptions: more low level visual information in Precuneus (PC) and more heterogeneity (uniqueness) of neural representations in ‘Voice Area’ for auditory stimuli in HFA. Further we also found similar neural substrates for social information in both groups: PC, Temporo-Parietal Junction (TPJ), and posterior medial Prefrontal Cortex (mPFC) - a region often associated with allocentric mentalizing, but not in anterior mPFC, a region associated with self-related mentalizing. Despite that, at the behavioral level, HFA show much less ability to agree on the shared social norm representation, i.e., lower between-subjects correlation of response patterns across runs, while having the same level of intra-subject consistency (within-subjects correlations) as controls. These results shed light on the similarities and differences of HFA brain organization during ‘social norm’ inferences.
Talk 12: Mindfulness-Based Stress Reduction Improves Fear Extinction: An fMRI Investigation
Gunes Sevinc1,2, Britta Hölzel3, Muhammed Milad1, Sara W. Lazar1,2; 1Massachusetts General Hospital, Division of Psychiatry, 2Harvard Medical School, 2Technical University of Munich, Klinikum rechts der Isar
Mindfulness based stress reduction (MBSR) programs have been widely utilized to ameliorate psychiatric and stress-related symptoms, however the neural mechanisms that underlie the reported improvements are still largely unknown. Mindfulness meditation involves refraining from cognitive avoidance and thus provides a basis for internal exposure to aversive stimuli. Thus, we hypothesized that mindfulness-based interventions create a context akin to behavioral exposure therapy and thereby alter participants’ neurobiological responses to aversive stimuli. In a randomized controlled trial with healthy but stressed meditation-naïve individuals, we tested this hypothesis and investigated neural activation patterns during the recall of aversive stimuli using a well-established 2-day fMRI fear-conditioning and extinction protocol. Participants completed either 8-week MBSR (n=42), or stress-management education (SME, n=25) programs and behavioral changes and alterations in neural activation patterns from pre to post interventions were assessed. The MBSR intervention resulted in significant activations in left inferior frontal gyrus (BA 44) during the recall of extinguished (as opposed to unextinguished) stimuli, suggesting heightened control of memory interference, while SME intervention did not. Moreover, both interventions were associated with enhanced brain activity from pre to post in vmPFC (BA 11) and the hippocampus during extinction recall, and changes in hippocampus marginally correlated with changes in perceived stress levels only for the MBSR intervention. These results indicate extinction learning as a potential mechanism underlying the positive psychological benefits ascribed to mindfulness meditation. These findings may enhance our understanding of how meditation-based interventions work and the role of extinction learning in stress-resilience.
Talk 13: Inter-subject representational similarity analysis reveals individual variations in affective experience when watching erotic movies
Pin-Hao Andy Chen1, Eshin Jolly1, Todd F. Heatherton1, Luke J. Chang1; 1Dartmouth College
We spend much of our life pursuing or avoiding affective experiences. However, surprisingly little is known about how these experiences are represented in the brain and whether everyone has an identical experience. In this study, 26 male participants watched 3.5 minutes of erotic pornography and completed an 8-item dyadic sexual desire scale as well as a 13-item brief self-control scale. We were interested in whether individual variations in sexual desire and self-control might be associated with individual variability in participants’ experiences watching the erotic movies. We used inter-subject representational similarity analysis (ISRSA) to identify regions of the brain in which similarity in participants self-reported preferences were associated with similarity in temporal brain dynamics. This technique calculates a pairwise similarity matrix of participant brain dynamics in each ROI and maps these variations to a pairwise similarity matrix in responses to a self-report scale using rank correlation and permutation tests. We calculated ISRSA separately for 200 parcels from a whole-brain parcellation of co-activation patterns from over 10,000 published studies (Bonferroni corrected). We found that similarity in preferences for sexual desire was significantly associated with similarity in brain dynamics in the MPFC, PCC, and NAcc, indicating possible similarity in endogenous reward processing. In contrast, similarity in preferences for self-control was correlated with bilateral DLPFC, indicating variations in the executive control network. These results suggest that individuals have substantial variability in their affective experiences, and that ISRSA is a useful technique for identifying brain regions where this individual variability maps onto preferences.
Talk 14: Children Engage Semantic Processes to Verify Arithmetic Facts: Evidence from the N400
Amandine E. Grenier1, Vanessa Cerda1, Danielle S. Dickson1, Bianca O. Obinyan1, Jacob P. Momsen2,3, Nicole Y.Y. Wicha1; 1The University of Texas at San Antonio, 2University of California San Diego, 2San Diego State University
Arithmetic facts, like multiplication tables, are thought to be encoded into verbal memory, and children who develop a memory retrieval strategy perform better in math through high school. Semantic memory retrieval is usually indexed by a modulation of the N400 component. However, when adults verify the solution to simple multiplication problems (2x3=6 versus 2x3=7), correct solutions elicit a P300. The adult P300 may reflect the overlearning of math facts, wherein the correct solutions are interpreted as target items. This study measured ERPs in children during a similar multiplication verification task to examine cognitive processing in early stages of learning math facts. Third through fifth graders judged the correctness of multiplication problems. Each problem was presented one number at a time, as Arabic numerals with no symbols (e.g., 2 4 8). Half of the solutions were correct and half were incorrect (e.g., 2 4 12). Results show that both correct and incorrect solutions elicit an N400, with larger (more negative) amplitude for the incorrect solutions. This N400 pattern is similar to what is observed when people process the meaning of words in a sentence. No P300 was observed for correct solutions, indicating that the development of this skill continues after fifth grade. This ERP pattern may reflect that children rely more on semantic level processes than adults when verifying math facts.
Talk 15: Alpha Oscillatory Synchrony underlying Working Memory Maintenance in Children
Julie Sato1,2, Sarah Mossad1,2, Simeon Wong2, Benjamin Hunt2, Benjamin Dunkley1,2, Mary Lou Smith1,2, Margot Taylor1,2; 1The Hospital for Sick Children, 2University of Toronto
Working Memory (WM) is a critical ability that supports a wide range of cognitive functions, and is highly associated with general intelligence and academic achievement. Although functional MRI studies have highlighted a network of regions involved in WM in adults, little is known about how these networks develop in children to support successful WM performance. Using magnetoencephalography (MEG), we examined the dynamic network structures underlying the maintenance of visual information in WM in 6-year-old children. We observed stronger and more stable connectivity in the alpha frequency band (8 to 14 Hz) during the retention interval associated with correct compared to incorrect trials. Importantly, this effect was found to be specific to the alpha band, with no other frequency bands showing significant differences between correct and incorrect trials. This difference in connectivity may be attributed to fluctuations in attention that are necessary to sustain representations in WM. Additionally, our network analysis revealed elevated alpha synchronization during WM maintenance compared to baseline, in a distributed network of frontal, parietal, and temporal regions. Central hubs in the network were lateralized to the left hemisphere, including the middle frontal gyrus, middle temporal gyrus and superior temporal gyrus. Our results suggest that alpha inter-regional synchrony is an important mechanism for sustaining memory of visual stimuli and is already active in young children.
Data Blitz Session 3
Saturday, March 24, 12:00 - 1:30 pm, Grand Ballroom
Speakers: Colleen Schneider, Esti Blanco-Elorrieta, Alexander Belden, Kevin P. Madore, Ekaterina Denkova, James A. Brissenden, Kelly A. Vaughn, Giulia V. Elli, Sara Cadavid, Jennifer Zuk, Nora Preuss, Rebecca Cutler, Heather Bruett, Trevor Brothers, Benjamin N. Conrad
Talk 1: Neuroplastic and Neurovascular Contributions to Visual Recovery in Post-Stroke Cortical Blindness
Colleen Schneider1,2, Emily Prentiss2, Zoe Williams1, Bogachan Sahin1, Bradford Mahon1,2; 1University of Rochester School of Medicine and Dentistry, 2University of Rochester School of Arts Sciences and Engineering
The underlying mediators of post-stroke recovery remain poorly understood. Previous studies suggest that cortical reorganization and perilesional repair are mediators of post-stroke recovery; however, none of those studies have been able to provide a clear, generalizable association between those changes and recovery of function across patients. The present study investigates the relation between post-stroke visual recovery and i) perilesional neurovascular re-coupling, and ii) changes in retinotopic organization. Patients with visual field cuts secondary to stroke were studied longitudinally with 24-2 Humphrey perimetry and fMRI at 1 week, 1 month, 3 months, and 6 months post-stroke. Perimetry and functional MRI measures were brought into register with each other by dividing the visual field into 12 wedges that matched the area of the visual field stimulated by 12 checkerboard wedges during polar angle retinotopic mapping. The response of perilesional voxels to a full-field visual stimulus was significantly delayed compared to the contralesional hemisphere at 1 week but recovered by 3 months post-stroke in most patients. In addition, there was a shift in the preferred retinotopic bias of perilesional voxels towards the original blind field. Normalization of the delayed response of perilesional voxels accounted for visual recovery in patients who were initially blind in a hemifield and whose blind field resolved to one quadrant, while shifts in retinotopy toward wedge locations that were originally blind were associated with visual recovery in those wedge locations. We argue that the neurovascular and neuroplastic, in conjunction, explain variability in visual recovery across patients.
Talk 2: Task switching decomposed: MEG evidence from bimodal language switching
Esti Blanco-Elorrieta1,2, Karen Emmorey3, Liina Pylkkänen1,2; 1New York University, 2NYUAD Institute, 2San Diego State University
A defining feature of human cognition is the ability to quickly and accurately alternate between complex behaviors. This switching process minimally comprises of goal recognition, disengagement from the previous task and engagement in a new task. Previous studies have associated task-switching with prefrontal activity; however, it is unknown how the sub-computations of task switching individually contribute to these activities. This is largely because few natural tasks allow for full separation of disengagement and engagement processes during switching. Here we recorded magnetoencephalography (MEG) from ASL-English bimodal bilinguals who often sign and speak simultaneously. This offers a unique opportunity to dissociate engagement and disengagement. Our MEG recordings showed that turning a language “off” (switching from simultaneous to single language production) led to increased activity in the ACC and PFC, while turning a language “on” (from one language to two simultaneously) did not differ from non-switch trials. The distinct representational nature of the “on” and “off” processes was also supported by multivariate decoding analyses. Granger causality analyses additionally revealed that i) compared to turning-on, turning-off required stronger connectivity between left and right dlPFC and ii) dlPFC activity predicted ACC activity, consistent with models in which the dlPFC is a top-down modulator of the ACC. These results suggest that in language switching, the burden of task-switching lies in disengagement from the previous task as opposed to engagement in a new task, and that in the absence of motor constraints, producing two languages simultaneously is not more cognitively costly than only producing only one.
Talk 3: Hearing Creatively: Default Network Selectively Synchronizes to Auditory Cortex in Jazz Improvising Musicians
Alexander Belden1, Tima Zeng1, Emily Przysinda1, Psyche Loui1; 1Wesleyan University
Jazz improvisation offers a model for creative cognition as it involves the real-time creation of a novel, information-rich product. Previous research has shown that when jazz musicians improvise, they recruit medial and dorsolateral prefrontal cortices, which are part of the Default Mode (DMN) and Executive Control (ECN) Networks respectively. Here, we ask whether these task-fMRI findings might arise from intrinsic differences in functional connectivity. We compare resting state fcMRI of ECN and DMN among jazz improvisers, classical musicians, and non-musicians. We seeded regions of interest in the medial prefrontal cortex, within the DMN, and the dorsolateral prefrontal cortex from the ECN, and compared the correlation patterns from each ROI across the three subject groups (all results p<0.05 cluster-corrected). We found higher resting state connectivity in jazz improvisers than classical musicians and non-musicians between mPFC and the superior temporal gyrus, including the auditory cortex. In contrast, all musicians showed increased connectivity from left dorsolateral prefrontal cortex, a region of the left ECN, to bilateral superior parietal lobule; this was especially higher in classical musicians compared to non-musicians. Results show that long-term training enhances functional connectivity in specific resting state networks. While general musical training is associated with executive control functions, the finding that the Default Mode Network is more synchronized with auditory regions in jazz improvisers is consistent with the hypothesis that real-time musical creativity relies on auditory access to spontaneous thought processes.
Talk 4: Neural mechanisms of episodic retrieval support divergent creative thinking
Kevin P. Madore1, Preston P. Thakral2, Roger E. Beaty2, Donna Rose Addis3, Daniel L. Schacter2; 1Stanford University, 2Harvard University, 2University of Auckland
Prior research has indicated that brain regions and networks that support semantic memory, attention, and cognitive control are all involved in divergent creative thinking. Kernels of evidence suggest that neural processes supporting episodic memory – the retrieval of particular elements of prior experiences – may also be involved in divergent thinking but such processes have typically been characterized as not very relevant for, or even a hindrance to, creative output. In the present study, we combine functional magnetic resonance imaging with an experimental manipulation to test formally, for the first time, episodic memory’s involvement in divergent thinking. Thirty-two young adults completed a within-subjects design, and task analyses focused on whole-brain univariate regression models and multivariate independent component analyses. Following a manipulation that facilitates detailed episodic retrieval, greater neural activity in the hippocampus and stronger connectivity between a core brain network linked to episodic processing and a frontoparietal brain network linked to cognitive control were observed during divergent thinking relative to an object association control task that requires little divergent thinking. Stronger coupling following the retrieval manipulation extended to a subsequent resting-state scan with univariate seed-to-voxel connectivity. Neural effects of the episodic manipulation were consistent with behavioral effects of enhanced idea production on divergent thinking. The results indicate that conceptual frameworks should accommodate the idea that episodic retrieval can function as a component process of creative idea generation, and highlight how the brain flexibly utilizes the retrieval of episodic details for tasks beyond simple remembering.
Talk 5: In Search of Mind Wandering: Dynamic Functional Connectivity during Rest and Task
Ekaterina Denkova1, Jason S. Nomi1, Shruti Gopal Vij1, Lucina Q. Uddin1, Amishi P. Jha1; 1University of Miami
Mind wandering (MW), defined as self-generated thinking disengaged from the external environment, has become a prominent topic in neuroscientific research. Emerging evidence suggests a critical role of three brain networks in MW: the default network (DN), the central executive network (CEN), and the salience network (SN). Advances in analytical methods for functional neuroimaging data (i.e., dynamic functional connectivity, DFC) demonstrate that the interactions among these networks are not static, but dynamically fluctuate over time. The majority of prior evidence comes from investigations focusing on the task-free resting state, which is often assumed to reflect unconstrained MW. Yet, it remains unclear if tasks characterized by fluctuations between attention and MW involve similar patterns of DFC. The present study applies the DFC method to neuroimaging data collected from 30 participants who completed a resting state scan followed by two sessions of a sustained attention to response task with embedded questions assessing MW. A DFC analysis revealed five dynamic network states common to both rest and task. Differences between rest and task appeared in the frequency of occurrence of two states. One state was more frequent during rest and characterized by weak connectivity between the three networks. The second state was more frequent during task and characterized by a negative correlation between the DN and CEN and SN, as well as a positive correlation between the CEN and SN. These findings suggest that some patterns of dynamic intercorrelations between networks may vary as a function of cognitive state.
Talk 6: Visual Field Representations in Human Cerebellum
James A. Brissenden1, Sean M. Tobyne1, David E. Osher2, Emily J. Levin3, Mark A. Halko4, David C. Somers1; 1Boston University, 2Ohio State University, 2Brown University, 2Harvard Medical School and Beth Israel Deaconess Medical Center
Maps of the visual field have been identified throughout the cerebral cortex. Visual field maps located in higher-order association areas such as the intraparietal sulcus and frontal eye fields are of particular interest as maps in these regions may serve as a critical substrate for spatial cognition. Recently, we showed that cerebellar lobules VIIb/VIIIa, which exhibit intrinsic connectivity with cortical attention areas, are recruited in a load-dependent fashion by visual attention and visual working memory tasks (Brissenden et al., JNeurosci., 2016). While somatotopic maps in the cerebellum are well characterized, cerebellar visual field maps have not been reported. To investigate whether visual field representations exist in cerebellum we conducted two fMRI experiments. In the first experiment, subjects performed an attentionally demanding visual field mapping task. A population receptive field (pRF) analysis found representations of the ipsilateral visual hemifield within lobules VIIb/VIIIa, consistent with the crossing of cortico-cerebellar fiber tracts. In the second experiment, subjects performed a lateralized visual working memory task. Stimuli were presented bilaterally and subjects were asked to covertly attend items in one visual hemifield. To determine whether the cerebellum is sensitive to the locus of attention, we trained support vector machines on cerebellar activity patterns to discriminate between the two possible locations of attentional deployment (left or right hemifield). A multivariate feature weight mapping analysis revealed clusters of informative voxels within lobule VIIb bilaterally. These results demonstrate the existence of visual field representations in human cerebellum and indicate that cerebellar lobules VIIb/VIIIa participate in visuospatial cognition.
Talk 7: Managing two languages relates to managing two goals: fMRI evidence from task-switching
Kelly A. Vaughn1, Arturo E. Hernandez1; 1University of Houston
Whether bilingualism is related to cognitive control in non-verbal tasks is a topic of debate among researchers (e.g., Paap, 2016). The current study used fMRI to compare brain activity in Spanish-English bilingual and English monolingual young adults during a shape-color switching task, in which each non-verbal switch or repeat cue was followed by 8-12 trials, and then another cue. The goal of the study was to understand whether task performance or fMRI activity was related to bilingual status. Bilinguals responded significantly more slowly and accurately than monolinguals across all trials. fMRI activity was analyzed at an FWE-corrected alpha of 0.05 for the switch > repeat and repeat > switch contrasts for the cues and for the trials. Monolinguals had the most significant fMRI activity in cognitive control regions (i.e., frontal cortex, anterior cingulate cortex, inferior parietal lobule, and basal ganglia) for the switch > repeat cues contrast, with additional significant activity in these regions during the trials. Bilinguals showed less significant fMRI activity than monolinguals for each task comparison. Overall, these findings lead to the conclusion that bilinguals handle a non-verbal switching task by monitoring conflict throughout the task and controlling speed to improve accuracy, whereas monolinguals handle the same task by responding to each cue as it comes and sacrificing accuracy for speed. This study suggests that bilingual experience, which may involve frequent monitoring of two languages, relates to successfully monitoring two task goals; monolingual language use, conversely, relates to a focus on one task goal at a time.
Talk 8: The neural basis of verb and noun semantic representations in congenitally blind individuals
Giulia V. Elli1, Rashi Pant1, Rebecca Achtman2, Marina Bedny1; 1Johns Hopkins University, 2DePauw University
How are the meanings of words influenced by sensory experience? We used multi-voxel pattern analysis (MVPA) to compare the neural basis of lexical semantic representations in congenitally blind (N=15) and sighted individuals (N=13). Specifically, we asked how noun- and verb-responsive cortical regions encode semantic distinctions among words within a grammatical category. Participants judged the similarity of pairs of nouns (birds, mammals, man-made places, natural places) and verbs (light emission, sound emission, hand action, mouth action). In each group, we identified regions in the left hemisphere that respond preferentially to nouns – inferior parietal lobule (IP), precuneus (PC) and inferior temporal cortex (IT) – and to verbs – middle temporal gyrus (MTG). A linear support vector machine (SVM) classifier was trained to decode among verbs and among nouns on half of the data, and then tested on the other half (e.g. even/odd runs). In PC, IP and MTG, classification was successful among verbs and among nouns in both groups (p's<0.05). Furthermore, blind and sighted individuals showed similar grammatical class effects: better classification for verbs in MTG and for nouns in IP and PC (WordClass x ROI: F(2,52)=15.38, p<0.000, Group x WordClass x ROI: F(2,52)=1.12, p=0.34). However, decoding among nouns in IT was successful only in sighted participants (Group main effect: F(1,26)=8.19, p=0.008). These results suggest that the lexical-semantic network is largely unchanged in blindness. However, inferior temporal areas that preferentially process concrete object nouns in sighted individuals appear to be less relevant for such processing in those who are born blind.
Talk 9: What happens in the human brain when explicit warnings reduce false memories?
Sara Cadavid1, M. Soledad Beato2, Mar Suarez2; 1Universidad del Rosario, Colombia, 2Universidad de Salamanca, Spain
We analyzed the effects of warnings on false recognition (FR) employing Deese/Roediger-McDermott paradigm. In this paradigm, words associated to a non-presented critical lure are studied and, subsequently, critical lures are often falsely remembered/recognized. We collected behavioral data, and, furthermore, we used event-related potentials (ERPs) to understand what happens in the human brain when this memory task is performed with warnings. ERPs were obtained to study frontal FN400 (300-500 ms), left-parietal (500-800 ms), and late right-frontal (1000-1500 ms) old/new effects (associated with familiarity, recollection, and monitoring processes, respectively). First, at the behavioral level, although it was not possible to eliminate FR in Warning condition, as expected, FR was higher in the No-Warning condition. Second, the ERP results regarding the FN400 old/new effect showed similar patterns for true recognition and FR. Therefore, true recognition and FR seemed to share common underlying familiarity-based processes. Differences between Warning and No-Warning conditions were only observed on true recognition patterns, and only in this particular epoch: Warning-condition participants presented a more pronounced familiarity-related effect. This outcome suggests that warnings actually led to strategic encoding. ERP similarities between true recognition and FR disappeared when recollection processes were examined, because only true recognition presented a left-parietal old/new effect. Later, with the onset of monitoring processes, true recognition and FR waveforms presented, again, clearly similar patterns, showing both a late right-frontal old/new effect associated with post-retrieval monitoring processes. Together, findings suggest that, even when participants are provided with warning instructions, true recognition and FR share some common underlying processes.
Talk 10: Relationships between white matter in infancy and subsequent language abilities in preschool
Jennifer Zuk1,2, Michael Figuccio1, Xi Yu1, Joseph Sanfilippo1, Jade Dunstan1, Clarisa Carruthers1, Ellen Grant1,2, Nadine Gaab1,2,3; 1Boston Children's Hospital, 2Harvard Medical School, 2Harvard Graduate School of Education
A rich body of evidence has identified the neural basis for language that is already present before birth, and longitudinal studies have found that behavioral and neural responses to language-related stimuli in infancy predict language abilities in preschool. To date, neuroimaging studies have predominantly employed electrophysiological methods with infants to characterize brain activity, whereas brain structure and its relation to subsequent language development remains understudied. Therefore, the present study investigates how brain structure in infancy relates to emerging language abilities in preschool. This study draws from an ongoing longitudinal investigation of infants with and without familial risk for dyslexia. Initially, structural neuroimaging was successfully acquired with infants (ages 4-18 months) using a natural sleep technique. Automated Fiber Quantification was employed to estimate white matter properties of language-related tracts from diffusion weighted images. Infants were then longitudinally enrolled and reinvited for follow-up assessment in preschool. To date, 25 follow-ups (mean age: 5.5 yrs, age range: 4–6.5 yrs) have completed a comprehensive language evaluation. Longitudinal analyses establish significant relationships between (i) the left arcuate fasciculus in infancy and vocabulary knowledge in preschool, and (ii) the posterior corpus callosum in infancy and phonological awareness abilities in preschool. Preliminary findings suggest that properties of white matter in language-related tracts predict language abilities in preschool. This research has the potential to uncover white matter properties in infancy that underlie the developmental trajectory of typical and atypical language development in early childhood, and further consider the role of early brain structure in shaping subsequent language outcomes.
Talk 11: Full-body ownership illusion elicited by visuo-vestibular integration
Nora Preuss1, Henrik Ehrsson1; 1Karolinska Institutet
Vestibular signals allow us to maintain balance and to orient ourselves in space. Integrating this input with information from our five senses contributes to our human self-consciousness and helps us to distinguish ourselves from the external world. The aim of the present study was to investigate how vestibular information contributes to the experience of body ownership using multisensory integration. We conducted three studies using a ‘full-body ownership illusion’ induced by virtual reality technology and galvanic vestibular stimulation, the latter a technique that allows for selective stimulation of vestibular afferents. Participants wearing head-mounted displays saw a mannequin’s body from first-person perspective that was performing a slow swinging movement. At the same time participants were exposed to galvanic vestibular stimulation that elicited vestibular sensations of swinging whole-body movements in the corresponding direction. Perceived ownership of the seen body was measured using questionnaire ratings and skin-conductance responses to a knife threat towards the mannequin. We demonstrated that when participants were exposed to congruent visuo-vestibular information, they perceived stronger ownership for the mannequin’s body compared to unimodal visual and vestibular conditions, a bimodal noise control condition, or an incongruent visuo-vestibular condition. The findings show for the first time that visuo-vestibular congruency is sufficient to induce illusory body ownership.
Talk 12: Searching for semantic knowledge: A vector space semantic analysis of the feature generation task
Rebecca Cutler1, Nate Klooster2, Melissa Duff1, Sean Polyn1; 1Vanderbilt University, 2University of Pennsylvania
A recent neuropsychological study found that amnesic patients with hippocampal damage (HD group; n=5) reliably produce fewer responses than healthy demographically matched comparison participants (NC group; n=15) in a semantic feature generation task (Klooster & Duff, 2015), consistent with the idea that semantic cognitive deficits are associated with hippocampal damage. Participants were presented with a target word and asked to produce as many features of that word as possible. We examined these data using a semantic vector space model to characterize semantic similarity between the target word and the response words comprising the features. Overall, HD patients generated features that were closer in semantic space to the target word, as compared to the wider range of features generated by the NC group. Both groups tended to initiate the search process with features close in semantic space to the target word, with a gradual decline in similarity to the target word over the first several responses. Adjacent features in the response sequence showed stronger similarity to each other than to non-adjacent features, suggesting that the search process follows a local trajectory in semantic space. Termination of feature generation in HD patients, but not the NC group, was preceded by a steady decline in similarity to the target word, suggesting that a patient’s search process is disrupted when the search ranges too far in semantic space from the target word. We consider these results in terms of a model in which hippocampus supports the probing of semantic memory.
Talk 13: The Role of Inter-region Information Synchrony in Processing Visual Stimuli
Heather Bruett1, Marc Coutanche1; 1University of Pittsburgh
The brain processes the many aspects of visual stimuli via the coordinated activity of a number of relevant regions. The processing targets of these regions can be uncovered by “decoding” multivoxel activity patterns, which can represent subtle distributed information. An approach that examines the timeseries of pattern discriminability – informational connectivity – can help determine which regions contain information in the same trials – in other words, which regions are acting in synchrony. I will present fMRI data that were analyzed via multivariate analysis tools and informational connectivity to determine how information synchrony plays a role in processing scenes and objects. We ask how regions within the scene and object processing networks can decode scenes and objects from “pseudo-scenes,” which contain certain elements present in typical scenes but lack other visual components. We find that the strength of informational connectivity within these networks differs based on the object or scene discriminations examined. The findings are particularly methodologically interesting, as they suggest that informational connectivity can identify functionally-relevant networks with greater specificity than can other measures, such as functional connectivity and psychophysiological interaction (PPI).
Talk 14: Two late positivities during sentence comprehension: The influence of wrap-up and cognitive control
Trevor Brothers1,2, Eddie Wlotko3, Simone Riley1, Margarita Zeitlin1, Connie Choi1, Gina Kuperberg1,2; 1Tufts University, 2Massachusetts General Hospital, 2Moss Rehabilitation Research Institute
During sentence comprehension, context can influence both the initial access of word meanings (300-500ms) and more prolonged, re-interpretive processes (500-1200ms). Here, we investigated the cognitive mechanisms underlying two late post-N400 positivities, which are known to vary as a function of plausibility and lexical constraint. In this study (N = 70), we measured ERPs as participants read sentences with predictable, unpredictable, or anomalous critical words (“Father carved the turkey with a knife/smile/beach…”). For 33 participants, critical words appeared in the sentence-final position, where “wrap-up” effects are maximal. For 37 participants, a few words were added to each sentence to delay sentence wrap-up. Finally, to probe the role of cognitive control in generating these late ERP components, we used the AX Continuous Performance Task (AX-CPT) to assess individual differences in cognitive control abilities. Unpredicted words produced a larger late anterior positivity, and anomalous words produced a larger late posterior positivity, relative to predictable words. Critically, the amplitude of the anterior positivity was larger for critical words appearing in sentence-final positions. This pattern was reversed for the late posterior positivity, with larger anomaly responses occurring in sentence-medial positions. Finally, improved performance on the AX-CPT appeared to selectively enhance the magnitude of the late anterior positivity, suggesting that the re-interpretation of unexpected (but plausible) events may depend on frontally-mediated cognitive control abilities.
Talk 15: Network Topology of Symbolic and Nonsymbolic Number Processing: A 7T fMRI Study
Benjamin N. Conrad1, Eric D. Wilkey1, Gavin R. Price1; 1Peabody College, Vanderbilt University
There is a longstanding debate regarding the extent to which symbolic (e.g. Arabic digits) and nonsymbolic (e.g. dot arrays) numbers engage shared versus distinct neural mechanisms. Previous functional magnetic resonance imaging (fMRI) studies have almost exclusively assessed regional activation with only a few evaluating functional connectivity via psychophysiological interaction analysis. Graph theory provides a methodological framework for describing the network architecture of the brain, with potential to shed new light on the symbolic/nonsymbolic number debate. Here, we employ graph theoretical measures to assess the networks engaged during symbolic and nonsymbolic number processing. We conducted an event-related 7T fMRI study with healthy subjects (n=40, 19.5 ± 0.9 years). Participants performed a number comparison task in which they decided whether Arabic digits or dot arrays were more/less than five. To assess task-related functional connectivity, we performed beta series correlations. Average beta series were extracted from a whole-brain, 246 region atlas and separated based on condition. Connectivity matrices were constructed for each subject/condition and regions were partitioned into functional modules according to consensus clustering. For both conditions, six modules were delineated including a fronto-parietal, default mode, visual, sensorimotor, temporal and subcortical network. 26 of 246 regions were assigned to different modules between conditions. These differences were primarily characterized by a more distributed fronto-parietal network in the nonsymbolic condition, with 19 additional regions from parietal, frontal, and temporal areas being included in this network compared to the symbolic condition. Our findings suggest organizational differences in whole-brain functional network architecture between nonsymbolic and symbolic processing.