12:00 - 1:30 pm
12:00 - 1:30 pm
12:00 - 1:30 pm
Data Blitz Session 1
Saturday, March 23, 12:00 - 1:30 pm, Bayview Room
Speakers: Irene van de Vijver, Hio-Been Han, Dongwei Li, Michael Freund, Joshua J. Volponi, Florian Fiebig, Rita Loiotile, Joshua D. Koen, Julia W. Y. Kam, Arianna N. LaCroix, Tanya Wen, Aaron Kucyi, Eriko Matsumoto, Phil Witkowski, Kristen Warren
Talk 1: Corticostriatal white-matter tracts supporting habitual behavior in the lab and in real life
Irene van de Vijver1, Aukje Verhoeven1, Sanne de Wit1; 1University of Amsterdam, the Netherlands
Goal-directed and habitual performance on an outcome-devaluation paradigm (the Slips-of-Action Task; SoAT) have been related to distinct frontostriatal pathways, between caudate nucleus and ventromedial prefrontal cortex, and between putamen and premotor cortex, respectively. Our first aim was to replicate this neural distinction in a larger sample, and our second aim was to extend this investigation to real-life habit formation. 207 young adults underwent structural MRI and performed the SoAT. We correlated individual differences in striatally-seeded white-matter tract probabilities with the tendency towards habitual behavior, or ‘slips’ of action. Preliminary results indicate no correlation with the previously reported frontostriatal pathways. Rather, connectivity between the striatum and sensorimotor, parietal, and occipital cortices predicted higher levels of goal-directed performance. Next, a subset of 65 participants was instructed to take a (placebo) pill every day for two weeks, and to report on the experienced automaticity. Automaticity was related to the number of pills successfully taken, and, interestingly, participants reporting higher automaticity also showed greater habit tendencies on the SoAT. The number of pills successfully taken correlated positively with tracts between the striatum and bilateral frontopolar cortex, in line with its supposed role in prospective memory. After correcting for the number of pills taken, automaticity correlated negatively with tracts between the striatum and bilateral insula and anterior cingulate cortex, areas that have been related to self-awareness. These results suggest that depending on the context, habitual behavior is subserved by a diverse network of brain areas, the co-operation of which requires further investigation.
Talk 2: Functional dissociation of EEG theta rhythms between prefrontal and visual cortices and their synchronization during sustained attention
Hio-Been Han1,2, Ka Eun Lee1,3, Jee Hyun Choi1,4; 1Korea Institute of Science and Technology, 2Korea Advanced Institute of Science and Technology, 2Seoul National University, 2Korea University of Science and Technology
Previous literature has identified the importance of a coherent oscillatory network between distal brain regions contributing to task performance, such as the fronto-visual theta (4-12 Hz) network. However, the conceptual framework has been challenged by recent evidence which suggests region-specifically different functions of cortical theta rhythms, especially those in the prefrontal cortex (PFC) and the visual cortex (VC). Because of their opposite functional correlates against attention, as PFC theta correlates while VC theta anti-correlates with attention, the underlying principle behind their functional network during sustained attention remains poorly understood. Here, we show that PFC/VC theta are different neuronal entities having distinct functional correlates, so that their connectivity increases during sustained attention when one (i.e., visual) is suppressed. Analyzing EEG signals from mouse brain during Go/No-Go task, we found PFC (VC) theta was stronger (weaker) in good-performance epochs than in bad-performance epochs. Interestingly, the synchronization between two oscillations increased during the epochs with good performance, despite the suppression of theta in VC. The fronto-visual phase-coherent theta network showed a predominant posterior-to-anterior direction on the order of few milliseconds. Along with the improved synchrony, the delay showed a subtle but systematic decrease, suggesting a boost of information relay from the posterior to the anterior brain region. Our findings not only provide empirical evidence for the distinction between the theta of PFC and VC, but also reveal the overlooked aspect of long-range synchrony between functionally different oscillators in the cerebral cortex.
Talk 3: Interaction between spatial attention and visual working memory from alpha oscillation and sustained potentials
Dongwei Li1, Chenguang Zhao1, Jialiang Guo1, Bingkun Li1, Qinyuan Chang1, Yulong Ding2, Yan Song1; 1Beijing Normal University, Beijing 100875, China, 2Sun Yat-Sen University, Guangzhou 510275, China
Selective attention helps us filter irrelevant information and allows limited relevant information into visual working memory for accomplishing the current task. We aimed to investigate the roles of alpha modulation and contralateral delay activity (CDA) during the retention of visual working memory. EEG signals were collected from 32 healthy adults while they performed a classical working memory task. We observed a sustained posterior CDA (351-1001ms) was presented and scaled with lateral working memory loads (one or two targets) but absent when the targets were presented across two visual hemifields. A strikingly same pattern emerged for modulations in the averaged alpha (8–12Hz) power (354-988ms). However, the alpha modulation in the early period of memory retention (354-600ms) was similar for lateral low- and high-load conditions, suggesting no difference in storage. In contrast, the alpha modulation was much larger in lateral high-load than low-load condition in the late period of memory retention (600-988ms). More importantly, the N2pc, a well-known electrophysiological index of attentional selection induced by the remembered targets in the following visual search task, showed the inverse gradations as compared with the late alpha modulation. Our results provide neurophysiological evidence for the dissociable roles of alpha oscillation and CDA during visual working memory, suggesting that CDA scales online maintaining of information but alpha oscillation reflects the allocation of visuospatial attentional resources and significantly influences the subsequent attentional selection of targets measured by N2pc.
Talk 4: A pattern-similarity analysis approach to cognitive control in color-word Stroop.
Michael Freund1, Todd Braver1; 1Washington University in St. Louis
Classic experimental paradigms such as Stroop have been instrumental for testing theories of cognitive control. Typically, brain imaging studies of such tasks have used univariate analysis approaches, with multivariate pattern analysis methods infrequently adopted. Here, we explore the utility of pattern similarity analysis for identifying dissociable components of neural coding in a spoken color-word Stroop fMRI study (N = 37). Three task dimensions were tested with three representational models: target (color/correct response), distractor (word/incorrect response), and congruency (target==distractor?). Analyses were conducted at the areal level, using an anatomically constrained cortical atlas recently developed for the Human Connectome Project (Glasser/Multi-Modal Parcellation). Our distractor model was selectively correlated with similarity structures from early visual cortex (V1, V2), whereas our target model was correlated with structures widely distributed across the brain, but strongest in sensorimotor areas. In contrast, our congruency model best described similarity structures in fronto-parietal and cingulo-opercular regions, including dorsolateral prefrontal (dlPFC), superior and inferior parietal, and anterior cingulate cortex. Supporting the functional relevance of these measures, individual differences in model fits were associated with enhanced behavioral performance. Specifically, stronger distractor decoding was associated with a smaller Stroop effect in early visual cortex, but a larger effect in perisylvian regions, possibly reflecting strength of stimulus encoding and depth of distractor processing. Conversely, stronger target decoding in dlPFC was associated with fewer Stroop errors, potentially reflecting better representation of task goals. These results highlight the potential of pattern similarity techniques for decomposing classic cognitive control tasks from a representational perspective.
Talk 5: Neural and behavioral transfer of a simultaneous cognitive-physical video game intervention in an older adult population
Joshua J. Volponi1, Alexander J. Simon1, Alana B. Colville1, Samirah V. Javed1, Brigid J. Larkin2, Karam K. Samplay1, Soo M. Park1, Jessica N. Schachtner1, Roger Anguera1, Christian J. Thompson2, Joaquin A. Anguera1, Adam Gazzaley1; 1UCSF, 2USF
Cognitive and physical interventions can have positive effects on cognitive control abilities (e.g. attention, working memory, goal management). Here we deployed a novel training video game (“Body-Brain Trainer: BBT”) for older adults that combines cognitive and physical features to assess the effects of transfer to cognitive outcome measures. During training both cognitive and physical task difficulty adapted on a trial-by-trial basis (through cognitive performance metrics and real-time heart rate measurements, respectively). In this study, cognitively normal older adults (n=24; 55-85 years of age) were asked to train for 24 hours over eight weeks, with a battery of cognitive outcome measures assessed prior to and following this training period. An age and expectancy-matched group of active controls (n=25) played a set of three placebo applications comprised of both cognitive and physical aspects. Evidence of behavioral transfer included a group X session interaction suggesting differential improvements in the training group’s impulsivity response time [p=0.017] and response time variability [p=0.033] compared to the control group. In the same assessment, the training group also showed a significant neural group X session effect through an increase in midline frontal theta compared to the control group [p=0.031] as well as a significant session effect [p=0.015]. These findings demonstrate the potential of transfer of improvements in cognitive control from a combined cognitive and physical intervention to an assessment in a different cognitive domain both in a behavioral and neural fashion.
Talk 6: An Indexing Theory for Working Memory based on Fast Hebbian Plasticity
Florian Fiebig1, Pawel Herman1, Anders Lansner1,2; 1Lansner Laboratory, Department of Computational Science and Technology, KTH Royal Institute of Technology, 10044 Stockholm, Sweden, 2Department of Mathematics, Stockholm University, 10691 Stockholm, Sweden
Working memory (WM) is a key component of human memory and cognitive function. Computational models have been used to uncover the underlying neural mechanisms. However, these studies have mostly focused on the short-term memory aspects of WM and neglected the equally important role of interactions between short- and long-term memory (STM, LTM). Here, we concentrate on these interactions within the framework of our new computational model of WM, which accounts for three cortical patches in macaque brain, corresponding to networks in prefrontal cortex (PFC) together with parieto-temporal cortical areas. In particular, we propose a cortical indexing theory that explains how PFC could associate, maintain and update multi-modal LTM representations. Our simulation results demonstrate how simultaneous, brief multi-modal memory cues could build a temporary joint memory representation linked via an “index” in the prefrontal cortex by means of fast Hebbian synaptic plasticity. The latter can then activate spontaneously and thereby reactivate the associated long-term representations. Cueing one long-term memory item rapidly pattern-completes the associated un-cued item via prefrontal cortex. The STM network updates flexibly as new stimuli arrive thereby gradually over-writing older representations. In a wider context, this WM model suggests a novel explanation for “variable binding”, a long-standing and fundamental phenomenon in cognitive neuroscience, which is still poorly understood in terms of detailed neural mechanisms.
Talk 7: Visual cortex activity during non-visual tasks is “cross-modal” in late but not congenital blindness.
Rita Loiotile1, Marina Bedny1; 1Johns Hopkins University
Studies of blindness provide a window into the timecourse of plasticity. In congenitally blind individuals, “visual” cortices are highly active during auditory and tactile tasks (Wanet-Defalque et al., 1988; Sadato et al., 1996). Some “cross-modal” activity is observed even in late blind and blindfolded sighted adults (Merabet et al., 2008, Burton et al., 2002). Do visual cortices support similar cognitive functions in these populations, albeit to different degrees? Alternatively, does cortex have distinctive cognitive potential early in life? Sighted (N=21), congenitally (N=21) and late blind (N=9; age of vision loss >= 17y) participants performed an auditory go-no/go task, while undergoing fMRI. Participants made fast button presses (within 900ms) in response to two “go” sounds (75% of all trials, frequent-go 50%, infrequent go 25%) and withheld responses to infrequent “no-go” sounds (25% of all trials). Resting state data were also collected. We observed a right-lateralized response-inhibition effect (no-go > infreq-go > freq-go) in the “visual” cortices of congenitally but not late blind or sighted groups and in the inferior frontal cortices of all groups. Crucially, “visual” cortices of sighted and late but not congenitally blind groups responded like sensorimotor cortices: all go i.e. with button press > no-go trials. We hypothesize that visual cortices are more functionally tethered to non-visual sensory-motor systems in sighted and adult-onset blind than in congenitally blind individuals. “Visual” cortices assume higher-cognitive functions in congenital blindness, but show “cross-modal” responses in late blindness.
Talk 8: An own-age bias in the hippocampus in young and older adults
Joshua D. Koen1, Nedra Hauck2, Michael D. Rugg2; 1University of Notre Dame, 2University of Texas at Dallas
The present study examines the neural correlates of the own-age bias – differential neural activity for age congruent (same age) relative to age incongruent (different age) faces. Own-age biases are present in recognition memory tasks for faces, as well as in ERP correlates of face processing (e.g., the N170). Here, we examined own-age biases while healthy young and older adults viewed unfamiliar young and older adult faces. A subset of the faces was repeated immediately (lag 0) or after a delay (lag 11), and participants were to report (via button press) when an occasional inverted face was presented. Face ‘repetition suppression’ effects were most prominent in canonical regions previously implicated in face processing (e.g., FFA) after the short lag. These effects were larger for older than younger adults in the FFA, but there was no evidence of an own-age bias in face repetition suppression. The only cluster demonstrating a significant (p<.05, FWE, k=5) own-age bias effect was located in the right anterior hippocampus. Whereas images of old adult faces elicited greater hippocampal activity than young faces in older participants, this difference was reversed in young adults. This finding is striking given the importance of the hippocampus in episodic memory, and prior behavioral evidence of an own-age bias in recognition memory for faces. We conjecture that the increased hippocampal activity for same relative to different age faces reflects differential engagement of neural processes supporting the episodic encoding of faces, and might provide a mechanism underlying own-age biases in face memory.
Talk 9: Human Frontal Cortex Modulates External and Internal Attention
Julia W. Y. Kam1, Randolph F. Helfrich1, Jack J. Lin2, Anne-Kristin Solbakk3,4, Tor Endestad3, Pal G. Larsson4, Robert T. Knight1; 1University of California – Berkeley, 2University of California – Irvine, 2University of Oslo, 2Oslo University Hospital
Prevailing theories of top-down control converge on the role of the frontal cortex in facilitating selective attention to behaviorally relevant external inputs. However, the neural architecture supporting sustained attention directed to either the external or internal environment is less well understood. To address this, we used intracranial EEG to characterize the spatiotemporal dynamics of external and internal attention during an auditory target detection task. Participants directed their attention externally to auditory tones and responded to infrequent target tones, or internally to their own thoughts while ignoring the tones. We compared high frequency band activity (HFA; 70-150Hz) in response to target and standard tones across the lateral, parietal and temporal cortices during external and internal attention. Increased HFA to target tones relative to standard tones, referred to as the target detection response, was present across all three cortices during both attention states. Notably, a larger percentage of frontal electrodes compared to temporal electrodes showed this target detection response during external attention. Direct comparisons between attention states revealed only the frontal cortex showed larger HFA response during external relative to internal attention, confirming that the frontal cortex was more sensitive to attentional modulations compared to other cortices. Taken together, these results provide evidence that the frontal cortex plays an important role in the top down control of attention to both the external and internal environments.
Talk 10: Alerting, orienting, and executive control: post-stroke effects of attention abilities on speech comprehension
Arianna N. LaCroix1, Corianne Rogalsky1; 1Arizona State University
Cognitive deficits often accompany language impairments post-stroke. Most past research has focused on working memory in aphasia, but attention is largely underexplored; to our knowledge no previous study has investigated auditory attention abilities in stroke patients. The present experiment explores the relationship between speech comprehension and three components of attention: alerting, orienting, and executive control, in 25 participants with left hemisphere stroke (all right-handed pre-stroke, native speakers of American English, and at least six-months post-stroke); participants’ aphasia diagnoses were heterogeneous and ranged from mild to severe. Participants completed visual and auditory versions of the Attention Network Test to provide measures of alerting, orienting and executive control in each modality. Speech comprehension was measured using a sentence-picture matching task. Regressions indicate attention modality affects the relationship between the three types of attention and sentence comprehension. Orienting in the auditory modality predicted sentence comprehension; participants with faster reaction times due to orienting cues demonstrated higher sentence comprehension accuracy. Neither auditory alerting nor executive control predicted sentence comprehension. Alternatively, alerting and executive control in the visual modality predicted sentence comprehension; participants who benefited less from alerting cues and exhibited slower executive control abilities also had slower sentence comprehension reaction times. Orienting in the visual modality did not significantly predict sentence comprehension. These preliminary findings suggest that alerting, orienting, and executive control support sentence comprehension in dissociable ways, and that measures of auditory attention provide additional information beyond typical visual tests of attention regarding the cognitive resources available to support speech comprehension after stroke.
Talk 11: The time-course of component processes of selective attention
Tanya Wen1,2, John Duncan1,2, Daniel Mitchell1,2; 1MRC Cognition and Brain Sciences Unit, 2University of Cambridge
Attentional selection shapes human perception, enhancing relevant information, according to behavioral goals. Here, we used multivariate decoding of electrophysiological brain responses (MEG/EEG) to examine real-time representation of the component processes of selective attention. Auditory cues instructed participants to respond to a particular visual target, embedded within a stream of single- and multi-item displays. Although the task logically required items to be compared to an attentional “template”, signals consistent with such a template were relatively weak and appeared to transition through an active sensory format before becoming quiescent. Subsequent stimulus processing evoked strong neural representation of multiple target features, evolving over different timescales. Combining single and multi-item displays with different types of distractors, allowed quantification of various components of attention. Following a visual choice display, we observed five distinguishable processing operations with different time-courses. First, visual properties of the stimulus were strongly represented. Second, the location of the candidate target was rapidly represented in multi-item displays, providing the earliest evidence of modulation by behavioral relevance. Third, biased competition continued to enhance the representation of the candidate target, including its identity, relative to distractors. Fourth, only later was the behavioral significance of the target explicitly represented in single-item displays. Finally, if the target was not identified and search was to be resumed, then an attentional template was weakly reactivated. The observation that an item’s behavioral relevance directs attention in multi-item displays prior to explicit representation of target/non-target status in single-item displays is consistent with two-stage models of attention.
Talk 12: Attentional state dependence of time-resolved inter-network anticorrelated brain activity
Aaron Kucyi1, Josef Parvizi1; 1Stanford University
Functional neuroimaging evidence suggests that the brain’s default mode network (DMN) exhibits antagonistic activity with dorsal attention (DAN) and salience (SN) networks across various behavioral states. We aimed to resolve the temporal dynamics and potential state-dependence of this activity using human intracranial electroencephalography (iEEG) with simultaneous recordings within core nodes of the DMN (posteromedial cortex), DAN (dorsal posterior parietal cortex), and SN (dorsal anterior insular cortex). Seven neurosurgical patients performed multiple sessions of the gradual-onset continuous performance task (a test of sustained attention), and we recorded spontaneous iEEG activity during wakeful rest and sleep as well as pre-operative resting state fMRI. During attentional task performance, the three networks showed clearly dissociable magnitudes and temporal profiles of iEEG high-frequency broadband (HFB; 70-170 Hz) activity. When accounting for inter-network temporal lags of activity, we found that better behavioral performance across sessions was tightly associated with greater DAN-DMN, but not SN-DMN, anticorrelation in the HFB range. During wakeful rest and sleep states, HFB anticorrelated activity was diminished, but wider network-level organization remained similar to task states and showed iEEG-fMRI correspondence within individuals. These findings have important implications for interpreting antagonistic network relationships found with functional neuroimaging and confirm the behavioral importance of time-lagged inter-network interactions.
Talk 13: The involuntary capture of visual attention by task-irrelevant ugly-beauty artificial faces: An ERP study
Eriko Matsumoto1, Tomoya Kawashima1, Tomoyuki Naito2; 1Graduate School of Intercultural Studies, Kobe University, 2Graduate School of Medicine, Osaka University
Visual attention is easily captured by peripheral faces particularly high attractive ones or threat-related ones although they present as completely task-irrelevant distractors. Several neuroimaging studies reported that the beauty and the ugly stimuli activated the emotional brain network or the social reward related areas such as orbitofrontal cortex, premotor cortex or basal ganglia. The function of these networks is thought to be related to attention capture by emotionally valuable faces. However, little is known that the mechanisms for ugly or beauty faces capture of attention involuntarily. The study employed RSVP task and measure event-related potentials (ERPs) to test how the task-irrelevant ugly/beauty faces capture attention. We use artificial faces (beauty and ugly) for distractor stimuli which were made by using classification image paradigm (Naito et al., 2017) to control the level of ugliness and attractiveness (beautifulness). Forty-one participants for behavioral study and twenty participants for ERP study. In each task trial, 20 letters with variable colors were presented rapid serially at the center of the monitor in every 100 ms. Participants were asked to detect target color letter ignoring the peripheral distractor stimulus. The results showed that the rate of correct target detection significantly decreased when ugly face presented as a distractor. N2pc peak amplitude enhanced ugly and beauty distractor condition compare to control condition, also, LPC (late parietal positivity) decreased in ugly condition. These results suggest that emotional valuable face could capture attention before aesthetic perception.
Talk 14: Learned Feature Distributions Predict Visual Search and Working Memory Precision
Phil Witkowski1,2, Joy Geng1,2; 1University of California, Davis, 2Center for Mind and Brain, University of California, Davis
Previous research has established that attention operates by selectively enhancing sensory processing of task-relevant target features held in working memory. Much of this literature uses search displays in which the target is an exact match of cued features. However, real world visual search rarely involves targets that are identical to our memory representations. The ability to deal with cue-to-target variability is a critical, but understudied aspect of visual attention. In these studies, we test the hypothesis that top-down attentional biases are sensitive to the reliability of target feature dimensions over time. In two experiments, subjects completed a visual search task where they saw a target cue composed of a certain motion direction and color, followed by a visual search display with multiple distractors. The target features changed from the cue, with one dimension drawn from a distribution narrowly centered over the cued feature (reliable dimension), while the other was broad (unreliable dimension). The results demonstrate that subjects learned the distributions of cue-to-target variability for the two dimensions and used that information to bias working memory and attentional selection: Reaction times and first saccades were better predicted by the similarity of the consistent feature than the inconsistent feature and the precision of working memory probe responses was greater for the consistent dimension. Moreover, the working memory precision predicted individual variation in search performance. Our results suggest that observers are sensitive to the learned reliability of individual features within a target and use this information adaptively to weight mechanisms of attentional selection.
Talk 15: Functionally specific effects of targeted noninvasive stimulation on hippocampal-cortical network connectivity
Kristen Warren1, Molly Hermiller1, Steven VanHaerents1, Joel Voss1; 1Northwestern University
Episodic memory is supported by the hippocampus and a distributed network of interacting cortical regions. We have shown that stimulation targeting this hippocampal-cortical network (HCN) increases resting-state fMRI correlations among network regions and influences episodic memory. However, connectivity of this network is also altered by memory retrieval. Here, we investigated the functional and regional specificity of stimulation effects on connectivity measured during a specific memory demand, autobiographical retrieval, relative to those observed via resting-state fMRI. Subjects (n=32) underwent resting-state and autobiographical retrieval-state fMRI scans following five days of high-frequency (20 Hz) transcranial magnetic stimulation to either a left lateral parietal cortex location of the HCN (n=16) or a left prefrontal cortex location that is not part of the HCN and not associated with autobiographical retrieval (n=16). We identified state-specific and network-specific effects of stimulation on connectivity. Parietal stimulation had a greater effect on retrieval-state relative to resting-state connectivity primarily in HCN regions, whereas prefrontal stimulation had a greater effect on resting-state relative to retrieval-state. fMRI connectivity changes in the medial temporal lobe due to stimulation predicted corresponding changes in episodic memory performance measured during a separate task, with greater modulation of retrieval-state connectivity relative to rest predicting greater context recollection improvement. These findings indicate that the expression of neuroplasticity generated by HCN-targeted stimulation is most robust when measured during memory processing. Additionally, recollection memory is causally related to memory-specific fMRI connectivity of the MTL. This result supports the utility of noninvasive stimulation for functionally selective modulation of brain networks.
Data Blitz Session 2
Saturday, March 23, 12:00 - 1:30 pm, Ballroom A
Speakers: Melissa Thye, Alina Leminen, Julien Dirani, Garret Kurteff, Bingjiang Lyu, Ben Maassen, Jie Lisa Ji, Suzanne Dikker, Yuan Tao, Sophia Vinci-booher, Erez Freud, Emily Kubicek, Lawrence Appelbaum, Delphine Oudiette, Eti Ben Simon
Talk 1: An Intracranial EEG Study of Taxonomic and Thematic Relations
Melissa Thye1, Jason Geller1, Diana Pizarro1, Jerzy P. Szaflarski1, Daniel Mirman1; 1University of Alabama at Birmingham
The hub-and-spoke model of semantic cognition posits that the anterior temporal lobe (ATL) serves as a critical hub, integrating multimodal information from surrounding spokes to arrive at semantic representations. The dual-hub account suggests that although ATL is important for processing taxonomic relations [defined by shared features (e.g., dog - bear)], an additional hub in the temporo-parietal cortex (TPC) may be specialized for processing thematic relations [based on event-related co-occurrence (e.g., dog - leash)]. Previous studies of semantic cognition have been restricted to neuroimaging methods with limited spatial or temporal resolution. In the current study, we overcome this limitation by recording from 84 electrode channels from 5 participants with refractory epilepsy undergoing intracranial EEGs from an array of multi-contact depth electrodes. Participants completed a semantic relatedness judgement task where critical word pairs varied in semantic type (taxonomic vs. thematic) and relatedness strength (high vs. low). Data were epoched, bandpass filtered to extract the high gamma range (70-110 Hz), transformed to generate the analytic signal, smoothed, and baseline corrected. Peak high gamma power (HGP) was greater for taxonomic compared to thematic relations in ATL. The opposite pattern was observed in TPC with greater peak HGP for thematic relations. Peak HGP was relatively consistent across the regions for the relatedness strength manipulation with greater peak HGP in response to the highly related trials. Our data support a dual-hub account of semantic cognition, with ATL as a critical hub for processing taxonomic relations, and TPC as a secondary hub for processing thematic relations.
Talk 2: Online build-up of neocortical memory traces for spoken words: specific facilitatory effects of novel semantic associations.
Alina Leminen1,2, Eino Partanen1,2, Andreas Højlund Nielsen2, Mikkel Wallentin2, Yury Shtyrov2,3; 1University of Helsinki, Finland, 2Aarhus University, Denmark, 2Saint Petersburg University, Russia
Recent research has shown that the brain is capable of a rapid build-up of novel cortical memory traces for words during mere perceptual exposure to new lexical items. This has been shown as an online increase in electrophysiological response to new word forms even when they have no specific meaning attached and are not attended to or rehearsed by the learners. However, the operation of this fast cortical language-learning mechanism in online acquisition of word meaning has not been thoroughly investigated yet. Here, we presented our participants with novel word forms in a word-learning task taking place during a short magnetoencephalography (MEG) recording session. Novel words were either learned perceptually through auditory exposure only or were assigned a clear semantic reference using a word-picture association task, in which novel words were presented in conjunction with novel objects. Real familiar words were used as controls. Our findings show that, already after approximately five presentations of each stimulus, novel stimuli learnt through semantic association demonstrated stronger activation over the left perisylvian cortices than perceptually acquired word forms that lacked semantic reference. Perceptual items demonstrated a linear learning-related amplitude increase throughout the 10-minute recording session. This result suggests a more efficient process of online novel word memory trace build-up in the presence of semantic reference. Our results confirm rapid formation of memory traces for novel words over a course of a short exposure and suggest facilitatory effects of acquisition of novel semantics on the neocortical memory trace formation.
Talk 3: Lexical Access in Comprehension vs. Production: Spatiotemporal localization of semantic facilitation and interference
Julien Dirani1, Liina Pylkkänen1,2; 1New York University Abu Dhabi, 2New York University
Humans understand words faster when they are preceded by semantically related words. This facilitation is thought to result from spreading activation between words with similar meanings. Interestingly, in language production, semantic relatedness often has the opposite effect: in object naming for example, a related prior word delays the naming time of the current object. This could be due to competition during conceptual search or later interference at the motor preparation stage. However, no study has systematically compared the facilitory and inhibitory effects and thus their neurobiological relationship is unknown. We contrasted maximally parallel production and comprehension tasks during magnetoencephalography and found that in comprehension (specifically word reading), semantic relatedness modulated activity in the left middle superior temporal gyrus at 180-335ms, consistent with prior findings on the spatiotemporal localization of lexical access. In contrast, a semantic interference pattern for the production task (object naming) occurred in a post-lexical time-window at 395-485ms in left posterior insular cortex, consistent with post-lexical motor preparation. Thus, our data show that semantic priming during comprehension and interference during production are not two sides of the same coin but rather they clearly dissociate in space and time, consistent with a lexical account for comprehension and a post-lexical one for production.
Talk 4: Behavioral and Neuroanatomical Characteristics of Stimulation-Induced Speech Arrest
Garret Kurteff1,2, Neal Fox1, Maansi Desai1,2, Alia Shafi1, Edward Chang1; 1University of California, San Francisco, 2University of Texas, Austin
For over a century, neurosurgeons performing awake cortical stimulation mapping have described the phenomenon of Speech Arrest, the temporary discontinuation of speech without simultaneous sensorimotor involvement. Despite its clinical and theoretical importance for the neuroanatomical localization of language function, Speech Arrest remains poorly characterized in the scientific literature. This study provides a comprehensive report of Speech Arrest in 34 patients who participated in clinical mapping tasks (e.g., counting to thirty) while undergoing awake language mapping during left hemisphere neurosurgery. We analyzed 291 speech disruptions using video and audio recordings acquired simultaneously with stimulation mapping, which were behaviorally classified as either instances of Speech Arrest or as Motor Errors. We found that Speech Arrest is characterized by a stimulation-induced delay in the onset of articulation until the termination of stimulation. Once initiated, the quality and duration of pronunciation is unaffected, and patients typically report subjectively feeling “unable to talk.” In contrast, stimulation-induced Motor Errors often begin while stimulation is still ongoing, and the resulting utterances are less intelligible and longer in duration than non-errors. A neuroanatomical dissociation between Speech Arrest and Motor Errors was also found. Most instances of Speech Arrest resulted from stimulation to pars opercularis and ventral/rostral precentral gyrus, while Motor Errors were the dominant error-type in dorsal/caudal quadrants of precentral gyrus. Interestingly, stimulation to pars triangularis (a region often considered part of anatomical Broca’s area) rarely elicited speech errors of either type. This study represents the first comprehensive quantitative characterization of Speech Arrest for neuroscientists and clinicians.
Talk 5: The spatiotemporal dynamics of flexible meaning: Neuromodulation of noun meaning by the preceding verb
Bingjiang Lyu1, Alex Clarke1, Hun Choi1, William Marslen-Wilson1, Lorraine Tyler1; 1Centre for Speech, Language, and the Brain, Department of Psychology, University of Cambridge
As utterances are heard, each word is integrated into an ongoing incremental interpretation, placing immediate constraints on following words. Here we investigate the spatial location and temporal dynamics of the neural mechanisms underpinning these processes, focusing on verb semantic constraints and their effect on the semantic interpretation of its direct object (DO) noun. Participants’ brain activity (EEG/MEG) was recorded while they listened to sentences like “The old man ate the apple”. Using Topic Modelling based on large corpora, we generated models which captured (a) verb semantic constraints (b) DO noun (e.g., ‘apple’) semantics and (c) the effect of verb semantics on the noun (verb-weighted noun semantics) as probability distributions over 200 semantic topics, and tested them against brain activity using Representational Similarity Analysis. Verb semantic constraints showed effects in LpMTG and LSMG from verb onset. Directed connectivity analysis revealed that information flowed continuously from LpMTG to LSMG with a 15 msec delay, whereas information flow from SMG only occurred after verb recognition point. There was no model fit for noun semantics alone, but significant verb-weighted noun semantic effects in LpMTG and LIFG around noun recognition point. While there is rapid and constant feedforward information flow from LpMTG to LIFG from noun onset, feedback from LIFG is relatively slower but long-lasting. These novel results reveal the distinct patterns of dynamic connectivity involved in the broader neural network involved in real-time semantic interpretation, with LpMTG & LSMG involved in semantic access and LpMTG & LIFG in semantic integration.
Talk 6: ‘Print tuning’ as neurophysiological marker of early typical and delayed reading acquisition.
Ben Maassen1,2, Toivo Glatz3; 1University of Groningen, The Netherlands, 2University Medical Center Groningen, The Netherlands, 2Catholic University Leuven, Belgium
The visual N170 is an occipitotemporal electrophysiological ERP-component which provides a measure for print tuning – the cortical specialization to processing of written text. Print tuning emerges in beginning readers during the first two years of reading education and is left-lateralized in advanced readers. In earlier studies with adults and adolescents we found the N170 component to be more pronounced in typical readers as compared to individuals with reading disability. We present two ERP studies which further explore the emergence of print tuning during the early stage of typical and delayed reading acquisition. In study 1 print tuning as measured from 21 typical and poor readers of Dutch in second grade of mainstream primary education, exhibited a robust, enhanced N170 response for words compared to symbols in the left hemisphere, and a positive, almost linear relation between reading fluency and the size of the print tuning effect. In study 2 the effect of a seven weeks computer-based literacy training was evaluated in a cross sectional sample of 36 first graders with behavioral tests and the visual print-tuning N170 paradigm. Conventional analyses of averaged ERPs pointed to an overall presence of left-lateralized print tuning, that was larger at the second than the first assessment. Single trial analyses showed evidence of print tuning in the more proficient readers, but returned null results for the experimental factors session and type of training (math versus literacy). Thus, further studies are needed to understand the precise relation between print tuning and reading development or delay.
Talk 7: Characterizing Individual Variation in Multivariate Connectivity and Behavior Along the Psychosis Spectrum
Jie Lisa Ji1, Joshua Burt1, Katrin Preller1,2, Brendan Adkinson1, Antonija Kolobaric1, Morgan Flynn1, Rick Adams3, Aleksandar Savic1,4, John Murray1, Alan Anticevic1; 1Yale University, 2University of Zurich, 2University College London, 2University of Zagreb
Understanding how neural processes give rise to the behavioral variation observed in patients is a key challenge in the field of neuropsychiatry. Notably, a great deal of heterogeneity in key cognitive domains, including working memory, social and emotional processing, and goal maintenance, exists even within patients with the same diagnosis. Here, we describe a multivariate neurobehavioral framework under which cognitive performance and behavioral variation in psychosis spectrum disorders can be mapped to features of specific neural systems and used to inform the identification of genetic/molecular targets associated with particular cognitive/behavioral deficits. We leverage fMRI-derived neural and behavioral data from 202 healthy controls and 436 patients from the Bipolar-Schizophrenia Network for Intermediate Phenotypes study. We first identify dimensions of symptom and cognitive variation and then demonstrate that variation along these dimensions relates robustly to variation in the global brain connectivity of specific neural systems. Importantly, these behavioral dimensions are not parallel to traditional symptom and cognitive scales from pre-existing clinical instruments; do not reflect conventional diagnostic boundaries; and are highly stable and robust to site and sample effects. We then demonstrate that these neurobehavioral relationships can be readily mapped to neural/cellular properties such as gene expression, thus informing the identification of pharmacological targets aimed at treating specific behavioral phenotypes in the psychosis spectrum at the individual subject level. We propose the Neuro-Behavioral Relationships In Dimensional Geometric Embedding (N-BRIDGE) framework as a key step towards unified mapping between the geometry of behavioral variation and the geometry of neural variation in psychiatry.
Talk 8: mindHIVE: An accessible cognitive neuroscience research platform for students and researchers
Suzanne Dikker1, Henry Valk1, Dano Morrison, Kimberly Burgas, Steven Azeka1, Teon Brooks, Wendy Suzuki1, Ido Davidesco1, David Poeppel1; 1New York University
Recent technological developments have made electroencephalography (EEG) equipment increasingly affordable and accessible. For example, our group has successfully introduced low-cost EEG devices into schools both for research and educational purposes, engaging students in scientific inquiry in a hands-on manner. However, EEG experimentation still requires extensive in-person support from someone with an advanced neuroscience degree and substantial programming skills: Each step—study design, data collection, analysis—is typically handled by different applications, often without a graphical user interface (GUI). To bridge this hardware-software gap, we developed an all-in-one GUI-based platform that guides users through the process of conducting cognitive neuroscience research via (a) intuitive experiment design features (e.g., sliders and ‘drag-and-drop’ components to adapt stimuli, duration, etc.); (b) a data visualizer that supports automatic and visual cleaning; (c) a simple analysis engine that computes Event-Related Potentials (ERPs), frequency spectra, and behavioral results; (d) a workspace that assists in generating research reports. The platform is offered alongside a comprehensive neuroscience curriculum, BrainWaves, currently implemented in 30 underserved schools in New York. We also developed a ready-to-use wireless EEG device (OpenBCI amplifier) that addresses remaining limitations of existing low-cost hardware (data access, accommodation for all hair/head types, electrode locations, fragile components). Initial testing demonstrates its ability to obtain several canonical ERPs and a signal-to-noise ratio comparable to research-grade EEG equipment. In sum, mindHIVE is a comprehensive software/hardware solution that enables students to independently conduct neuroscience/behavioral studies, paving the way for a range of real-world citizen science research and educational applications.
Talk 9: The effects of lesions on the modular organization of the brain: A comparison of simulated and real lesions
Yuan Tao1, Brenda Rapp1; 1Johns Hopkins University
Simulation studies directed at understanding the effects of lesions on functional organization have shown that damage to nodes supporting cross-module or within-module integration have opposite effects on whole-brain modular organization (Honey & Sporns, 2008). However, the consequences of actual lesions have been scarcely studied. In this work we examined the consequences of brain lesions in chronic stroke (n=15) and the simulated impact of these lesions in healthy individuals (n=23). A reference modular structure was computed from controls and, on this basis, global (participation coefficient, or PC) and local (within-module density, or WD) integration coefficients were calculated for each node. For each lesion mask, we computed its PC and WD damage scores by averaging the respective integration coefficients. Simulated lesions were created by applying every lesion mask to each control dataset and modularity (Newman’s Q) was calculated for all datasets. Finally, lesion-mask PC and WD damage scores were correlated with modularity. Consistent with previous studies, simulated lesions with larger PC damage resulted in higher modularity (Pearson r=0.43, p<0.05), while WD damage had the reverse effect (r=0.46, p<0.05). However, for actual lesions, PC damage was negatively correlated with modularity (r=-0.48, p<0.05), and WD damage was uncorrelated with modularity (r=0.07, n.s.). The discrepancy between simulated and real lesions indicates that lesion-driven functional re-organization cannot be explained as a simple subtraction of nodes from the healthy brain. Instead, the findings indicate that lesions lead to functional connectivity changes in which global connectors play a pivotal role.
Talk 10: Sensorimotor contingency leads to developmental changes in the neural mechanisms supporting visual recognition
Sophia Vinci-booher1, Anastasia Nikoulina1, Thomas W. James1, Karin H. James1; 1Indiana University, Bloomington
Letter production leads to increases in functional connectivity among visual and motor brain systems as well as gains in visual letter recognition. We hypothesized that the coupling between hand movements and visual feedback during early learning stages would result in the emergence of visual-motor functional connectivity during perception and, further, that this connectivity would support gains in recognition. Twenty literate adults were trained on four sets of novel symbols over one week. Training conditions were designed to manipulate the contingency between the motor and visual experiences of a letter that occur during letter production: writing with ink, writing without ink, watching a handwritten symbol unfold, and watching a static handwritten symbol. Participants were presented with the training symbols during functional magnetic resonance (fMRI) scanning at three time points: one pre-training, one post-training, and one after a week-long no-training delay. Recognition was tested after each training session and after the third scan. We found that the contingency between visual and motor experiences during production changed the pattern of functional connectivity among visual, motor, and auditory brain systems and resulted in recognition gains at post-training. Recognition gains were maintained after the no-training delay, but the functional connections observed immediately after training returned to their pre-training baselines. Our results suggest that behaviors that couple sensory and motor systems result in temporary increases in functional connectivity that contribute to longer-lasting changes in the neural mechanisms supporting recognition.
Talk 11: Preserved shape sensitivity in the dorsal pathway of a visual agnosia patient
Erez Freud1, Marlene Behrmann2; 1York University, Toronto, ON, Canada, 2Carnegie Mellon University, Pittsburgh, PA, USA
Shape processing is a cornerstone for various perceptual behaviours such as object recognition, face perception and orthographic processing. For decades, these perceptual behaviours were considered to be and investigated as the product of one visual pathway- i.e. the ventral pathway. In contrast, the dorsal visual pathway was assumed to support the visuomotor control of objects. Importantly, however, recent research has revealed novel evidence for a contribution of the dorsal pathway to shape processing. An obvious outstanding question, then, is whether dorsal shape processing mechanisms are dissociable from computations carried out by the ventral pathway. To address this issue, we utilized a parametric scrambling manipulation that has been used successfully in healthy individuals (Freud, Culham, Plaut & Behrmann, 2017; eLife) to map the large-scale organization of shape processing mechanisms in a patient with object agnosia. As expected, neural indices of shape sensitivity along the ventral pathway were remarkably altered in the patient, particularly in regions proximal and anterior to the location of the injury. In contrast, shape sensitivity along the dorsal pathway was preserved and followed a similar topographical organization to that observed in controls. These findings were corroborated when the experiment was repeated two-years after the initial scans were acquired. Together, these findings challenge the binary segregation between the two visual pathways and suggest that the dorsal pathway derives shape representations, that might contribute to perception, independently of the ventral pathway.
Talk 12: Neural correlates of biological motion perception in sign language users
Emily Kubicek1, Lorna C. Quandt1; 1Gallaudet University
Although widely studied in typically developing populations, the neural basis of biological motion perception has not yet been studied amongst a group that uses action as their primary mode of communication: sign language users. We hypothesized that the continuous perception of biological motions used in sign language may mean that native signers show an increased ability to extract relevant action information. With this EEG study we test whether Deaf signers’ (N = 19) sensorimotor systems are differentially sensitive to biological motion presented in two conditions (scrambled vs. unscrambled) compared to hearing non-signers. We predicted greater central alpha event-related desynchronization (ERD) for the unscrambled condition, due to greater demands on sensorimotor cortices when understanding movement. Everyday actions (e.g., jumping jacks, jump rope) were presented using point light displays (PLD). Time-frequency activity in the alpha and beta ranges was computed for each condition at frontal electrodes and central sites overlying the sensorimotor cortex. Paired comparisons showed significantly greater ERD at central electrode sites in response to scrambled PLDs as compared to unscrambled PLDs (p<.05, bootstrapped). This finding suggests that deaf signers may recruit sensorimotor systems more strongly in response to unintelligible actions than coherent action, contrary to our prediction. Frontal electrodes showed the same pattern of ERD (p<.05, bootstrapped), suggesting that executive functions are involved in parsing scrambled PLDs. The results from Deaf native signers were statistically compared to the EEG responses of hearing non-signers. This work provides the first investigation of sensorimotor EEG in Deaf signers during PLD observation.
Talk 13: Transcranial Direct Current Stimulation to Enhance Laparoscopic Technical Skill Learning: A Preregistered Randomized Controlled Trial
Lawrence Appelbaum1, Hannah Palmer1, Zhi-De Deng2, Lysianne Beynel1, Amanda Watts1, Jonathan Young1, Sarah Lisanby2, John Migaly1, Morgan Cox1; 1Duke University, 2National Institute of Mental Health
Surgical skills are technically challenging and require extensive deliberate practice to master. Because trainees have limited time to learn the complex visual-motor skills necessary for successful passage of board certification, there is a strong need for technologies that can expedite learning. Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation technique that can modify neuronal excitability, leading to enhanced motor learning, and potentially accelerate surgical skill acquisition. This pre-registered (NCT03083483), double-blinded and controlled study aimed to test whether active tDCS, applied during laparoscopic surgical skill practice, would lead to larger learning gains relative to sham. Sixty subjects were randomized into three training cohorts: active stimulation to bilateral primary motor cortex (bM1), active stimulation to supplementary motor area (SMA), and sham stimulation. Participants performed 6, 20-minute training blocks of the Fundamental of Laparoscopic Surgery Peg Transfer task over 3 days. For each block, electrical current was ramped up to 2mA over the first 30 seconds and remained constant for active tDCS, while it was immediately ramped off for sham tDCS. Learning curves calculated on scores accounting for timing and errors showed significantly greater improvement for the bM1 group (t=2.07, p=0.039), achieving the same skill level in 4 sessions compared to 6 sessions for the sham group. While the SMA group also produced greater learning, these data were highly variable and didn’t differ significantly from sham (t=0.85, p=0.4). This study therefore demonstrates the potential for tDCS to enhance manual surgical training, meriting further investigation to replicate and extend these findings.
Talk 14: REM sleep respiratory behaviours match mental content in narcoleptic lucid dreamers
Delphine Oudiette1,2,3, Pauline Dodet2, Thomas Similowski2,3, Isabelle Arnulf1,2,3; 1Brain and Spine Institute, 2Sorbonne Universités, 2Pitie-Salpetriere Hospital
Breathing is irregular during rapid eye-movement (REM) sleep, whereas it is stable during non-REM sleep. Why this is so remains a mystery. We propose that irregular breathing has a cortical origin and reflects the mental content of dreams, which often accompany REM sleep. We tested 21 patients with narcolepsy who had the exceptional ability to lucid dream in REM sleep, a condition in which one is conscious of dreaming during the dream and can signal lucidity with an ocular code. Sleep and respiration were monitored during multiple naps. Participants were instructed to modify their dream scenario so that it involved vocalizations or an apnoea, -two behaviours that require a cortical control of ventilation when executed during wakefulness. Most participants (86%) were able to signal lucidity in at least one nap. In 50% of the lucid naps, we found a clear congruence between the dream report (e.g., diving under water) and the observed respiratory behaviour (e.g., central apnoea) and, in several cases, a preparatory breath before the respiratory behaviour. This suggests that the cortico-subcortical networks involved in voluntary respiratory movements are preserved during REM sleep and that breathing irregularities during this stage have a cortical/subcortical origin that reflects dream content.
Talk 15: Under slept and Overanxious: The neural correlates of sleep-loss induced anxiety in the human brain
Eti Ben Simon1, Matthew Walker1,2; 1Department of Psychology, University of California, Berkeley, USA., 2Helen Wills Neuroscience Institute, Berkeley, University of California, USA.
Losing sleep causally triggers anxiety. Disturbed sleep, including Non-REM slow-wave activity (SWA), is comorbid with all anxiety disorders, while sleep-deprivation increases anxiety in healthy individuals. Still, the underlying brain mechanisms of this effect remain unknown. Here, we test the hypothesis that sleep-loss induced anxiety is triggered by a neural profile observed in anxiety disorders: amplified activity within the limbic network due to impaired top-down regulation by medial-PFC. We further examined the palliative role of Non-REM SWA in restoring this network and thus reducing anxiety. 18 healthy adults (20.2±1.5y, 9F) participated in two counterbalanced experimental sessions: (1) after a rested night of polysomnography-recorded sleep, and (2) after 24-hours of sleep-deprivation. Anxiety states were measured in each session prior to a functional-MRI scan assessing affective brain reactivity. Findings reveal that sleep-deprivation triggered a 30% increase in anxiety (p < 0.01) and led to amplified reactivity within the amygdala and dorsal-anterior cingulate, yet marked hypoactivity in medial-PFC (p<0.05). Critically, the degree of medial-PFC disengagement predicted 1) the magnitude of sleep deprivation-induced anxiety across individuals and 2) the loss of top-down mPFC-amygdala connectivity following sleep-loss (both p<0.05). Finally, greater amounts of Non-REM SWA in the sleep rested night predicted the palliative overnight reduction in anxiety, further associated with greater re-engagement of medial-PFC activity the next day (p<0.05). Together, these data establish a neuropathological model explaining the anxiogenic impact of sleep loss and emphasize sleep intervention, specifically deep Non-REM sleep, as a novel therapeutic target for the amelioration of anxiety in non-clinical and clinical populations.
Data Blitz Session 3
Saturday, March 23, 12:00 - 1:30 pm, Ballroom B/C
Speakers: Erhan Genc, Celia Lacaux, Andrew Gordon, Muireann Irish, Rachel G. Pizzie, James H. Kryklywy, Joanna E. Witkin, Debbie Yee, Adam Krause, Milou Sep, Myrthe G. Rijpma, Anne S. Berry, Chris Martin, Jessica A. Collins, David Clewett
Talk 1: Diffusion markers of dendritic density and arborization in gray matter predict differences in intelligence
Erhan Genc1, Christoph Fraenz1, Onur Güntürkün1, Rex Jung2; 1Biopsychology, Department of Psychology, Ruhr University Bochum, Germany, 2Department of Psychology, University of New Mexico, Albuquerque, New Mexico, USA
Previous research has demonstrated that individuals with higher intelligence are more likely to have larger gray matter volume in brain areas predominantly located in parieto-frontal regions. These findings were usually interpreted to mean that individuals with more cortical brain volume possess more neurons and thus exhibit more computational capacity during reasoning. In addition, neuroimaging studies have shown that intelligent individuals, despite their larger brains, tend to exhibit lower rates of brain activity during reasoning. However, the microstructural architecture underlying both observations remains unclear. In this study we assessed microstructural brain anatomy within a large sample of 259 healthy individuals using advanced multi-shell diffusion tensor imaging also known as neurite orientation dispersion and density imaging (NODDI). Further, we conducted a culture-fair matrix-reasoning test in order to measure fluid intelligence. We found that higher intelligence is related to lower values of dendritic density and arborization. We were able to cross-validate our results with data from 506 individuals provided by the Human Connectome Project. Most likely, these findings demonstrate that the neuronal circuitry associated with higher intelligence is organized in a sparse and efficient manner, fostering more directed information processing and less cortical activity during reasoning. In conclusion, this study substantially extends our knowledge about the biological basis of human intelligence differences, by providing insights to efficient information processing during reasoning at the level of axons or dendrites.
Talk 2: Increased creative thinking in narcolepsy
Celia Lacaux1,2, Giuseppe Plazzi3, Isabelle Arnulf1,2, Delphine Oudiette1,2; 1Sorbonne University, IHU@ICM, INSERM, CNRS UMR7225, F-75013 Paris, France, 2AP-HP, Hôpital Pitié-Salpêtrière, Service des Pathologies du Sommeil, F-75013 Paris, France, 2University of Bologna, Bologna, Italy
Rapid eye movement (REM) sleep and dreams have been linked to creativity. However, most studies relied on simple associative tasks and did not assess creativity per se, a process that is unlikely to be achieved in a single nap. To tackle this methodological hurdle, we called upon experts of REM sleep and dreams: subjects with narcolepsy, who present excessive daytime sleepiness, symptoms of dissociated wakefulness and REM sleep (e.g., lucid dreams), and a high dream recall frequency. Given their life-long privileged access to REM sleep and dreams, we hypothesised that subjects with narcolepsy have developed high creative abilities. To test this assumption, we evaluated, with two questionnaires, the creative achievements and the creative profiles of 185 subjects with narcolepsy and of 126 healthy matched-controls. Then, we objectively tested the creative performance of 30 subjects with narcolepsy and of 30 matched-controls. This test assesses the two key modes of creative thinking, namely divergent-exploratory thinking (i.e., finding the greatest number of solutions based on a given stimulus) and convergent-integrative thinking (i.e., integrating several elements into a coherent and original synthesis) on two different domains of expression (graphic and verbal). Subjects with narcolepsy obtained higher scores than controls on the two questionnaires of creativity and on the objective test of creative performance in all the dimensions tested. Additionally, most symptoms of narcolepsy (but not treatments) were associated with higher scores of creativity. These results highlight a higher creative potential in subjects with narcolepsy and further support a role of REM sleep in creativity.
Talk 3: Keeping Track of ‘Alternative Facts’: The Neural Correlates of Processing Misinformation Corrections
Andrew Gordon1,2, Susanne Quadflieg2, Jonathan Brooks2,3, Ullrich Ecker4, Stephan Lewandowsky2,4; 1University of California, Davis, MIND Institute, 2University of Bristol, 2Clinical Research and Imaging Centre, University of Bristol, 2University of Western Australia
Upon receiving a correction, initially presented misinformation often continues to influence people’s judgement and reasoning. Whereas some researchers believe that this so-called continued-influence effect of misinformation (CIEM) simply arises from the insufficient encoding and integration of corrective claims (mental-models account), others assume that it arises from a competition between the correct information and the initial misinformation in memory during retrieval (concurrent-storage account). To examine these possibilities, we conducted two functional magnetic resonance imaging (fMRI) studies. In each study, participants were asked to (a) read a series of brief news reports that contained confirmations or corrections of prior information and (b) evaluate whether subsequently presented memory probes matched the reports’ correct facts rather than the initial misinformation. Behavioural results from both studies revealed that following correction-containing news reports, participants struggled to refute mismatching memory probes, especially when they referred to initial misinformation (as opposed to mismatching probes with novel information). In contrast to the mental-models account of the CIEM, we found little evidence that the encoding of confirmations and corrections produced systematic neural processing differences indicative of distinct encoding strategies. Instead, in both studies, we discovered that following corrections, participants exhibited increased activity in the angular gyrus and the precuneus in response to mismatching memory probes that contained prior misinformation, compared to novel mismatch probes. These findings favour the notion that people’s susceptibility to the CIEM arises from the concurrent retention of both correct and incorrect information in memory.
Talk 4: Elevation of episodic-based mind-wandering in semantic dementia – evidence for functional reorganisation of the brain’s default network
Muireann Irish1,2, Daniel Roquet1,2, Zoë-Lee Goldberg1, Jessica Andrews-Hanna3, John Hodges2,4; 1The University of Sydney, Brain and Mind Centre and School of Psychology, Sydney, Australia, 2Australian Research Council Centre of Excellence in Cognition and its Disorders, Sydney, Australia, 2Department of Psychology, University of Arizona, Tucson, Arizona, USA, 2The University of Sydney, Central Clinical School, Sydney, Australia
Mind-wandering refers to the quintessentially human capacity to perceptually decouple from the immediate surroundings to consider perspectives distinct from the here and now. Reductions in mind-wandering may constitute an important neurocognitive endophenotype across clinical disorders, reflecting alterations in default and frontoparietal brain networks. Here, we explored mind-wandering in semantic dementia, a neurodegenerative disorder characterised by progressive deterioration of the conceptual knowledge base due to anterior temporal lobe degeneration. Thirteen SD patients were contrasted with 28 healthy older Controls on the Shape Expectations task, a minimally demanding cognitive task which reliably promotes mind-wandering. An episodic-semantic mind-wandering score was derived, representing an individual’s tendency to mind-wander in a predominantly episodic versus semantic style. Despite marked semantic deficits, SD patients displayed significantly elevated episodic-based mind-wandering relative to Controls (p < .001). Voxel-based morphometry analyses demonstrated that the co-occurrence of atrophy in the left inferior temporal gyrus and left putamen correlated with mind-wandering in SD. Resting-state functional connectivity analyses further revealed that decreased functional connectivity between inferior parietal regions of the default mode network and the anterior cingulum correlated with SD mind-wandering performance. This study is the first, to our knowledge, to document a striking facilitation of episodic-based mind-wandering in a population characterised by marked cognitive decline and significant neural insult. This increased proclivity for episodic mind-wandering reflects the disconnection between posterior brain regions of the default mode network and prefrontal regions essential for executive control, which may allow intact episodic representations to be harnessed by the SD patient in an unconstrained manner.
Talk 5: Neural Evidence for Cognitive Reappraisal as a Strategy to Alleviate the Effects of Math Anxiety
Rachel G. Pizzie1,2, Cassidy L. McDermott3,2, Tyler G. Salem2, David J.M. Kraemer2; 1Georgetown University, 2Dartmouth College, 2National Institutes of Health
Investigating the neural underpinnings of math anxiety provides an unique opportunity to study the behavioral and neural interactions between cognition and affect. Math anxiety describes feelings of tension, apprehension, and fear that interfere with math performance. High math anxiety (HMA) is correlated with a number of negative consequences, including lower math grades, less self-confidence in math, and ultimately an avoidance of quantitative classes and careers. Given these adverse consequences, it is essential to explore effective interventions to reduce math anxiety. In the present fMRI study, we investigated the efficacy of cognitive reappraisal as a strategy to alleviate the effects of math anxiety. Cognitive reappraisal, an emotion regulation strategy, has been shown to decrease negative affect and amygdala responsivity to negative emotion-eliciting stimuli. Here, we compared an instructed reappraisal strategy to participants’ natural strategy for solving math problems. We found that individuals with high math anxiety showed an increase in accuracy and a decrease in negative affect on the math trials during the reappraise condition as compared to the control condition. During math reappraise trials, higher activity in a network of regions associated with arithmetic was correlated with improved math performance for HMA individuals. These results demonstrate that increased recruitment of arithmetic regions may underlie the benefits of reappraisal in this context. Overall, cognitive reappraisal is a promising strategy for improving math performance and reducing anxiety in math anxious individuals.
Talk 6: Dissociating the neural representations of tactile and hedonic information.
James H. Kryklywy1, Mana R. Ehlers1, Andre O. Beukers2, Sarah M. Moore1, Rebecca M. Todd1, Adam K. Anderson3; 1University of British Columbia, 2Princeton University, 2Cornell University
With the increasing development of multivariate analyses for neuroimaging data, we are beginning to get an deeper and more intricate understanding of how mental states manifest as specific patterns of neural activation. Recent work has outlined activation patterns associated with hedonic and tactile pain representation, yet the degree to which this is overlapping or independent of positive hedonic value and stimulation remains unknown. In the current study, functional magnetic resonance imaging data was collected from 67 participants from Cornell University, who received tactile stimulation to induce painful or pleasurable experience. Representational similarity analyses conducted on BOLD signal from six regions of interest identified the similarity of neural representation across pressure-pain, appetitive brushing and null trials. The resulting similarity matrices were tested against 9 unique theoretical models of potential hedonic and tactile responding to determine the contribution of each area to phenomenological mental states. Of particular note, by individually titrating pain-pressure to a constant hedonic value across participants, while holding appetitive tactile stimulation consistent (with varied hedonic liking), we are able to disembed tactile and hedonic value with unprecedented accuracy. Results demonstrate unique contributions from primary somatosensory cortex, ventromedial prefrontal cortex and insula to both modeled factors. This work gives unique insight into the integration of sensory and affective experience and suggests that while dissociable brain areas may contribute to each experiential feature, there is also great deal of interaction and codependence within these systems.
Talk 7: Neural and Behavioral Mechanisms Underlying the Relationship between Everyday Pain and Cognitive Performance
Joanna E. Witkin1, Steven R. Anderson1, Taylor Bolt2, Maria M. Llabre1, Elizabeth A. Reynolds Losin1; 1University of Miami, 2Emory University
While the extant literature examining the relationship between pain and cognition has primarily involved patients with chronic pain and healthy subjects undergoing experimental pain inductions, few studies have examined the relationship between everyday pain complaints and cognitive performance. The relationship between everyday pain (measured via self-reported pain intensity and pain interference in daily activities) and cognitive performance (measured via accuracy on a working memory n-back task) was examined utilizing a structural equation modeling approach. Neuroimaging and behavioral outcomes were accessed from the Human Connectome Project in a large sample of participants (N=416). Task-related brain activation in the ventromedial prefrontal cortex (vmPFC, previously associated with the evaluative aspects of pain) and self-reported negative affect were included as mediators to investigate potential mechanisms underlying the relationship between everyday pain and cognitive performance. Based on recent findings by Bolt and colleagues (2018), we sought to disentangle brain region from network-level influences on the relationship between everyday pain and cognition by specifying a separate structural model containing a latent construct for default mode network (DMN) activity. Our results demonstrated greater everyday pain intensity was associated with worse cognitive performance via increases in both DMN and vmPFC activity. In contrast, greater everyday pain interference was associated with worse cognitive performance via increased self-reported negative affect. These findings demonstrate that the negative correlation between pain and cognition extends to everyday pain complaints in healthy populations. Furthermore, our results suggest distinct contributions of ROI and network-level brain activity on the association between everyday pain and cognitive performance.
Talk 8: Neural Mechanisms of Motivational Incentive Integration and Cognitive Control
Debbie Yee1, Todd Braver1; 1Washington University in St. Louis
Humans regularly consider a wide variety of incentives when pursuing behavioral goals. Further, our prior work indicates different motivational incentives combine to influence cognitive task performance (Yee et al, 2016 Frontiers). Yet there is currently a poor understanding of the mechanisms by which motivational integration occurs in the brain. We developed an innovative fMRI task paradigm that quantifies dissociable and integrative effects of liquid motivational valence (appetitive = juice, neutral = tasteless solution, aversive = saltwater) and monetary rewards (low, medium, high) on cognitive control. Healthy young adults (N=51) performed a cued task-switching paradigm to earn varying monetary reward amount (low, medium, high), with oral liquid delivery serving as post-trial performance feedback that signaled successful attainment of monetary reward (for fast/accurate responses). Because the symbolic meaning of the liquid was constant, blocked effects of liquid valence on behavior and brain activity were taken to reflect motivational integration effects. Monetary reward effects were widespread, but most reliable in frontoparietal regions, while liquid valence effects were present primarily in value-sensitive brain regions, such as dorsal striatum, and dorsomedial frontal cortex (ACC / pre-SMA). Moreover, in this latter region, strong evidence was found for motivational integration, which moreover predicted both behavioral performance and self-reported motivation. These results are consistent with theoretical accounts suggesting the ACC encodes the expected value of cognitive control, and moreover point to the utility of studying motivational integration processes. Planned follow-up analyses will utilize multivariate decoding to examine the relationship between value-based and task-coding in frontostriatal and frontoparietal circuits.
Talk 9: The pain of sleep loss: A brain characterization in humans.
Adam Krause1, Aric Prather2, Tor Wager3, Martin Lindquist4, Matthew Walker1; 1University of California, Berkeley, 2University of California, San Francisco, 2University of Colorado, Boulder, 2Johns Hopkins University
Sleep loss increases the experience of pain. However, the brain mechanisms underlying altered pain processing following sleep deprivation are unknown. Moreover, it remains unclear whether ecologically modest night-to-night changes in sleep within an individual confer consequential day-to-day changes in experienced pain. Here, we first demonstrate that acute sleep-deprivation amplifies pain reactivity within human (male and female) primary somatosensory cortex, yet blunts pain-reactivity in higher-order valuation and decision-making regions of the striatum and insula cortex. Consistent with this altered neural signature, we further show that sleep deprivation expands the temperature range for classifying a stimulus as painful, specifically through a lowering of pain thresholds. Moreover, the degree of amplified reactivity within somatosensory cortex following sleep deprivation significantly predicts this expansion of experienced pain across individuals. Finally, outside of the laboratory setting, we similarly show that even modest night-to-night changes in sleep quality (increases and decreases) within an individual determine consequential day-to-day changes in experienced pain (decreases and increases, respectively). These data reveal a central brain mechanism underlying the impact of sleep loss on pain perception, and furthermore, establish that the association between sleep and pain is expressed in a night-to-day, bidirectional relationship within individuals. More generally, such findings highlight sleep as an important therapeutic target for pain management within and outside the clinic, including circumstances where sleep is frequently short, yet pain is abundant (e.g. the hospital setting).
Talk 10: Cognitive functioning in post-traumatic stress disorder: a meta-analysis of evidence from animal models & clinical studies
Milou Sep1,2, Elbert Geuze1,2, Marian Joëls2,3; 1Military Mental Healthcare, Dutch Ministry of Defence, 2University Medical Center Utrecht, the Netherlands, 2University Medical Center Groningen, the Netherlands
After a traumatic experience, some people develop post-traumatic stress disorder (PTSD). To improve the prevention and treatment of PTSD in the future, (fundamental) research efforts need to align with clinical reality so that translational findings can foster clinical progress. In the case of PTSD, this implies attention for research in the cognitive domain (including learning & memory), which plays a crucial role in current clinical diagnosis and treatment. In a meta-analysis, we compared current knowledge on learning, memory and fear conditioning (FC) in PTSD patients to healthy controls. Subsequently, data from animal models of PTSD was compared to patient-data, to investigate how preclinical data relates to clinical data. Data searches were performed in Pubmed and the PRISMA guidelines were followed throughout the project. 184 articles were included in this study (60.9% preclinical; 53.8% FC). PTSD patients show enhanced learning and memory of emotional or fearful information but perform worse than healthy controls in learning and memory of neutral information. Preclinical data showed comparable results, with even stronger associations between PTSD and parameters of learning and memory. FC was predominantly assessed in preclinical studies, whereas clinical studies focussed mainly on learning and memory of emotional or neutral information. These discrepancies could inspire future (pre)clinical studies to adopt a more translational, thereby more valuable, set-up. Overall, the results underline the importance of the learning and memory performance in PTSD and suggest that animal models can be used to model the cognitive domain of PTSD.
Talk 11: Salience-Driven Attention is Pivotal to Understanding Others’ Intentions
Myrthe G. Rijpma1, Suzanne M. Shdo1, Gianina Toller1, Joel H. Kramer1, Bruce L. Miller1, Katherine P. Rankin1; 1Memory and Aging Center, University of California, San Francisco, 675 Nelson Rising Ln, Suite 190
A large body of research has been generated to explain the cognitive and neural processes related to inferring other people’s mental states, often called Theory of Mind (ToM). However, most research does not use ecologically realistic test paradigms that can address how an individual selects socially important stimuli to process, and only a subset has focused on practical ToM-related constructs, such as understanding other people’s intentions. In this study, we investigated the structural neuroanatomic contributions to understanding intentions in complex social situations, using the realistic video vignette-based Awareness of Social Inference Test (TASIT) enriched version (SI-E) in 179 participants, including 102 patients with known neurodegenerative diagnoses and 77 older neurologically healthy controls. Statistical Parametric Mapping (SPM12) was used for voxel-based morphometry and structural regions of interest (ROIs) were derived to correspond with key nodes in three functional intrinsically connected networks (ICNs). In concordance with previous ToM studies, distinct neural correlates for understanding others’ intentions included ICN ROIs known to mediate executive functioning (i.e., frontoparietal network, FPN), and memory manipulation (i.e., default mode network, DMN). However, when regions related to salience-driven attention (i.e., salience network, SN) were included in the models, only the SN ROIs independently predicted the ability to infer intentions (p < 0.001), suggesting the FPN and DMN regions may only play a secondary role. Thus, online attention-attribution and selection of socially important information were found to be vital to correctly infer other people’s intentions from realistically complex stimuli, supporting the essential role of attention for normal social cognition.
Talk 12: The influence of catecholamine function on reward-related memory in aging
Anne S. Berry1, Theresa M. Harrison1, A.J. Whitman1, Kaitlin N. Swinnerton1, Ming Hsu1, William J. Jagust1, Anne Berry; 1UC Berkeley
The hippocampus receives dopamine and norepinephrine inputs that may influence memory encoding. Aging is accompanied by parallel declines in neurochemical systems and in memory function. However, memory for positive information may be relatively preserved, or even enhanced compared to young adults. Defining the mechanisms underlying such positive memory biases in aging is an area of active investigation. Our prior research using the PET tracer 6[18F]fluoro-L-m-tyrosine revealed that dopamine synthesis capacity ([18F]FMT Ki) in the striatum is elevated in older adults, rather than reduced. Here, we examined relationships between hippocampal [18F]FMT Ki, which reflects both dopamine and norepinephrine synthesis capacity, and memory enhancement for reward-related stimuli. We examined subsequent memory for houses presented simultaneously with monetary reward, loss, or neutral feedback in young (n = 21) and cognitively normal older adults (n = 36). Older adults, but not young adults, showed better memory for houses paired with rewards relative to neutral outcomes. PET analyses were performed on a subset of amyloid negative older adults (n = 28) who had undergone both [18F]FMT imaging and [18F]AV1451 imaging, which measures aggregated tau. While previous studies have linked [18F]AV1451 binding with memory, we found no significant relationships for the current incidental encoding task. However, higher hippocampal [18F]FMT Ki was associated with greater reward-related memory enhancement (reward-neutral; adjusted r2 = .21, B=168.71, SE= 76.96, p = .04; controlling for age and sex). These findings provide preliminary evidence that age-related memory enhancement for positively-valenced information is associated with higher catecholamine synthesis in the hippocampus.
Talk 13: Reviewing autobiographical memory cues promotes distinctive neural coding in older adults
Chris Martin1, Rachel Newsome1, Bryan Hong1, Andrew Xia1, Christopher Honey2, Morgan Barense1,3; 1University of Toronto, 2Johns Hopkins University, 2Rotman Research Institute
Digital memory augmentation (DMA) is one promising approach to mitigating age-related memory decline. In DMA, portable devices are used to capture information about everyday episodes, making them available for later review. Here, we developed a novel, smartphone-based DMA application that allowed older adults to create and review rich autobiographical memory cues, which were randomly assigned to one of two conditions: replayed or hidden. Over a two-week period, participants created an average of 67.83 autobiographical cues, approximately half of which were reviewed an average of 8.47 times each. Content in the hidden condition was never replayed. A behavioural cued-recall test and fMRI scanning was completed after a delay of one-week and again after a delay of three-months. We revealed a 30% boost in the number of event-specific autobiographical details retrieved in the replayed as compared to the hidden condition, while holding event significance, event frequency, and memory age constant. This effect persisted after a three-month delay. Replayed memories exhibited more distinct neural representations. A representational similarity analysis focused on the hippocampus revealed that voxel patterns of replayed memories were more dissimilar during retrieval than were voxel patterns of hidden memories. A whole-brain searchlight analysis revealed reliable differences in activity patterns between when retrieving hidden and replayed events within the core episodic memory network: posterior medial regions, angular gyrus, medial temporal, and ventromedial prefrontal cortex. Taken together, these findings indicate that reviewing rich autobiographical memory cues can boost retrieval of event-specific detail and promote orthogonal coding of pertinent information in the hippocampus.
Talk 14: Data-driven analysis of whole-brain connectivity reveals post-encoding network dynamics.
Jessica A. Collins1, Bradford C. Dickerson1, J. Benjamin Hutchinson2; 1Massachusetts General Hospital and Harvard Medical School, 2University of Oregon
Previous work from our own lab has provided evidence that reactivation mediated memory consolidation (reviewed in Tambini et al., 2010) modulates connectivity between brain regions selective for the encoding of faces and scenes (Collins & Dickerson, 2018). In the present study, we followed up on this work using Full Correlation Matrix Analysis (FCMA), a method for performing unbiased multivariate analysis of whole-brain functional connectivity. Participants completed a baseline resting-state scan followed by two encoding tasks in which they responded to images of faces or scenes. Each encoding task was followed by an additional resting state scan. The temporal correlation of BOLD activity for every voxel pair in the brain was calculated for the baseline, post-face encoding, and post-scene encoding resting state scans. These matrices were then submitted to multivoxel pattern analysis to identify regions where whole-brain connectivity discriminated between conditions. The connectivity of clusters in perirhinal and posterior cingulate cortex discriminated post-face encoding from baseline rest with 85% accuracy. The connectivity of clusters in parahippocampal gyrus, angular gyrus, and retrosplenial cortex discriminated post-scene encoding from baseline rest with 85% accuracy. A follow-up analysis using network based statistics identified a domain-general brain network, centered on angular gyrus, where connectivity discriminated both post-face encoding and post-scene encoding resting state scans from baseline rest. Our results support the existence of both category selective and domain general brain network changes following the encoding of visual stimuli. In addition, our results support the utilization of unbiased multivariate analysis for studying state-dependent changes in whole-brain connectivity.
Talk 15: Arousal Modulates the Temporal Structure of Episodic Memory
David Clewett1, Camille Gasser2, Lila Davachi2,3; 1New York University, 2Columbia University, 2Nathan Kline Institute
Although everyday life unfolds continuously, we tend to remember our experiences as being more discrete and episodic. But what creates an ‘episode’ in episodic memory? Research suggests that a stable context supports the ongoing integration of sequential information, compressing experience into memories of tight-knit events. By contrast, when the current context changes, a theoretical ‘event boundary’ disrupts sequential integration, instead separating experiences into more distinct memory representations. Although increasing evidence shows that boundaries influence episodic memory organization, little is known about the brain mechanisms that support these processes. Across three studies (N = 34; N = 35; N = 30), we found that event boundaries expanded estimates of temporal distance between item pairs, impaired order memory for recent item pairs, and enhanced item-source memory binding. Using eye tracking, we also found that boundaries elicited significantly greater pupil dilation, an index of physiological arousal, than other novel item presentations. A principle component analysis revealed distinct temporal profiles of this pupil response that were modulated by boundaries. Moreover, distinct pupil components also predicted different effects of boundaries on temporal memory, with parasympathetic inhibition being related to greater time dilation effects in memory and sympathetic activation being related to greater impairments in temporal order memory. In a fourth experiment, we are combining high-resolution fMRI of the medial temporal lobe/brainstem with eye tracking to see how neural memory representations are modulated by arousal at event boundaries. Taken together, these findings shed new light on how arousal mechanisms modulate memories of time and everyday events.