Infraslow dynamic patterns in human cortical networks track a spectrum of external to internal attention

Early efforts to understand the human cerebral cortex focused on localization of function, assigning functional roles to specific brain regions. More recent evidence depicts the cortex as a dynamic system, organized into flexible networks with patterns of spatiotemporal activity corresponding to attentional demands. In functional MRI (fMRI), dynamic analysis of such spatiotemporal patterns is highly promising for providing non-invasive biomarkers of neurodegenerative diseases and neural disorders. However, there is no established neurotypical spectrum to interpret the burgeoning literature of dynamic functional connectivity from fMRI across attentional states. In the present study, we apply dynamic analysis of network-scale spatiotemporal patterns in a range of fMRI datasets across numerous tasks including a left-right moving dot task, visual working memory tasks, congruence tasks, multiple resting state datasets, mindfulness meditators, and subjects watching TV. We find that cortical networks show shifts in dynamic functional connectivity across a spectrum that tracks the level of external to internal attention demanded by these tasks. Dynamics of networks often grouped into a single task positive network show divergent responses along this axis of attention, consistent with evidence that definitions of a single task positive network are misleading. Additionally, somatosensory and visual networks exhibit strong phase shifting along this spectrum of attention. Results were robust on a group and individual level, further establishing network dynamics as a potential individual biomarker. To our knowledge, this represents the first study of its kind to generate a spectrum of dynamic network relationships across such an axis of attention.

Time-varying functional connectivity predicts fluctuations in sustained attention in a serial tapping task

The mechanisms for how large-scale brain networks contribute to sustained attention are unknown. Attention fluctuates from moment to moment, and this continuous change is consistent with dynamic changes in functional connectivity between brain networks involved in the internal and external allocation of attention. In this study, we investigated how brain network activity varied across different levels of attentional focus (i.e., "zones"). Participants performed a finger-tapping task, and guided by previous research, in-the-zone performance or state was identified by low reaction time variability and out-of-the-zone as the inverse. In-the-zone sessions tended to occur earlier in the session than out-of-the-zone blocks. This is unsurprising given the way attention fluctuates over time. Employing a novel method of time-varying functional connectivity, called the quasi-periodic pattern analysis (i.e., reliable, network-level low-frequency fluctuations), we found that the activity between the default mode network (DMN) and task positive network (TPN) is significantly more anti-correlated during in-the-zone states versus out-of-the-zone states. Furthermore, it is the frontoparietal control network (FPCN) switch that differentiates the two zone states. Activity in the dorsal attention network (DAN) and DMN were desynchronized across both zone states. During out-of-the-zone periods, FPCN synchronized with DMN, while during in-the-zone periods, FPCN switched to synchronized with DAN. In contrast, the ventral attention network (VAN) synchronized more closely with DMN during in-the-zone periods compared with out-of-the-zone periods. These findings demonstrate that time-varying functional connectivity of low frequency fluctuations across different brain networks varies with fluctuations in sustained attention or other processes that change over time.

The interaction between random and systematic visual stimulation and infraslow quasiperiodic spatiotemporal patterns of whole brain activity

One prominent feature of the infraslow BOLD signal during rest or task is quasi-periodic spatiotemporal pattern (QPP) of signal changes that involves an alternation of activity in key functional networks and propagation of activity across brain areas, and that is known to tie to the infraslow neural activity involved in attention and arousal fluctuations. This ongoing whole-brain pattern of activity might potentially modify the response to incoming stimuli or be modified itself by the induced neural activity. To investigate this, we presented checkerboard sequences flashing at 6Hz to subjects. This is a salient visual stimulus that is known to produce a strong response in visual processing regions. Two different visual stimulation sequences were employed, a systematic stimulation sequence in which the visual stimulus appeared every 20.3 secs and a random stimulation sequence in which the visual stimulus occurred randomly every 14~62.3 secs. Three central observations emerged. First, the two different stimulation conditions affect the QPP waveform in different aspects, i.e., systematic stimulation has greater effects on its phase and random stimulation has greater effects on its magnitude. Second, the QPP was more frequent in the systematic condition with significantly shorter intervals between consecutive QPPs compared to the random condition. Third, the BOLD signal response to the visual stimulus across both conditions was swamped by the QPP at the stimulus onset. These results provide novel insights into the relationship between intrinsic patterns and stimulated brain activity.

The interaction between random and systematic visual stimulation and infraslow quasiperiodic spatiotemporal patterns of whole brain activity

One prominent feature of the infraslow BOLD signal during rest or task is quasi-periodic spatiotemporal pattern (QPP) of signal changes that involves an alternation of activity in key functional networks and propagation of activity across brain areas, and that is known to tie to the infraslow neural activity involved in attention and arousal fluctuations. This ongoing whole-brain pattern of activity might potentially modify the response to incoming stimuli or be modified itself by the induced neural activity. To investigate this, we presented checkerboard sequences flashing at 6Hz to subjects. This is a salient visual stimulus that is known to produce a strong response in visual processing regions. Two different visual stimulation sequences were employed, a systematic stimulation sequence in which the visual stimulus appeared every 20.3 secs and a random stimulation sequence in which the visual stimulus occurred randomly every 14∼62.3 secs. Three central observations emerged. First, the two different stimulation conditions affect the QPP waveform in different aspects, i.e., systematic stimulation has greater effects on its phase and random stimulation has greater effects on its magnitude. Second, the QPP was more frequent in the systematic condition with significantly shorter intervals between consecutive QPPs compared to the random condition. Third, the BOLD signal response to the visual stimulus across both conditions was swamped by the QPP at the stimulus onset. These results provide novel insights into the relationship between intrinsic patterns and stimulated brain activity.

Beyond mind wandering: Performance variability and neural activity during off-task thought and other attention lapses

To study the characteristics of attention lapses, a metronome response task and experience sampling were employed while recording fMRI data. Thought prompts queried several attention states (on-task, task-related interference, off-task, inattention). Off-task thoughts were probed on whether they arose in a spontaneous or constrained (i.e., directed) manner. Increased fMRI activation was observed in the default mode network during off-task thought and in subregions of the anterior cingulate cortex and inferior frontal gyrus during inattention. Activation also increased in the left hippocampus during constrained thoughts. Functional connectivity increased between the left superior temporal sulcus and right temporoparietal junction for constrained compared to spontaneous thoughts. Overall, behavioral results indicated a monotonic increase in performance variability from on-task to inattention. However, subtle but consistent differences were observed between self-reported attention state and performance. Results are discussed from perspectives of mind wandering frameworks, the function of brain networks, and the role of engagement in off-task thought.

Measuring task structure with transitional response times: Task representations are more than task sets

The structure of task representations is widely studied with task-switching procedures in which the experimenter compares performance across predetermined categories of trial transitions (viz., switch costs). This approach has been productive, but relies on experimental assumptions about the relationships among stimulus-response mappings that define a set. Here, we develop a novel method of evaluating structure without relying on such assumptions. Participants responded to centrally presented stimuli and we computed the transitional response times (RTs; changes in RT as a function of specific response sequences) for each response combination. Conventional task-switch analyses revealed costs when the response switched from the left-side to the right or vice versa, but this switch cost was not affected by whether the stimuli belonged to a single category or to two distinct categories. In contrast, the transitional RT analysis provided fine-grained information about relationships among responses and how these relationships were affected by stimulus and response manipulations. Specifically, tasks containing a single stimulus category produced response chains in which neighboring responses had lower transitional RTs, while these chains were broken when two stimulus categories were used. We propose that the transitional RT approach offers a more detailed picture of the underlying task representation that reveals structure not detectable by conventional switch cost measures and does not require a priori assumptions about task organization.

Dissociating the Neural Correlates of Planning and Executing Tasks with Nested Task Sets

Task processing (e.g., the preparation and execution of responses) and task representation (e.g., the activation and maintenance of stimulus-response and context information) are two facets of cognitive control supported by lateral frontal cortex (LFC). However, the mechanistic overlap (or distinction) between these two facets is unknown. We explored this by combining a complex task mapping with a precueing procedure. Participants made match/nonmatch judgments on pairs of stimuli during fMRI recording. Precues on each trial gave variable amounts of information to the participant in anticipation of the stimulus. Our results demonstrated that regions throughout LFC were more active at the stimulus (when responses could be executed) than at the cue (when they could only be prepared), indicating that they supported execution of the task agnostic to the specific task representation. A subset of regions in the left caudal LFC showed increased activity with more cue information at the cue and the reverse at the stimulus, suggesting their involvement in reducing uncertainty within the task representation. These results suggest that one component of task processing is preparing and executing the task according to the relevant representation, confined to left caudal LFC, whereas nonrepresentational functions that occur primarily during execution are supported by different regions throughout the rest of LFC. We further conducted an exploratory investigation of connectivity between the two groups of regions in this study and their potential relationship to the frontoparietal and cingulo-opercular networks. Regions with both patterns of activity appear to be part of the frontoparietal network.

Reduced interference in working memory following mindfulness training is associated with increases in hippocampal volume

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 whether and 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. 79 participants were randomized to a 4-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. No group differences were found in hippocampal volume, yet proactive interference improvements following mindfulness training were significantly associated with volume increases in the left hippocampus. These results provide the first evidence to suggest that (1) mindfulness training can protect against proactive interference, and (2) 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.

Investigating the Intersession Reliability of Dynamic Brain-State Properties

Dynamic functional connectivity metrics have much to offer to the neuroscience of individual differences of cognition. Yet, despite the recent expansion in dynamic connectivity research, limited resources have been devoted to the study of the reliability of these connectivity measures. To address this, resting-state functional magnetic resonance imaging data from 100 Human Connectome Project subjects were compared across 2 scan days. Brain states (i.e., patterns of coactivity across regions) were identified by classifying each time frame using k means clustering. This was done with and without global signal regression (GSR). Multiple gauges of reliability indicated consistency in the brain-state properties across days and GSR attenuated the reliability of the brain states. Changes in the brain-state properties across the course of the scan were investigated as well. The results demonstrate that summary metrics describing the clustering of individual time frames have adequate test/retest reliability, and thus, these patterns of brain activation may hold promise for individual-difference research.

Task structure boundaries affect response preparation

Does cognitive control operate globally (across task sets) or locally (within a task set)? Recently, two of the current co-authors (Hazeltine and Schumacher 2016; Schumacher and Hazeltine 2016) proposed that humans represent tasks as task files: hierarchically structured, compartmentalized subsets of our current goals and motivations, task instructions, and relevant stimuli and responses that are selected during task performance according to associated contextual rules. Here, we hypothesize that these task representations bound the implementation of cognitive control at distinct levels of this hierarchical structure. To investigate how task structure influences the implementation of control processes, we conducted a pair of experiments that utilized a precuing procedure. To manipulate task structure, we gave participants mappings in which two stimulus sets were either mapped so that each set was separated by response hand or both sets were interleaved across hands. In Experiment 1, participants responded to sets of images distinguished by their semantic category; in Experiment 2, they responded to sets based on different perceptual features (viz., location or color). During each experiment, precues could give information about the stimulus category or response hand for the upcoming target. The results indicate that participants with separated mappings represented the task hierarchically, while those with interleaved mappings did not. This pattern was consistent across experiments, despite the differences in the way that each set of stimuli influenced representation of the low-level task features. These findings suggest that task structure can be represented hierarchically, and that this structure supports distinct cognitive control processes at different hierarchical levels.

Quasi-periodic patterns contribute to functional connectivity in the brain

Functional connectivity is widely used to study the coordination of activity between brain regions over time. Functional connectivity in the default mode and task positive networks is particularly important for normal brain function. However, the processes that give rise to functional connectivity in the brain are not fully understood. It has been postulated that low-frequency neural activity plays a key role in establishing the functional architecture of the brain. Quasi-periodic patterns (QPPs) are a reliably observable form of low-frequency neural activity that involve the default mode and task positive networks. Here, QPPs from resting-state and working memory task-performing individuals were acquired. The spatiotemporal pattern, strength, and frequency of the QPPs between the two groups were compared and the contribution of QPPs to functional connectivity in the brain was measured. In task-performing individuals, the spatiotemporal pattern of the QPP changes, particularly in task-relevant regions, and the QPP tends to occur with greater strength and frequency. Differences in the QPPs between the two groups could partially account for the variance in functional connectivity between resting-state and task-performing individuals. The QPPs contribute strongly to connectivity in the default mode and task positive networks and to the strength of anti-correlation seen between the two networks. Many of the connections affected by QPPs are also disrupted during several neurological disorders. These findings contribute to understanding the dynamic neural processes that give rise to functional connectivity in the brain and how they may be disrupted during disease.

Dual-Task Processing With Identical Stimulus and Response Sets: Assessing the Importance of Task Representation in Dual-Task Interference

Limitations in our ability to produce two responses at the same time – that is, dual-task interference – are typically measured by comparing performance when two stimuli are presented and two responses are made in close temporal proximity to when a single stimulus is presented and a single response is made. While straightforward, this approach leaves open multiple possible sources for observed differences. For example, on dual-task trials, it is typically necessary to identify two stimuli nearly simultaneously, whereas on typical single-task trials, only one stimulus is presented at a time. These processes are different from selecting and producing two distinct responses and complicate the interpretation of dual- and single-task performance differences. Ideally, performance when two tasks are executed should be compared to conditions in which only a single task is executed, while holding constant all other stimuli, response, and control processing. We introduce an alternative dual-task procedure designed to approach this ideal. It holds stimulus processing constant while manipulating the number of “tasks.” Participants produced unimanual or bimanual responses to pairs of stimuli. For one set of stimuli (two-task set), the mappings were organized so an image of a face and a building were mapped to particular responses (including no response) on the left or right hands. For the other set of stimuli (one-task set), the stimuli indicated the same set of responses, but there was not a one-to-one mapping between the individual stimuli and responses. Instead, each stimulus pair had to be considered together to determine the appropriate unimanual or bimanual response. While the stimulus pairs were highly similar and the responses identical across the two conditions, performance was strikingly different. For the two-task set condition, bimanual responses were made more slowly than unimanual responses, reflecting typical dual-task interference, whereas for the one-task set, unimanual responses were made more slowly than bimanual. These findings indicate that dual-task costs occur, at least in part, because of the interfering effects of task representation rather than simply the additional stimulus, response, or other processing typically required on dual-task trials.

Infraslow Electroencephalographic and Dynamic Resting State Network Activity

A number of studies have linked the blood oxygenation level dependent (BOLD) signal to electroencephalographic (EEG) signals in traditional frequency bands (δ, θ, α, β, and γ), but the relationship between BOLD and its direct frequency correlates in the infraslow band (<1 Hz) has been little studied. Previously, work in rodents showed that infraslow local field potentials play a role in functional connectivity, particularly in the dynamic organization of large-scale networks. To examine the relationship between infraslow activity and network dynamics in humans, direct current (DC) EEG and resting state magnetic resonance imaging data were acquired simultaneously. The DC EEG signals were correlated with the BOLD signal in patterns that resembled resting state networks. Subsequent dynamic analysis showed that the correlation between DC EEG and the BOLD signal varied substantially over time, even within individual subjects. The variation in DC EEG appears to reflect the time-varying contribution of different resting state networks. Furthermore, some of the patterns of DC EEG and BOLD correlation are consistent with previous work demonstrating quasiperiodic spatiotemporal patterns of large-scale network activity in resting state. These findings demonstrate that infraslow electrical activity is linked to BOLD fluctuations in humans and that it may provide a basis for large-scale organization comparable to that observed in animal studies.

Quasi-periodic patterns of intrinsic brain activity in individuals and their relationship to global signal

Quasiperiodic patterns (QPPs) as reported by Majeed et al., 2011 are prominent features of the brain's intrinsic activity that involve important large-scale networks (default mode, DMN; task positive, TPN) and are likely to be major contributors to widely used measures of functional connectivity. We examined the variability of these patterns in 470 individuals from the Human Connectome Project resting state functional MRI dataset. The QPPs from individuals can be coarsely categorized into two types: one where strong anti-correlation between the DMN and TPN is present, and another where most areas are strongly correlated. QPP type could be predicted by an individual's global signal, with lower global signal corresponding to QPPs with strong anti-correlation. After regression of global signal, all QPPs showed strong anti-correlation between DMN and TPN. QPP occurrence and type was similar between a subgroup of individuals with extremely low motion and the rest of the sample, which shows that motion is not a major contributor to the QPPs. After regression of estimates of slow respiratory and cardiac induced signal fluctuations, more QPPs showed strong anti-correlation between DMN and TPN, an indication that while physiological noise influences the QPP type, it is not the primary source of the QPP itself. QPPs were more similar for the same subjects scanned on different days than for different subjects. These results provide the first assessment of the variability in individual QPPs and their relationship to physiological parameters.

The effect of visual and musical suspense on brain activation and memory during naturalistic viewing

We tested the hypothesis that, during naturalistic viewing, moments of increasing narrative suspense narrow the scope of attentional focus. We also tested how changes in the emotional congruency of the music would affect brain responses to suspense, as well as subsequent memory for narrative events. In our study, participants viewed suspenseful film excerpts while brain activation was measured with functional magnetic resonance imaging. Results indicated that suspense produced a pattern of activation consistent with the attention-narrowing hypothesis. For example, we observed decreased activation in the anterior calcarine sulcus, which processes the visual periphery, and increased activity in nodes of the ventral attention network and decreased activity in nodes of the default mode network. Memory recall was more accurate for high suspense than low suspense moments, but did not differ by soundtrack congruency. These findings provide neural evidence that perceptual, attentional, and memory processes respond to suspense on a moment-by-moment basis.

Hierarchical Task Representation

Human behavior is remarkably complex—even during the performance of relatively simple tasks—yet it is often assumed that learned associations between stimuli and responses provide the representational substrate for action selection. Here, we introduce an alternative framework, called a task file, that includes hierarchical associations between stimulus features, response features, goals, and drives, which may overcome the limitations inherent in the conceptualization of response selection as being based solely on associations between stimuli and responses. We then review evidence from our own experimental research showing that even in the context of performing relatively easy tasks, the stimulus-response-association approach to response selection is inadequate to account for the interactions between discrete responses. Instead, response selection may emerge from competition between linked representations at multiple levels.

Errors on interrupter tasks presented during spatial and verbal working memory performance are linearly linked to large-scale functional network connectivity in high temporal resolution resting state fMRI

The brain is organized into networks composed of spatially separated anatomical regions exhibiting coherent functional activity over time. Two of these networks (the default mode network, DMN, and the task positive network, TPN) have been implicated in the performance of a number of cognitive tasks. To directly examine the stable relationship between network connectivity and behavioral performance, high temporal resolution functional magnetic resonance imaging (fMRI) data were collected during the resting state, and behavioral data were collected from 15 subjects on different days, exploring verbal working memory, spatial working memory, and fluid intelligence. Sustained attention performance was also evaluated in a task interleaved between resting state scans. Functional connectivity within and between the DMN and TPN was related to performance on these tasks. Decreased TPN resting state connectivity was found to significantly correlate with fewer errors on an interrupter task presented during a spatial working memory paradigm and decreased DMN/TPN anti-correlation was significantly correlated with fewer errors on an interrupter task presented during a verbal working memory paradigm. A trend for increased DMN resting state connectivity to correlate to measures of fluid intelligence was also observed. These results provide additional evidence of the relationship between resting state networks and behavioral performance, and show that such results can be observed with high temporal resolution fMRI. Because cognitive scores and functional connectivity were collected on nonconsecutive days, these results highlight the stability of functional connectivity/cognitive performance coupling.

Dissociation of Storage and Rehearsal in Verbal Working Memory: Evidence From Positron Emission Tomography

Current cognitive models of verbal working memory include two components a phonological store and a rehearsal mechanism that refreshes the contents of this store We present research using positron emission tomography (PET) to provide further evidence for this functional division In Experiment 1, subjects performed a variant of Sternberg's (1966) item recognition task Experiment 2 used a continuous memory task with control conditions designed to separate the brain regions underlying storage and rehearsal The results show that independent brain regions mediate storage and rehearsal In Experiment 3, a dual-task procedure supported the assumption that these memory tasks elicited a rehearsal strategy

Neural evidence that suspense narrows attentional focus

The scope of visual attention changes dynamically over time. Although previous research has reported conditions that suppress peripheral visual processing, no prior work has investigated how attention changes in response to the variable emotional content of audiovisual narratives. We used fMRI to test for the suppression of spatially peripheral stimuli and enhancement of narrative-relevant central stimuli at moments when suspense increased in narrative film excerpts. Participants viewed films presented at fixation, while flashing checkerboards appeared in the periphery. Analyses revealed that increasing narrative suspense caused reduced activity in peripheral visual processing regions in the anterior calcarine sulcus and in default mode network nodes. Concurrently, activity increased in central visual processing regions and in frontal and parietal regions recruited for attention and dynamic visual processing. These results provide evidence, using naturalistic stimuli, of dynamic spatial tuning of attention in early visual processing areas due to narrative context.

Audience preferences are predicted by temporal reliability of neural processing

Naturalistic stimuli evoke highly reliable brain activity across viewers. Here we record neural activity from a group of naive individuals while viewing popular, previously-broadcast television content for which the broad audience response is characterized by social media activity and audience ratings. We find that the level of inter-subject correlation in the evoked encephalographic responses predicts the expressions of interest and preference among thousands. Surprisingly, ratings of the larger audience are predicted with greater accuracy than those of the individuals from whom the neural data is obtained. An additional functional magnetic resonance imaging study employing a separate sample of subjects shows that the level of neural reliability evoked by these stimuli covaries with the amount of blood-oxygenation-level-dependent (BOLD) activation in higher-order visual and auditory regions. Our findings suggest that stimuli which we judge favourably may be those to which our brains respond in a stereotypical manner shared by our peers.

Encephalographic brain recordings are often used to characterize neuronal dynamics at the network level in relation to specific behaviours. Here, Dmochowski et al. show that neural activity from a few individuals viewing popular media can predict population-level neural activity in thousands of individuals.