Schedule of Events | Symposia

Symposium Session 10 - What can(‘t) oscillations tell us about cognition?

Symposium Session 10: Tuesday, April 1, 2025, 1:30 – 3:30 pm EDT, Independence Ballroom

Chair: Agatha Lenartowicz¹; ¹UCLA
Presenters: Agatha Lenartowicz, Jan R. Wessel, Bradley Voytek, Cory Inman

Neural oscillations clearly play an important role in cognition and neural computation. Thought to capture fundamental mechanisms of neural communication, they have featured prominently in models of attention, memory, and long-range communication in the brain. Being measurable non-invasively and across species, they also offer a powerful bridge for cross-species inference in brain research. However, the interpretation of oscillatory effects in relation to cognitive constructs is not without debate, as new data challenge existing interpretations. In this symposium four speakers will present and discuss data that highlight limits and emerging trends in oscillatory research as it relates to cognition. Dr. Agatha Lenartowicz will present EEG data showing that attentional interpretations of alpha range oscillations depend on experimental context, suggesting multiple contributing sources. Dr. Jan Wessel will present new intracranial recordings from the basal ganglia that support the notion of beta-gamma activity as a domain-general signature of inhibitory control, which is intractable in time-domain recordings. Additionally, Dr. Bradley Voytek will present work showing that both oscillatory and aperiodic activity are dynamically modulated during memory and cognitive control tasks. Finally, Dr. Cory Inman will present wireless ambulatory intracranial EEG data to show that complementary oscillatory and aperiodic temporal lobe dynamics depend on changes in environmental context and naturalistic behaviors as participants navigate through real-world environments. The symposium thus highlights emerging trends, including integration of context and multidimensional profiles that include periodic and aperiodic activity, in bridging neural oscillations and cognition.

Presentations

Multiple faces of alpha oscillations in constructs of attention.

Agatha Lenartowicz¹; ¹University of California Los Angeles

Alpha oscillations are a robust neurophysiological phenomenon associated with cortical suppression and synaptic input gating, often interpreted as a mechanism of selective attention. Recently dissociations between alpha oscillations and selective attention processes have emerged that question the specificity of this interpretation. In this talk I will present data from a series of experiments in which EEG data were recorded under several conditions, in the laboratory, during naturalistic activities and concurrently with fMRI, during tasks of working memory and sustained attention in children and adults, including those with ADHD. Across these experiments alpha oscillations were evaluated as a marker of attention and associated attention deficits in ADHD. The results indicate: (i) variability in interpretation of alpha oscillations when manipulated within individual versus measured between individuals, as well as when assessed within laboratory-based versus more naturalistic recording conditions, (ii) structured variability in BOLD-functional-connectivity of alpha oscillations across cognitive contexts, and (iii) functional dissociations between alpha power and non-oscillatory features of EEG. Together these results are consistent with multiple sources of variability in typically measured alpha oscillations (including bottom-up and top-down interactions, internal processes, and regulatory system influences on cortical excitability), and suggest that attention-based interpretations of such oscillations depend on the analytical and experimental context in which they are acquired. Our data echo the need for better understanding of mechanisms across different instantiations of alpha oscillations in cognition.

Subthalamic beta-gamma activity as a circuit motif underlying domain-general inhibitory control.

Jan R. Wessel¹; ¹University of Iowa

Inhibitory control is a key executive function. It is used to suppress outdated or inappropriate information. Over the last two decades, intracranial neurophysiological work in humans has detailed the dynamics of a fronto-basal ganglia neural circuit underlying inhibitory control of movement. This circuit prominently features the subthalamic nucleus (STN) of the basal ganglia and operates predominantly in the beta frequency band. Recently, we have hypothesized that this same circuit could also be involved in the inhibitory control of non-motor activity. This was motivated both by the neuroanatomy of the fronto-basal ganglia circuit, as well as by the empirical observation of beta-band activity during the inhibition of sensory and mnemonic processes. Here, I will present two new studies that lend further support to this theory. In both studies, we recorded STN local field potentials from implanted deep-brain stimulators of Parkinson’s patients, alongside scalp EEG. In the first study, we measured the inhibition of active visual attention after distracting sounds. We found that sound-related beta-gamma activity in the STN mediated the sounds’ suppressive influence on visual attention, measured by the steady-state visual evoked potential. In the second study, we used semantic violations to study the inhibitory control of language. Once again, beta-gamma activity in the STN was observed when preactivated, but subsequently violated semantic representations had to be inhibited. Together, this work points towards a consistent circuit motif signifying the domain-general inhibitory control of active neural processes.

Dynamic cortical and oscillatory activity in visual working memory and cognitive control.

Bradley Voytek¹; ¹University of California San Diego

Biological neural networks translate sensory information into a neural code that is transmitted across a hierarchy of cortical brain regions, is held in memory over long timescales, and is relatively robust against distractions. Here we provide evidence that both neural oscillations and aperiodic activity are functionally dynamic in distinct ways during human cognition. We provide evidence for this from three different studies, one using intracranial electroencephalography (iEEG) while participants perform a visual working memory task and two using noninvasive EEG while participants perform a visual hierarchical cognitive control task. In the iEEG study, we leverage a novel time-resolved parameterization of neural spectral activity and find clear, event-related changes in both oscillations and aperiodic activity during memory encoding, changes that are largely independent from one another. During memory encoding, visual cortical alpha oscillatory power significantly decreases while, simultaneously, visual cortical aperiodic activity “flattens out”. In the cognitive control EEG experiments, we found that frontal aperiodic activity becomes “steeper” at stimulus onset. This coincides with a decrease in posterior, aperiodic-adjusted alpha oscillatory power. These changes occurred across all conditions but were greater with increasing task abstraction. These results suggest that aperiodic activity is a dynamic feature of neuronal computation, capturing task-specific reorganization of functional network dynamics when a greater degree of contextual information must be integrated. We interpret these results within the context of current, though not necessarily distinct, theories regarding the physiology of aperiodic neural activity as reflecting neuronal excitation / inhibition balance and / or neuronal timescales.

Intracranial neural dynamics during human navigation in the wild.

Cory Inman¹; ¹University of Utah

The ultimate goal of cognitive neuroscience is to understand and explain real-world behavior in terms of brain activity, and to use these insights to develop therapeutic approaches for neural disorders. By using wearable sensors synchronized with intracranial EEG recordings in epilepsy patients with a permanently implanted deep brain recording system, we can begin to explore the electrophysiological basis of human activities such as real-world navigation and memory encoding in a way that captures the complexity, scale, and functional characteristics of real-world experiences. We hypothesized we’d observe changes in medial temporal lobe (MTL) activity based on changes in spatial context, task-relevant behavior, and movement speed. In this initial observational study, we asked five participants to learn a 0.75-mile route around campus while MTL electrophysiology was recorded. Subjects walked the route 7-8 times across two days, with the 1st walk guided (encoding) and 6-7 of the walks navigated by the participants themselves (navigation retrieval). Using a linear mixed effects model, temporal lobe theta power (5-8 Hz) during real-world spatial navigation is best explained by changes in environmental context, velocity, eye movements, and observer-defined event boundaries. Specifically, low-frequency broadband power (5-12 Hz) significantly increases when participants navigate outdoors relative to indoors. We also find evidence that temporal lobe aperiodic, theta, and gamma activity changes immediately around spatial and cognitive event boundaries. Taken together, we find evidence that neural oscillations and aperiodic activity in the medial and lateral temporal lobe changes around shifts in real-world contexts and cognitive event boundaries.