Widge, A.S., Boggess, M., Rockhill, A.P., Mullen, A., Sheopory, S., Loonis, R. Freeman, D.K., and Miller, E.K.  (2018)  Altering alpha-frequency brain oscillations with rapid analog feedback-driven neurostimulation. PLOS ONE

Abstract
Oscillations of the brain’s local field potential (LFP) may coordinate neural ensembles and brain networks. It has been difficult to causally test this model or to translate its implications into treatments because there are few reliable ways to alter LFP oscillations. We developed a closed-loop analog circuit to enhance brain oscillations by feeding them back into cortex through phase-locked transcranial electrical stimulation. We tested the system in a rhesus macaque with chronically implanted electrode arrays, targeting 8–15 Hz (alpha) oscillations. Ten seconds of stimulation increased alpha oscillatory power for up to 1 second after stimulation offset. In contrast, open-loop stimulation decreased alpha power. There was no effect in the neighboring 15–30 Hz (beta) LFP rhythm or on a neighboring array that did not participate in closed-loop feedback. Analog closed-loop neurostimulation might thus be a useful strategy for altering brain oscillations, both for basic research and the treatment of neuro-psychiatric disease

An article in Scientific American discusses the debate about whether brain waves have a function or whether they are an epiphenomenon:
Do Brain Waves Conduct Neural Activity Like a Symphony?    There are so many things wrong with this debate.

We know so little about how the brain functions. To dismiss observable signals as an epiphenomenon assumes a level of knowledge that no one has.  It is merely a defense of the status quo paradigm, straight out of Thomas Kuhn.  The attitude boils down to “they don’t play a role because we kind of already know how the brain works”.  No, you don’t.  No one does.

What do neurons do?: They spike when their membrane potential reaches the spiking threshold.  What do oscillations do?  They move the membrane potential towards and away from the spiking threshold.  You start with the assumption that oscillations matter, not that they don’t.  The latter is just a defense of what you think you already know.  That attitude holds back progress.

The article says that “critics point out that oscillations arise everywhere one looks in nature” so therefore brain oscillations are not functional.  To my mind, that is an argument *for* a functional role.  Evolution builds on what is already available.

Critics also say that there is not a lot of evidence for a link between oscillations and mental states. Yes, there is. There are more and more papers published each week.  There might be more evidence for spiking but that is just because spiking has been studied longer.  The same type of correlational evidence is there for both spikes and oscillations.  As for the lack of a causal role between oscillations and brain function, the evidence for a causal role between spiking and function is equally flimsy.   Spiking is not the gold standard just because it is the first and easiest thing we could measure when we only had single-channel amplifiers and slow or no computers.

Skepticism is a good thing but where is the skepticism about the notion that spikes do it all?  For example, consider the following quote from the article: “evidence amassed so far is not based on rigorous tests looking for a cause-and-effect relationship between gamma waves and specific neural processes.”  The same statement is equally true for cause-and-effect between spikes and function.  Somehow that gets a free pass?  The oscillation naysayers accept their spikes-only model without question but set a high bar for others.  I refer you back to Thomas Kuhn.

Spikes vs oscillations is not an either/or thing.  They both work together.  Indeed, it is hard to imagine how one would decouple them.  The bottom line is that both spikes and oscillations are both signals.  No one knows enough about how the brain works to dismiss the measuring of a signal. Since when is more data a bad thing?

Our new paper describing a new model of working memory.  Actually, not so much a new model as an update to the classic model. The classic model posited that we hold thoughts “in mind” (i.e., in working memory) via the persistent spiking of neurons.  That is not wrong.  It is right to a certain level of approximation. There is little doubt that spikes help maintain working memories. However, a closer examination revealed that there is much more going on than persistent spiking.

It is important to keep in mind (pun intended) that virtually all evidence for persistent spiking comes from experiments that averaged neural activity across trials.  The assumption was that averaging boosts signal and decreases noise (“noise” meaning changes in activity from trial to trial). But what if that noise was not noise but real neural dynamics?  We don’t always think about the same thing in the same way.  Averaging assumes we do.

With this in mind, we and others have been leveraging multiple-electrode recording to examine neural activity in “real time” (on individual trials).  This has revealed that working memory-related spiking occurs in sparse, coordinated bursts of activity.  It also revealed oscillatory dynamics between brain waves in two frequency bands, beta and gamma.  Gamma seems to act as a carrier wave that maintains the contents of working memory. Beta seems carry the top-down control signals that allow us to exert volitional control over working memory.

Miller, E.K., Lundqvist, M., and Bastos, A.M. Working Memory 2.0 Neuron  DOI:https://doi.org/10.1016/j.neuron.2018.09.023  (Download a *free* copy for the next 50 days)

Here’s the abstract:
Working memory is the fundamental function by which we break free from reflexive input-output reactions to gain control over our own thoughts. It has two types of mechanisms: online maintenance of information and its volitional or executive control. Classic models proposed persistent spiking for maintenance but have not explicitly addressed executive control. We review recent theoretical and empirical studies that suggest updates and additions to the classic model. Synaptic weight changes between sparse bursts of spiking strengthen working memory maintenance. Executive control acts via interplay between network oscillations in gamma (30–100 Hz) in superficial cortical layers (layers 2 and 3) and alpha and beta (10–30 Hz) in deep cortical layers (layers 5 and 6). Deep-layer alpha and beta are associated with top-down information and inhibition. It regulates the flow of bottom-up sensory information associated with superficial layer gamma. We propose that interactions between different rhythms in distinct cortical layers underlie working memory maintenance and its volitional control.

V1-V4 gamma coherence before stimulus change predicts reaction time to detect the change while deviations from the phase relation increases reaction times.  Nice.

Gamma Synchronization between V1 and V4 Improves Behavioral Performance
Gustavo Rohenkohl, Conrado Arturo Bosman, Pascal Fries

A trial by trial analysis showed that beta bursts, as opposed to power averaged across trials, is a good predictor of variations in motor behavior.

Torrecillos, F., Tinkhauser, G., Fischer, P., Green, A. L., Aziz, T. Z., Foltynie, T., … & Tan, H. (2018). Modulation of beta bursts in the subthalamic nucleus predicts motor performance. Journal of Neuroscience, 38(41), 8905-8917.

Nice review from Miler Lab alumnus Joni Wallis arguing for the importance of single-trial analyses.  Variability across trials may not be noise, it may be cognition.   Joni argues that ensembles, not single neurons, are the fundamental unit in the brain.  One needs to record from many neurons simultaneously to understand cognitive processes.

Wallis, J. D. (2018). Decoding Cognitive Processes from Neural EnsemblesTrends in Cognitive Sciences.

 

This review highlights work showing that spectrally distributed oscillations and their coupling have functional relevance for sensorimotor processing.

Palva, S., & Palva, J. M. (2018). Roles of brain criticality and multiscale oscillations in temporal predictions for sensorimotor processing. Trends in Neurosciences, 41(10), 729-743.

Nice FMRI study showing that working memory delay activity is primarily in the superficial, feedforward, cortical layers while behavioral response-related activity is primarily in deep, feedback layers.

Layer-dependent activity in human prefrontal cortex during working memory
Emily S. Finn, Laurentius Huber, David C. Jangraw, Peter A. Bandettini
doi: https://doi.org/10.1101/425249

This is very consistent with our recent work:
Bastos, A.M., Loonis, R., Kornblith, S., Lundqvist, M., and Miller, E.K. (2018)  Laminar recordings in frontal cortex suggest distinct layers for maintenance and control of working memory.  Proceedings of the National Academy of Sciences.  View PDF

 

Holmes, C.D., Papadimitriou, C.,  Snyder, L.H.(2018)  Dissociation of LFP Power and Tuning in the Frontal Cortex during Memory  Journal of Neuroscience

Nice paper. Well done.  But with a caveat. The authors show that absolute power is dissociated from neural tuning in spiking activity.  From this, they conclude that “oscillatory activity by itself is likely not a substrate of memory” and “may be an epiphenomenon of a rate code in the circuit, rather than a direct substrate”.

Not quite.  No one is claiming that absolute power alone carries specific information. Rather, it is *patterns of coherence* that carry information (e.g., Buschman et al., 2012; Salazar et al 2012; Antzoulatos and Miller, 2014).  If so, there is no reason to think that information would be carried by absolute power.  For example, two different patterns of coherence for two different items could have equal global power because it is the pattern, not the global power, that matters.  In fact, we and others have shown that coherence and power can be dissociated (Buschman et al., 2012).  Using absolute power as a proxy to argue against a functional role for oscillations is a “straw man” argument. It tests a hypothesis that does not reflect the state-of-the-art of thinking on this matter.

Further reading:
Antzoulatos, E.G. and Miller, E.K. (2014) Increases in functional connectivity between the prefrontal cortex and striatum during category learning.  Neuron, 83:216-225. View PDF »

Buschman, T.J., Denovellis, E.L., Diogo, C., Bullock, D. and Miller, E.K. (2012) Synchronous oscillatory neural ensembles for rules in the prefrontal cortex.  Neuron. 76: 838-846. View PDF »

Salazar, R.F., Dotson, N.M., Bressler, S.L., and Gray, C.M. (2012). Content-Specific Fronto-Parietal Synchronization During Visual Working Memory. Science 1224000

Another point:  The reason they see “tuning” for contra vs ipsilateral targets in power is not because of stimulus tuning per se, it is because the right vs left visual hemifields are somewhat independent.  See:
Buschman,T.J., Siegel, M., Roy, J.E. and Miller, E.K. (2011) Neural substrates of cognitive capacity limitations. Proceedings of the National Academy of Sciences. 108(27):11252-5. View PDF »

Kornblith, S., Buschman, T.J., and Miller, E.K. (2015)  Stimulus load and oscillatory activity in higher cortex. Cerebral Cortex. Published online August 18, 2015  doi: 10.1093/cercor/bhv182. View PDF »

Nice study showing that anterior parts of the prefrontal cortex and more plastic than posterior parts.

Anterior-posterior gradient of plasticity in primate prefrontal cortex
Mitchell R. Riley, Xue-Lian Qi, Xin Zhou & Christos Constantinidis
Nature Communications volume 9, Article number: 3790 (2018)

Zanos et al show that beta oscillations play a role in short-term synaptic plasticity in primate neocortex that may explain the role of oscillations in attention, learning, and cortical reorganization.

Zanos, S., Rembado, I., Chen, D., & Fetz, E. E. (2018). Phase-locked stimulation during cortical beta oscillations produces bidirectional synaptic plasticity in awake monkeys. Current Biology.

See discussion of this paper by Womelsdorf and Hoffman:
Latent Connectivity: Neuronal Oscillations Can Be Leveraged for Transient Plasticity

Bouchacourt and Buschman describe a two-layer model of working memory. A sensory layer feeds into an unstructured layer of neurons with random connections (i.e., “mixed-selectivity” type neurons).  It is flexible but interference between representations results in a capacity limit.  Sounds like working memory to me.

Bouchacourt, F., & Buschman, T. J. (2018). A Flexible Model of Working Memory. bioRxiv, 407700.

More about mixed-selectivity:
Fusi, S., Miller, E.K., and Rigotti, M. (2016) Why neurons mix: High dimensionality for higher cognition.  Current Opinion in Neurobiology. 37:66-74  doi:10.1016/j.conb.2016.01.010. View PDF »

Rigotti, M., Barak, O., Warden, M.R., Wang, X., Daw, N.D., Miller, E.K., & Fusi, S. (2013) The importance of mixed selectivity in complex cognitive tasks. Nature, 497, 585-590, doi:10.1038/nature12160. View PDF »

04 Sep 2018
September 4, 2018

Phase-coding memories in mind

Neuroscience

Nice summary of phase coding models of working memory by Hakim and Vogel, including a recent paper by Bahramisharif et al.

Hakim, N., & Vogel, E. K. (2018). Phase-coding memories in mindPLoS biology16(8), e3000012.

Bahramisharif, A., Jensen, O., Jacobs, J., & Lisman, J. (2018). Serial representation of items during working memory maintenance at letter-selective cortical sitesPLoS biology16(8), e2003805.

28 Aug 2018
August 28, 2018

Attention is rhythmic!

Neuroscience

Two new, exciting papers in Neuron that “put the last nail(s) in the coffin of sustained attention.”  They present compelling evidence that sustained attention is not sustained at all but fluctuates with theta rhythms and alpha/beta rhythms. This provides yet more evidence that the brain works by rhythmic switching between representations.

Ian C. Fiebelkorn, Mark A. Pinsk, Sabine Kastner

A Dynamic Interplay within the Frontoparietal Network Underlies Rhythmic Spatial Attention
Neuron, Volume 99, Issue 4, 22 August 2018, Pages 842-853.e8

Randolph F. Helfrich, Ian C. Fiebelkorn, Sara M. Szczepanski, Jack J. Lin, Josef Parvizi, Robert T. Knight, Sabine Kastner

Neural Mechanisms of Sustained Attention Are Rhythmic
Neuron, Volume 99, Issue 4, 22 August 2018, Pages 854-865.e5

An excellent Preview by Rufin VanRullen: Attention Cycles

For further reading:
Buschman, T.J. and Miller, E.K. (2009) Serial, covert, shifts of attention during visual search are reflected by the frontal eye fields and correlated with population oscillations. Neuron, 63: 386-396. View PDF »

Buschman,T.J. and Miller, E.K. (2010) Shifting the Spotlight of Attention: Evidence for Discrete Computations in Cognition. Frontiers in Human Neuroscience. 4(194): 1-9. View PDF »

Beta rhythms play a role in synaptic plasticity.

Zanos, S., Rembado, I., Chen, D., & Fetz, E. E. (2018). Phase-Locked Stimulation during Cortical Beta Oscillations Produces Bidirectional Synaptic Plasticity in Awake MonkeysCurrent Biology.

Super-cool paper by Andreas Nieder and crew.  Frontal-parietal beta synchrony encodes the most recent numerical input.  Theta synchrony distinguishes between different numerosities held in working memory.  The spiking of mixed-selectivity neurons multiplexed both task-relevant and irrelevant stimuli but they were separated in different phases of theta oscillations.  Powerful support that neural oscillations functionally organize spiking activty.

Jacob, S. N., Hähnke, D., & Nieder, A. (2018). Structuring of Abstract Working Memory Content by Fronto-parietal Synchrony in Primate CortexNeuron99(3), 588-597.

More evidence for mixed-selectivity in the cortex.  This time with voxels in the human brain.

Jackson, J., & Woolgar, A. (2018). Adaptive coding in the human brain: Distinct object features are encoded by overlapping voxels in frontoparietal cortex. Cortex.

Read more about mixed selectivity:
Rigotti, M., Barak, O., Warden, M.R., Wang, X., Daw, N.D., Miller, E.K., & Fusi, S. (2013) The importance of mixed selectivity in complex cognitive tasks. Nature, 497, 585-590, doi:10.1038/nature12160. View PDF »        

Fusi, S., Miller, E.K., and Rigotti, M. (2016) Why neurons mix: High dimensionality for higher cognition.  Current Opinion in Neurobiology. 37:66-74  doi:10.1016/j.conb.2016.01.010. View PDF »

Enhanced prefrontal-hippocampal spike-LFP coupling during learning of a spatial strategy (but not other strategies).

Negrón-Oyarzo, I., Espinosa, N., Aguilar, M., Fuenzalida, M., Aboitiz, F., & Fuentealba, P. (2018). Coordinated prefrontal–hippocampal activity and navigation strategy-related prefrontal firing during spatial memory formationProceedings of the National Academy of Sciences, 201720117.

A computational model of visual categorization in cortex that has properties similar to our lab’s results.  It must be true.

Abe, Y., Fujita, K., & Kashimori, Y. (2018). Visual and Category Representations Shaped by the Interaction Between Inferior Temporal and Prefrontal CorticesCognitive Computation, 1-16.

Big-ass survey of cortex by Gray and crew:

Dotson, N. M., Hoffman, S. J., Goodell, B., & Gray, C. M. (2018). Feature-Based Visual Short-Term Memory Is Widely Distributed and Hierarchically OrganizedNeuron.

Nice paper by Bressler and colleagues showing that top-down influences on visual cortex are mediated by beta-band oscillations.

Richter, C. G., Coppola, R., & Bressler, S. L. (2018). Top-down beta oscillatory signaling conveys behavioral context in early visual cortex. Scientific reports, 8(1), 6991.

Further reading on beta oscillations mediating top-down processing:
Buschman, T.J. and Miller, E.K. (2007) Top-down versus bottom-up control of attention in the prefrontal and posterior parietal cortices. Science. 315: 1860-1862  View PDF »

Bastos, A.M., Loonis, R., Kornblith, S., Lundqvist, M., and Miller, E.K. (2018)  Laminar recordings in frontal cortex suggest distinct layers for maintenance and control of working memory.  Proceedings of the National Academy of Sciences.  View PDF

 

Nice result from Buschman Lab.  Error-correcting dynamics introduce bias into working memory while reducing noise.

Error-correcting dynamics in visual working memory
Matthew F Panichello, Brian DePasquale, Jonathan W Pillow, Timothy Buschman
doi: https://doi.org/10.1101/319103

Press release for our new paper:
A heavy working memory load may sink brainwave ‘synch’

The paper:
Pinotsis, D.A., Buschman, T.J. and Miller, E.K. (2018) Working Memory Load Modulates Neuronal Coupling. Cerebral Cortex.  https://doi.org/10.1093/cercor/bhy065  View PDF

Pinotsis, D.A., Buschman, T.J. and Miller, E.K. (2018) Working Memory Load Modulates Neuronal Coupling. Cerebral Cortex, 2018 https://doi.org/10.1093/cercor/bhy065

Abstract: There is a severe limitation in the number of items that can be held in working memory. However, the neurophysiological limits remain unknown. We asked whether the capacity limit might be explained by differences in neuronal coupling. We developed a theoretical model based on Predictive Coding and used it to analyze Cross Spectral Density data from the prefrontal cortex (PFC), frontal eye fields (FEF), and lateral intraparietal area (LIP). Monkeys performed a change detection task. The number of objects that had to be remembered (memory load) was varied (1–3 objects in the same visual hemifield). Changes in memory load changed the connectivity in the PFC–FEF–LIP network. Feedback (top-down) coupling broke down when the number of objects exceeded cognitive capacity. Thus, impaired behavioral performance coincided with a break-down of Prediction signals. This provides new insights into the neuronal underpinnings of cognitive capacity and how coupling in a distributed working memory network is affected by memory load.

Freedman and Ibos give us a new general framework to think about the functions of the parietal cortex.

Freedman, D. J., & Ibos, G. (2018). An Integrative Framework for Sensory, Motor, and Cognitive Functions of the Posterior Parietal CortexNeuron97(6), 1219-1234.