The Miller Lab

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?

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.

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 »

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

Interesting study showing that there are decreases in the frequency of alpha oscillations when a task requires require integration of two inputs that are separated in time.  The slowing fosters integration by making it more likely that two stimuli fall within one alpha cycle and are thus integrated.  Cool.

Wutz, A., Melcher, D., & Samaha, J. (2018). Frequency modulation of neural oscillations according to visual task demands. Proceedings of the National Academy of Sciences, 201713318.

The authors find that dopamine increased power of beta-low gamma oscillations in cortex.  During visual stimulation, dopamine increased information encoding over a wide range of frequencies but most prominently in the feedforward supragranular layers and in the gamma band (50-100 Hz).

Zaldivar, D., Goense, J., Lowe, S. C., Logothetis, N. K., & Panzeri, S. (2018). Dopamine Is Signaled by Mid-frequency Oscillations and Boosts Output Layers Visual Information in Visual Cortex. Current Biology.

This must be correct.  It is very remarkably consistent with our recent study 🙂
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. doi:10.1073/pnas.1710323115   View PDF

as well as with our previous work showing that gamma is associated with bottom-up 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 »

A thoughtful review and discussion of the issues involved in analyzing brain rhythms.

Jones, S. R. (2016). When brain rhythms aren’t ‘rhythmic’: implication for their mechanisms and meaning. Current opinion in neurobiology, 40, 72-80.