Brain Rhythms in Cognition

Role of Brain Rhythms in Learning, Memory, Decision Making, and Psychiatric Disease:                                                              Implications for Novel Therapies

Thomas W. Elston

Baylor University


Neuroscientists have long recognized the existence of synchronized, rhythmic oscillations of electrophysiological activity in the brain but until recently the role of these rhythms has been poorly understood. In this paper I characterize these rhythms in terms of neural synchrony and demonstrate that (1) brain rhythms undergird decision making and working memory tasks, (2) the magnitude of neural synchrony and dysynchrony is a direct predictor of performance in cognitive tasks, (3) neural synchrony is a new way of conceptualizing neurodegenerative diseases and (4) treating a disease from the perspective of reestablishing synchrony is a validated approach with major implications for the development of novel therapies. I will conclude with future directions of research in the brain rhythm field.

Keywords: synchrony, brain rhythms, learning, memory, decision making, MIA model, schizophrenia

Schedule of Contents


I.            Introduction

II.            Synchrony and Decision Making

III.            Synchrony and Working Memory

IV.            Synchrony and Schizophrenia

V.            Synchrony and Other Psychiatric Disease

VI.            Future Directions

VII.            References 

I. Introduction

Neuroscientists strive to understand the relationship between behavior and the physical workings of the brain. The circuitry of the brain and its functional organization gives rise to our consciousness and, to a great extent, mediates our quality of life. Therefore it is crucial to consider: how do our brains organize themselves? What does it look like when this goes wrong? And can anything be done to rescue brains with dysfunctional processing? In light of the worldwide $513.65 billion dollar burden brain disease places on society each year (MIT, 2013), movement towards answers is imperative.

The first progress towards answers came in Hebb’s (1949) book The Organization of Behavior: A Neuropsychological Theory. In it Hebb puts forth his now famous Hebbian Postulate: cells that fire together wire together. The diagram, detailing synapse formation in Pavlovian conditioning, and data below will clarify the advancement in modern terms (Dickerson, Restieaux, & Bilkey, 2012).

image001 image003



Let’s call the “Bell” cell Cell A and the “Salivation” cell Cell B. Hebb proposed that when an axon of cell A is near enough to excite Cell B and repeatedly or persistently take part in firing it, metabolic changes occur in one or both cells such that Cell A’s efficacy, as one of the cells firing Cell B, is increased. Long-since validated, and renamed Long-Term Potentiation (LTP), the Hebbian Postulate is the basis for modern neuroelectrophysiological investigations of learning, memory, decision-making, and pathologies associated with neurological disease. The Hebbian Postulate provided the first traction for the principle by which the brain organizes itself: neural synchrony. Synchrony, coherent action potential firing between anatomically distinct brain regions, is the binding mechanism that integrates neural processing in the brain. Electrophysiological measures of synchrony, such as the data on the right, inform us as to how the various brain regions work together to facilitate a fluid and unified perception of the world.

II. Synchrony and Decision Making

Benchenane and coworkers (2010) observed the developmental trajectory of neural synchrony between CA1 hippocampus and medial prefrontal cortex in rodents in the Y-maze decision making task through a combination of single-unit and EEG techniques over 60 experimental sessions in four rats placed in a two-contingency Y-maze decision task (first go to the right arm and then go to the randomly selected lit arm). Medial prefrontal electrophysiological data was collected via multi-spike single-unit recording from 1475 medial prefrontal neurons and CA1 hippocampal event-related field potentials were measured via EEG before, during, and after the learning process. Coherence was assessed as the synchronized action potential firing in the two brain regions. Prior to the experimental learning, hippocampal-neocortical theta rhythm coherence was 40±4ms, meaning that once every 40ms the medial prefrontal cortex and CA1 hippocampus concomitantly depolarized. During and after training, coherence in single-unit spike trains was 30ms, demonstrating that during high synchrony periods theta-band modulated cell pairs tend to fire together within the same oscillatory cycle more frequently. Put more simply, the brain regions synchronized their rates of activity and as a result of that synchronization depolarized at an elevated rate.

Theta coherence peaked when the rat was at the choice point (i.e. the vertex of the Y-maze) and increased directly with acquisition of the rule and behavioral measures of performance. In other words, as the rate of medial prefrontal cortex synchronized to the rate of CA1 activity, performance of the learned rule improved. Interestingly, the initial hippocampal-mPFC-theta-coherence at the choice point strongly predicted the rat’s learning performance: rats displaying higher initial coherence at the choice point in the learning stage had, on average, better performance. Elevated coherence/performance was not a function of greater incidences of reward following rule acquisition: coherence was not significantly different between rewarded and unrewarded trials.

Neuroplasticity, the brain’s capacity to modify itself as a function of experience, includes neural synchrony. Synchronicity is a valid measure of the neural correlates of learning, binding widely distributed sets of neurons into functionally coherent ensembles and may serve as a component of the framework undergirding the functional organization and reorganization of the brain.

III. Syncrony and Working Memory

Hyman and colleagues (2010) investigated the role of synchrony in functional connectivity between CA1 hippocampus and medial prefrontal cortex in a rodent delayed-matching-to-sample task, measuring working memory (n=3). Taking a more computational, integrative approach, using electrophyiological methods similar to Benchenane et al. (2010)), the authors were able to able to statistically isolate individual cases of depolarization within reciprocally connected neurons in CA1 and mPFC and identify the percentage of neurons which respond to failed, successful, and were outcome independent of trials. 74 reciprocally connected cells met the criteria for detailed analysis. Over all correct trials 46% (34/74) of cells were uniquely theta-entrained, meaning that the peaks of the depolarization in CA1 and mPFC units were synchronized. Over all error trials 17% (13/74) of cells were uniquely theta-entrained. 45% (33/74) were entrained during both successful and failed trials. Observed synchrony in failed trials was nearly 93% different (smaller) than successful trials, indicating that the breakdown of synchrony underlies the breakdown of functional connectivity which then produces the errors in behaviour.

Consistent with Benchenane et al. (2010), the CA1 was determined to be the synchronizing agent and the magnitude of theta-band synchrony, more so than the actual depolarization rate, was a very strong predictor of performance in the working memory task. The authors suggest that the failure-theta-entrained cells serve as a synchrony reset circuit. Recalling the large reciprocal connectivity between CA1 and mPFC and a significant proportion of neurons (the 17% associated with failed trials) are inversely synchronized with the 46% of neurons associated with successful trials, there is an empirical platform for this hypothesis. Taken together, these results suggest that phase-locking (i.e. theta-entrainment, synchrony) of different regions directs attention towards relevant internal representations and may represent the functional connectivity giving rise to working memory. The loss of phase-locking at the onset of what became failed trials may represent a shift of attention away from those representations.

IV. Synchrony and Schizophrenia

Given that synchrony is deeply involved in learning, memory, decision making, and the establishment of functional connectivity, the question becomes: how this knowledge can be practically applied? Bilkey and his New Zealander coworkers offer a meaningful first step towards the practical biomedical application of synchrony. Bilkey researches the role of neural synchrony in the rodent maternal immune activation (MIA) model of schizophrenia. The model consists of injecting a poly IC virus into the pregnant dam’s womb which induces a robust immune response. Using a high throughput microRNA analysis to determine the molecular genetic effects of maternal immune responses, Overeem et al. (2013) determined that the MIA model produces functional changes in 148 genes, most notably those associated with the KEGG and mTOR signaling pathways, which, among other things, code for axonal guidance proteins, glutamatergic receptors, and dopaminergic receptors. Functional changes in these genes are well in line with the current glutamate-dopamine hypothesis of schizophrenia, which postulates the disorder as a dramatic upregulation of these transmitter systems (Bradford, 2009; Reynolds, 2008). The MIA model also produces the pre-pulse inhibition deficits in both juvenile and adult rats (Wolff and Bilkey, 2010) hallmarking of schizophrenic behaviour. Given that the MIA model is well characterized and fits well within the accepted molecular and behavioural profiles of schizophrenia, it is a strong candidate model for the study of synchrony in neurodegenerative disease.

Using a unique fixed single-unit microdrive multi-shank developed by Bilkey (2003), Dickerson et al. (2012) investigated the aberrance and reestablishment of synchrony in the MIA rodent model via acute clozapine administration using in-vivo single-unit recording from the CA1 hippocampus and medial prefrontal cortex. Nine MIA rodents were bred specifically for use in this experiment and eight control animals were used (17 subjects total). Adult (>5 months of age) animals were anaesthetized and implanted with two non-moveable recording microelectrodes in the medial prefrontal cortex (+3.2mm AP; -.06mm ML to bregma; 3.2mm DV from dura mater) and the CA1 region of the hippocampus (-3.8mm AP; -2.5mm ML to bregma; 2.5mm DV from dura mater). Animals were given two weeks recovery and then food deprived to 85% of ad-libitum weight. The animals were then placed in a black, plastic, circular tub (d=74cm; height=56cm) for 100 minutes while single-unit recordings were taken in 10 minute phases (this was to establish a baseline level of synchrony) while the animal foraged for chocolate hail (a treat). Following this baseline phase either clozapine or a placebo (saline) was acutely administered (IP) at 1mg/kg and 5mg/kg doses. Fifteen minutes after the injection, recordings were taken for 30 minutes. Raw data, the instances of depolarization by each single neuron being recorded from in mPFC and CA1, was fed directly into and analyzed with Matlab (Mathworks, Natick, MA) to determine the synchronized co-occurrence of activity between these regions. The investigators found that clozapine enhances long-range synchrony, the synchronized rates of depolarization of anatomically distant brain regions, in a dose dependent manner. This is the first investigation that has approached both the conceptualization and treatment of a neurodegenerative disorder as matters of re-establishing synchrony. The importance of this study and its implications for synchrony are realized when one considers the previous explanations for variability for both schizophrenia and clozapine.

Clozapine, first synthesized in 1958 by the Swiss pharmaceutical company Wander AG, is an atypical antipsychotic, meaning that the ligand has a strong binding affinity for both dopaminergic and serotinergic receptors (Crilly, 2007) and is an antagonist to glutamatergic NMDA receptors (Xi et al., 2011).  Clozapine’s rapid adoption as a therapeutic is due to the lack of extra-pyramidal symptoms (EPS), Parkinsonian-like movement disorders characteristic of antipsychotics, and being a fully effective antipsychotic. Recalling the dopamine-glutamate hypothesis of schizophrenia, that schizophrenic behaviour arises as a result of deeply misregulated dopaminergic and glutamatergic neurotransmission (c.f. Bradford, 2009; Reynolds, 2008), clozapine fits the pharmacological bill for schizophrenia. And this is the extent to which previous literature was able to account for clozapine’s efficacy: it impacts dopaminergic and glutamatergic neurotransmission which alleviated the various behavioural symptoms associated with schizophrenia. The problem is that this explanation accounts only for clozapine’s mechanism of action but does not address the deeper circuitry of its efficacy.

Dickerson et al. (2012) recharacterized schizophrenia with an electrophysiological, rather than pharmacological, profile. This recharacterization of schizophrenia was able to account for the clozapine’s efficacy in terms of synchrony, an emergent property of neural networks, which is vastly more complex with more explanatory power than single molecule explanations because it addresses the interconnected nature of the brain. Their study validates conceptualizing of a neurodegenerative disease as an aberrance of synchrony because the reestablishment of synchrony, via clozapine, led to the alleviation of behavioural symptoms. This is a vastly new “top-down” paradigm for approaching the treatment of neurodegenerative disorder, which has traditionally focused on the alleviation of single molecule or single gene aberrances (a “bottom-up” paradigm).

V. Synchrony and Other Psychiatric Diseases

            Using a variety of electrophysiological techniques similar to  those mentioned above, Ulhlhaas and Singer (2006) have created characteristic profiles for the aberrance of synchrony in epilepsy, autism, Alzheimer’s Disease, Parkinson’s Disease, schizophrenia, and other pathological brain states. The authors contend that measures of neural synchrony will become increasingly important for the diagnosis of neuropsychiatric diseases. Already the analysis of changes in the brain rhythms of epileptic patients has proven a strong predictor of seizures (Le Van Quyen et al., 2003). If we are to approach neurodegenerative diseases anew from the perspective of synchrony, future research into the mechanisms which give rise to synchrony is imperative.

VI. Future Directions

            In the world of science, knowing the right questions to ask is often just as important as their answers. Buzsaki et al. (2013) has identified several future research questions key to the preservation of brain rhythms. Coming synchrony research must address (1) how modulatory GABAergic interneurons precisely modulate principle neurons, (2) the identification of the different types of pyramidal neurons and their characteristic connectivity as a function of cerebral layer, (3) the patterns of connectivity within local circuits, and (4) the functional connectivity between anatomically distinct processing modules. The work ahead will require investigations across many species and the development of large scale electrophysiological methods capable of resolving the activity of single units at each instance of depolarization. The challenges ahead are as enormous as the implications: an alternative to brute force pharmaceutical approaches in the treatment of neuropsychiatric disorders. In keeping with previous work demonstrating that the reestablishment of normal brain rhythms leads to the restoration of normal behavior (Dickerson et al., 2012) new treatments, beyond drugs, might include pattern-guided, closed-loop deep-brain stimulation, sensory feedback, and transcranial magnetic and electrical stimulation. The future of brain rhythm research is bright and expansive with vast implications for both researchers and clinicians.

VII. References

Benchenane, K., Peyrache, A., Khamassi, M., Tierney, P. L., Gioanni, Y., Battaglia, F. P., & Wiener, S. I. (2010). Coherent theta oscillations and reorganization of spike timing in     the hippocampal prefrontal network upon learning. Neuron, 66, 921-936.

Bilkey, D. K., Russel, N., & Colombo, M. (2003). A lightweight microdrive for single-unit recording in freely moving rats and pigeons. Methods, 30, 152-158.

Bradford, A. (2009). The dopamine and glutamate theories of schizophrenia: A short review. Current Anaesthesia & Critical Care, 20, 240-241.

Buzsaki, G., Logothetis, N., & Singer, W. (2013). Scaling brain size, keeping timing: Evolutionary preservation of brain rhythms. Neuron, 410, 751-764.

Crilly, J. (2007). The history of clozapine and its emergence in the us market: a review and analysis. History of Psychiatry, 18, 39-60.

Dickerson, D. D., Restieaux, A. M., & Bilkey, D. K. (2012). Clozapine administration ameliorates disrupted long-range synchrony in a neurodevelopmental animal model of schizophrenia. Schizophrenia Research, 135, 112-115.

Horváth, S., & Mirnics, K. (2013). Immune system disturbances in schizophrenia. Biological Psychiatry,1, 1-8.

Hyman, J. M., Hasselmo, M. E., & Seamans, J. K. (2011). What is the functional relevance of prefrontal cortex entrainment to hippocampal theta rhythms? Frontiers in       Neuroscience5, 1-13.

Hyman, J. M., Zilli, E. A., Paley, A. M., & Hasselmo, M. E. (2010). Working memory performance correlates with prefrontal-hippocampal theta interactions but not with prefrontal neuron firing rates. Frontiers in Integrative Neuroscience, 4, 1-13.

Le Van Quyen,M., Navarro, V., Martinerie, J., Baulac, M., and Varela, F.J. (2003). Toward a neurodynamical understanding of ictogenesis. Epilepsia 44, 30–43.

MIT. (2013, March 13). Brain disorders: by the numbers. Retrieved from

Overeem, K., Wolff, A. R., Bilkey, D. K., & Williams, J. M. (2013, November). In Larry Swanson (Chair). Aberrant expression of microrna in blood plasma and CA1 of adult    animals exposed to maternal immune activation. Presentation at the Society for            Neuroscience.

Reynolds, G. P. (2008). The neurochemistry of schizophrenia. Psychiatry, 7, 425-429.

Uhlhaas, P. J., & Singer, W. (2006). Neural synchrony in brain disorders: Relevance for cognitive dysfunctions and pathophysiology. Neuron, 52, 155-168.

Wolff, A. R., & Bilkey, D. K. (2010). The maternal immune activation (mia) model of schizophrenia produces pre-pulse inhibition (ppi) deficits in both juvenile and adult rat but these effects are not associated with maternal weight loss. Behavioural Brain Research, 213, 323-327.

Xi, D., Li, Y. C., Snyder, M. A., Gao, R. Y., Adelman, A. E., Zhang, W., Shumsky, J. S., & Gao, W. J. (2011). Group ii metabotropic glutamate receptor agonist ameliorates mk801- induced dysfunction of nmda receptors via the akt/gsk-3b pathway in adult rat prefrontal cortex. Neuropsychopharmacology, 36, 1260-1274.