The gamma band effect for episodic memory encoding is absent in epileptogenic hippocampi
Introduction
Episodic memory, in which specific memory items are placed within temporal context during encoding and retrieval, strongly depends on the contribution of the hippocampus (Burgess et al., 2002, Tulving, 2002). Correspondingly, episodic memory shows selective and marked degradation in patients with temporal lobe epilepsy (Dupont et al., 2000). The implantation of electrodes in human patients to localize seizures has provided memory researchers with the opportunity to explore patterns of brain activity that occur during episodic memory encoding with high spatial and temporal resolution (Jacobs and Kahana, 2010). By comparing features of brain activity during successful versus unsuccessful item encoding, researchers can identify oscillatory patterns that uniquely predict memory formation (Sederberg et al., 2003, Sederberg et al., 2007). These patterns are termed subsequent memory effects (SME).
In the human hippocampus, the most robust and consistent SME is an increase in gamma band (40–200 Hz) oscillatory power during successful item encoding (Sederberg et al., 2007). This gamma band effect is considered a positive SME because power is relatively higher when memories are formed versus when they are not. At other frequency bands by contrast, changes in power often show the opposite pattern, with decreased power during successful encoding (a negative SME). Using invasive recordings, a hippocampal gamma band effect has also been observed during working memory tasks, spatial navigation, and autobiographical memory in human participants (Ekstrom et al., 2005, Fell et al., 2001, van Vugt et al., 2010, Axmacher et al., 2007). These studies have generally excluded abnormal hippocampi, especially epileptogenic hippocampi, because these investigations typically seek to establish generalizable human patterns analogous to animal data. There is evidence of diminished BOLD responsiveness in abnormal hippocampi in epilepsy patients during episodic memory; this effect may predict memory decline after surgery (Jokeit et al., 2001). An analysis of recognition memory in epilepsy patients identified differences in interictal spiking throughout the brain for recognized memory items, suggesting that memory processes interact with epileptic networks at the neurophysiological level (Matsumoto et al., 2013). However, no direct comparison of the gamma band SME between epileptogenic and non-epileptogenic hippocampi has been undertaken.
To examine the gamma band SME in epilepsy, we analyzed a set of nine patients (ten hippocampal implantations, age 18–62, seven female) who underwent stereoelectroencephalography (sEEG) for seizure localization and then performed an episodic memory task. We compared activity during the encoding of episodic memories between epileptogenic and non-epileptogenic hippocampi, hypothesizing that the gamma band SME would be absent or diminished in the epileptogenic hippocampi. We demonstrate that gamma band activity is significantly different between the two groups: in epileptogenic hippocampi gamma power decreases during successful item encoding, while non-epileptogenic hippocampi exhibit the expected + SME. We discuss some theoretical underpinnings and implications for this finding.
Section snippets
Methods
Electrode implantation was performed using the ROSA stereotactic robot after a double-contrast high resolution volume acquisition MRI (Gonzalez-Martinez et al., 2013). Patients were tested from 48 h after implantation when clinical obligations permitted and when they had not suffered seizure activity for at least four hours while awake and fully able to participate. Sessions were not initiated if patients were drowsy or otherwise unwilling to engage in the task. Sessions in which a patient
Results
To directly compare the gamma band subsequent memory effect between epileptogenic and non-epileptogenic hippocampi, we recorded hippocampal activity from nine patients undergoing pre-operative sEEG for seizure localization (eight unilateral, one bilateral implantation). The patients received an average of 10.6 implanted arrays, all inserted with millimeter accuracy using the ROSA stereotactic robot (Gonzalez-Martinez et al., 2013). The ictal onset location and pattern of spread were determined
Discussion
Our data show for the first time that a robust pattern of oscillatory activity seen during human memory formation, the hippocampal gamma band SME, is absent in epileptogenic hippocampi. In humans, a negative gamma band effect has been observed only rarely for the lowest edge of the gamma frequency range (Fell et al., 2001, Sederberg et al., 2007); gamma frequencies above 40 Hz exhibit a reliable positive effect in the non-epileptogenic hippocampi that comprise the dataset for existing analyses (
Conflict of interest
None of the authors have potential conflicts of interest to be disclosed.
References (38)
- et al.
The human hippocampus and spatial and episodic memory
Neuron
(2002) - et al.
Mechanisms of gamma oscillations in the hippocampus of the behaving rat
Neuron
(2003) - et al.
High-frequency oscillations – where we are and where we need to go
Prog Neurobiol
(2012) Freesurfer
Neuroimage
(2012)- et al.
Direct brain recordings fuel advances in cognitive electrophysiology
Trends Cogn Sci
(2010) - et al.
Single pulse electrical stimulation of the hippocampus is sufficient to impair human episodic memory
Neuroscience
(2010) - et al.
Observing the transformation of experience into memory
Trends Cogn Sci
(2002) - et al.
Localization of interictal spikes using sam (g2) and dipole fit
Neurol Clin Neurophys
(2003) - et al.
Entrainment of neocortical neurons and gamma oscillations by the hippocampal theta rhythm
Neuron
(2008) - et al.
Comparison of spectral analysis methods for characterizing brain oscillations
J Neurosci Meth
(2007)
Sustained neural activity patterns during working memory in the human medial temporal lobe
J Neurosci
Episodic memory in left temporal lobe epilepsy: a functional MRI study
Brain
Human hippocampal theta activity during virtual navigation
Hippocampus
High-frequency oscillations: what is normal and what is not?
Epilepsia
Human memory formation is accompanied by rhinal-hippocampal coupling and decoupling
Nat Neurosci
N-acetylaspartate and creatine levels measured by 1h mrs relate to recognition memory
Neurology
Stereoelectroencephalography in the difficult to localize refractory focal epilepsy: early experience from a north american epilepsy center
Epilepsia
High frequency oscillations: the new eeg frontier?
Epilepsia
Hyperactivation of parahippocampal region and fusiform gyrus associated with successful encoding in medial temporal lobe epilepsy
Epilepsia
Cited by (23)
Theta and gamma hippocampal–neocortical oscillations during the episodic-like memory test: Impairment in epileptogenic rats
2022, Experimental NeurologyCitation Excerpt :Multiple investigations have shown that cell death in the hippocampus and related structures leads to a reorganization of circuits essential for encoding and retrieval, which may be the main cause of cognitive dysfunction in temporal lobe pathology (Glowinski, 1973; Jokeit and Ebner, 1999; Helmstaedter, 2002). Recently, it has been suggested that theta/gamma coupling plays a critical role in episodic memory formation in the healthy brain and the lack of this coordination in the epileptogenic hippocampus (Shirvalkar et al., 2010; Lega et al., 2015, 2016). However, there is much to be learned and discussed: the oscillatory processes underlying the episodic memory and their disturbances in seizure pathology remain poorly understood.
An inventory of basic research in temporal lobe epilepsy
2021, Revue NeurologiqueCitation Excerpt :For the generation of seizures, the hippocampus might play a key role in the cognitive impairments that are associated with TLE. First, various lines of evidence suggest that physiological hippocampal theta and gamma band rhythms, which are substrates for several cognitive functions, are disturbed in TLE [28–30]. Second, hippocampal sclerosis seems to be an important correlate for the severity of memory decline [31,32].
Inhibition and oscillations in the human brain tissue in vitro
2019, Neurobiology of DiseaseCitation Excerpt :These pathological high-frequency oscillations (HFOs) are associated with epileptogenic areas and seizure onset zones and linked with hippocampal sclerosis (Staba et al., 2007; Zijlmans et al., 2012). Other oscillatory bands, including theta and gamma, are equally impaired in the epileptic hippocampus and slow theta-gamma modulation supporting episodic memory encoding is deficient in the human TLE (Lega et al., 2015). In experimental models of TLE, spatiotemporal features of theta, gamma and ripple oscillations are associated with a range of cognitive co-morbidities (Lopez-Pigozzi et al., 2016; Shuman et al., 2017; Valero et al., 2017).
Task-induced gamma band effect in type II focal cortical dysplasia: An exploratory study
2018, Epilepsy and BehaviorCitation Excerpt :In studies using task-induced HFA to identify large-scale functional networks in different experimental paradigms, it is usually recommended to consider only recording sites away from the epileptic zone and signals free of epileptiform activity such as epileptic spikes and to concentrate on results from functionally intact regions as assessed by neuropsychological testing and neuroimaging [10]. Even in studies investigating cognitive functions within the epileptogenic zone such as episodic memory in patients with mesial temporal lobe epilepsy [22], electrodes exhibiting interictal or ictal activity were excluded from the analysis. In this context, the possibility of using task-induced HFA for functional mapping in patients with type II FCD might be difficult for several reasons.