Theta–gamma coupling during a working memory task as compared to a simple vigilance task
Highlights
► Theta–gamma coupling (TGC) was analyzed as measured by scalp EEG. ► A visuo-spatial 2-back task and a control task (vigilance task) were performed. ► An instantaneous phase method proposed by Cohen was used. ► TGC increased during the working memory task in the frontal area. ► Differences in TGC were observed at high-gamma frequency.
Introduction
Systemic modulation of the power of gamma oscillations over the course of the theta cycle suggests that there is a relationship between theta–gamma coupling (TGC) and working memory processes [7]. Cross-frequency interaction between theta and gamma oscillations during working memory operations (theta phase–gamma power interaction) is consistent with the model of Lisman and Idiart [14], which holds that gamma oscillations divide the theta cycle into “time slots” that segment several active representations in time. The set of items is reactivated within each theta cycle at a rate of one item per gamma cycle.
Studies involving intracranial electroencephalography (EEG) recordings in animals and humans suggest that enhanced theta gamma coupling is associated with working memory processes. For example, the firing rate of place cells, which respond when an animal is in a specific location [17], is modulated by the phase of gamma band activity [9], [20]. Intracranial EEG recordings in patients with epilepsy [1], [12], [16], [19] indicate that cross-frequency coupling is enhanced during working memory maintenance [1] and related to memory encoding and retrieval in the rhinal cortex and hippocampus [16]. However, it has not been established whether TGC as recorded by scalp EEG reflects working memory processes.
Previously, we reported that TGC in the parietal area correlates significantly with both visuo-spatial short-term memory scores and accuracy in a short-term memory task in healthy older individuals at low-gamma frequency (30–60 Hz) [18], suggesting that TGC as recorded by scalp EEG reflects memory processes. However, additional confirmatory research is necessary. First, “working memory” refers to a brain system that provides temporary storage and manipulation of information relevant to complex cognitive tasks such as language comprehension, learning, and reasoning [2]. The cognitive component that we evaluated in our previous study was short-term memory rather than working memory per se. Second, it is not possible to determine based on our previous study [18] whether increased TGC reflects working memory processes or simple vigilance, as there was no control task to assess simple vigilance. Considering that gamma oscillatory activity and theta–gamma phase locking is affected by attention as well as memory [11], [15], it is important to clarify this point in order to better understand the properties of TGC. Finally, because the participants in our earlier study were older adults, the generalizability of the findings to people of other ages is unclear.
We hypothesized that the degree of TGC as measured by scalp EEG would increase during working memory processing as compared to simple vigilance. To test this hypothesis, we measured TGC while participants completed a working memory task and a vigilance task and examined differences in the TGC index at approximately 40 Hz (reference), the frequency known to reflect memory processes as measured by scalp EEG.
In addition, previous studies involving the use of intracranial EEG have reported a relationship between TGC and memory processing at high-gamma frequency (80–200 Hz). However, there are no reports to date regarding whether high gamma with such small amplitudes can survive and be reliably measured when passing through the skull and scalp. Thus, for explorative purposes, we also analyzed and compared TGC in high-gamma areas during a working memory task and a control vigilance task.
Section snippets
Methods
Thirteen right-handed volunteers (5 females; age, 20.8 ± 3.1 years; education, 12.9 ± 1.8 years) participated in this study. All participants were screened for past or present medical, neurological, or psychiatric illness. This study was performed following the ethical guidelines of the Institutional Review Board of Severance Mental Health Hospital and the Declaration of Helsinki.
All participants performed a visuo-spatial working memory task (2-back task) and a control task (vigilance task) during
Results
SIm at approximately 40 Hz, the frequency of interest, increased in the frontal area during the working memory task as compared to the vigilance task. SIm at the FP2 electrode increased significantly under the FDR threshold during the working memory task as compared to the vigilance task [t(12) = 2.558, p = 0.025 uncorrected] (Fig. 1a and Supplementary Table 1). Theta and gamma power did not differ between the two tasks at any electrode (Fig. 1b and c and Supplementary Table 1).
The mean hit rate was
Discussion
In this study, we explored low-gamma frequencies of approximately 40 Hz, which are well known to reflect memory processes [11], [21], was enhanced significantly during a working memory task vs. a vigilance task. This result can be understood in the same context as previous findings that TGC reflects memory processes [1], [7], [16], [19], and suggests that TGC as measured by scalp EEG reflects working memory processes specifically. TGC was enhanced in the frontal area during the working memory
Acknowledgements
This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (No. 2010-0026833). The funding source had no role in the design of the study, data collection, data analysis, data interpretation, writing of the report, or the decision to submit the paper for publication.
References (21)
- et al.
A breakthrough in neuroscience needs a “Nebulous Cartesian System” Oscillations, quantum dynamics and chaos in the brain and vegetative system
International Journal of Psychophysiology
(2007) - et al.
Oscillations in the prefrontal cortex: a gateway to memory and attention
Current Opinion in Neurobiology
(2011) - et al.
Coupled oscillations mediate directed interactions between prefrontal cortex and hippocampus of the neonatal rat
Neuron
(2011) Assessing transient cross-frequency coupling in EEG data
Journal of Neuroscience Methods
(2008)- et al.
Mechanisms of gamma oscillations in the hippocampus of the behaving rat
Neuron
(2003) - et al.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
Journal of Neuroscience Methods
(2004) - et al.
Cognitive functions of gamma-band activity: memory match and utilization
Trends in Cognitive Sciences
(2004) - et al.
Frontal-midline theta from the perspective of hippocampal “theta”
Progress in Neurobiology
(2008) - et al.
The relationship between theta–gamma coupling and spatial memory ability in older adults
Neuroscience Letters
(2011) - et al.
Brain oscillatory substrates of visual short-term memory capacity
Current Biology
(2009)
Cited by (36)
Frontal midline theta and cross-frequency coupling during short term memory and resting state
2022, Neuroimage: ReportsIs short-term memory capacity (7±2) really predicted by theta to gamma cycle length ratio?
2021, Behavioural Brain ResearchCitation Excerpt :The results of several previous studies suggest cross-frequency coupling as a promising candidate when looking for a link between EEG activity and short-term memory capacity. This parameter should be measured and calculated both for resting state and as a functional measurement during cognitive tasks and is nowadays often used as a proxy for brain activity during similar types of experiments (Barr et al., 2017; Park et al., 2013; Park et al., 2011; Rajji et al., 2017, and others). Kristýna Malenínská: conceptualization, data curation, writing – original draft, writing - review & editing.
Training state and performance evaluation of working memory based on task-related EEG
2019, Biomedical Signal Processing and ControlImpaired theta-gamma coupling during working memory performance in schizophrenia
2017, Schizophrenia ResearchCitation Excerpt :It is hypothesized that gamma oscillations encode items of information and modulation of gamma amplitude by theta phase (theta-gamma coupling) order these items providing a “neural code” for WM (Lisman and Idiart, 1995; Lisman and Jensen, 2013). Emerging lines of evidence from animal (Tort et al., 2009; Shirvalkar et al., 2010; Li et al., 2012), electrocorticography (Mormann et al., 2005; Canolty et al., 2006; Axmacher et al., 2010; Jacobs and Kahana, 2010; Khursheed et al., 2011), and electroencephalography (EEG) (Park et al., 2013; Rajji et al., 2016) have demonstrated the role of coupling with WM performance. Recently, coupling has been suggested to be important for the correct ordering of information during N-Back performance (Rajji et al., 2016) whereby coupling independently contributed to WM load on trials which required ordering of information (i.e., target correct trials) compared to trials in which the correct order was neither necessary nor advantageous in healthy participants (i.e., non-target correct trials; (Rajji et al., 2016)).
Resting-state EEG gamma power and theta–gamma coupling enhancement following high-frequency left dorsolateral prefrontal rTMS in patients with depression
2017, Clinical NeurophysiologyCitation Excerpt :Indeed, a heightened modulation index (MI) of TGC was observed in the hippocampal region (Tort et al., 2009) and medial prefrontal cortex (Li et al., 2012) of rats following learning and memory tasks. The strength of MI was also enhanced during a working memory task in healthy humans (Park et al., 2013). To date, there have been no clinical studies investigating the resting-state gamma power and MI of TGC changes before and after rTMS treatment in patients with depression, which may be associated with clinical and cognitive outcomes.