Spectral Analysis for Physical Applications

Percival, Donald B.; Walden, A. T.

doi:10.1017/cbo9780511622762

Cited by 1,810 publications

(1,089 citation statements)

References 0 publications

Supporting

Mentioning

1,073

Contrasting

Unclassified

Order By: Relevance

“…Spectral analysis provides a unified framework for the characterization of hybrid processes and we use multitaper methods of spectral estimation developed in Thomson (1982) to construct estimators for all spectral quantities (see Appendix). These estimates are presented and evaluated in (Percival and Walden, 1993) and are applied to a wide range of neurobiological data in (Mitra and Pesaran, 1999). Correlation function measures such as the auto-and cross-correlation function characterize the same statistical structure in time series as spectra and cross-spectra, however, as we discuss in a later section, spectral estimates offer significant advantages over their time domain counterparts.…”

Section: Resultsmentioning

confidence: 99%

Temporal structure in neuronal activity during working memory in macaque parietal cortex

et al. 2002

View full text Add to dashboard Cite

A number of cortical structures are reported to have elevated single unit firing rates sustained throughout the memory period of a working memory task. How the nervous system forms and maintains these memories is unknown but reverberating neuronal network activity is thought to be important. We studied the temporal structure of single unit (SU) activity and simultaneously recorded local field potential (LFP) activity from area LIP in the inferior parietal lobe of two awake macaques during a memory-saccade task. Using multitaper techniques for spectral analysis, which play an important role in obtaining the present results, we find elevations in spectral power in a 50-90Hz (gamma) frequency band during the memory period in both SU and LFP activity. The activity is tuned to the direction of the saccade providing evidence for temporal structure that codes for movement plans during working memory. We also find SU and LFP activity are coherent during the memory period in the 50-90Hz gamma band and no consistent relation is present during simple fixation. Finally, we find organized LFP activity in a 15-25Hz frequency band that may be related to movement execution and preparatory aspects of the task. Neuronal activity could be used to control a neural prosthesis but SU activity can be hard to isolate with cortical implants. As the LFP is easier to acquire than SU activity, our finding of rich temporal structure in LFP activity related to movement planning and execution may accelerate the development of this medical application.Keywords: parietal, prosthesis, local field potential, gamma band, coherence, temporal structure.Pesaran et. al. 3Working memory is a brain system requiring the temporary storage and manipulation of information necessary for the performance of complex cognitive tasks (Baddeley, 1992). The neurophysiological basis of working memory is studied in non-human primates by recording neural activity during delayed-response tasks (Fuster, 1995). Cue-selective elevated single unit firing rates have been recorded during the delay period in many brain areas during different versions of the task (Fuster and Jervey, 1982;Bruce and Goldberg, 1985;Gnadt and Andersen, 1988;Miyashita and Chang, 1988;Funahashi et al., 1989;Koch and Fuster, 1989;Miller et al., 1996;Zhou and Fuster, 1996). How this neural activity is sustained is unknown but may be important to understanding the neural basis of working memory (Goldman-Rakic, 1995). Converging evidence points to the importance of a distributed recurrent neuronal network (Goldman-Rakic, 1988) and reverberating network activity has long been suggested as a possible mechanism for short-term memory (Lorente de No, 1938;Hebb, 1949;Amit, 1995;Seung, 1996;Wang, 1999).Measures with the potential to capture correlated neural activity on a millisecond time scale may be needed to resolve reverberating memory activity. The dynamical structure of neuronal activity has been the source of much interest as a temporal code (for a review see Singer and Gray (1995) ) however ...

show abstract

Section: Resultsmentioning

confidence: 99%

Temporal structure in neuronal activity during working memory in macaque parietal cortex

et al. 2002

View full text Add to dashboard Cite

show abstract

“…Spectra were calculated with a window length of 2 s, fast Fourier transform (FFT) length of 8 s, and bandwidth parameter nw = 2 and k = 3 tapers (Percival and Walden, 1993). Spectral peaks were determined, and the peak position was classified with respect to the frequency bands: theta (4-8 Hz), alpha (8-13 Hz), beta (13-30 Hz).…”

Section: Power Spectral Densitymentioning

confidence: 99%

EEG alpha distinguishes between cuneal and precuneal activation in working memory

Michels¹,

et al. 2008

View full text Add to dashboard Cite

EEG alpha distinguishes between cuneal and precuneal activation in working memory AbstractIn the literature on EEG during working memory (WM), the role of alpha power (8-13 Hz) during WM retention has remained unclear. We recorded EEG while 18 subjects retained sets of consonants in memory for 3 s; setsize (ss4, ss6, ss8) determines memory workload. Theta power (4-8 Hz) increased with workload in all subjects in middle frontal electrodes. Using ICA, the increase in theta could be attributed to one component whose generators were localized by sLORETA in the medial frontal gyrus. Alpha power in parietal electrode Pz showed a mean increase during retention as compared to prestimulus fixation (event-related synchronization, ERS). On an individual basis, alpha power increased with workload in 9 subjects (WL+ group) and decreased in 9 subjects (WL-group). The alpha increased in upper alpha for the WL+ group (mean: 10.4 Hz) and decreased in lower alpha for the WLgroup (mean: 8.9 Hz). Time-frequency representations show high alpha power early during retention for the WL+ group and high alpha power late during retention for the WL-group. sLORETA revealed maximal contrast for the WL+ group in the cuneus and for the WL-group in the precuneus. In subjects with WL+, alpha increase in the cuneus may reflect WM maintenance or active inhibition of task-irrelevant areas. In subjects with WL-, alpha decrease in the precuneus may reflect release of inhibition associated with attentional demands. Thus, alpha EEG characterizes two aspects of processing in the same WM task. EEG alpha distinguishes between cuneal and precuneal activation in working memoryLars In the literature on EEG during working memory (WM), the role of alpha power (8-13 Hz) during WM retention has remained unclear. We recorded EEG while 18 subjects retained sets of consonants in memory for 3 s; setsize (ss4, ss6, ss8) determines memory workload.Theta power (4-8 Hz) increased with workload in all subjects in middle frontal electrodes. Using ICA, the increase in theta could be attributed to one component whose generators were localized by sLORETA in the medial frontal gyrus.Alpha power in parietal electrode Pz showed a mean increase during retention as compared to prestimulus fixation (event-related synchronization, ERS). On an individual basis, alpha power increased with workload in 9 subjects (WL+ group) and decreased in 9 subjects (WL− group). The alpha increased in upper alpha for the WL+ group (mean: 10.4 Hz) and decreased in lower alpha for the WL− group (mean: 8.9 Hz). Time-frequency representations show high alpha power early during retention for the WL+ group and high alpha power late during retention for the WL− group. sLORETA revealed maximal contrast for the WL+ group in the cuneus and for the WL− group in the precuneus.In subjects with WL+, alpha increase in the cuneus may reflect WM maintenance or active inhibition of task-irrelevant areas. In subjects with WL−, alpha decrease in the precuneus may reflect release of inhibition associated with attention...

show abstract

“…A widely-used current spectral estimation method is multitaper spectral analysis (e.g., Percival & Walden, 1993). We make use of a set of K real-valued orthonormal taper sequences {h k,t , t = 0, .…”

Section: Estimation Of the Spectral Matrixmentioning

confidence: 99%

“…We see that without up-weighting, (ζ = 0), the partial coherence is large for frequencies from zero to Nyquist, even though the time series were bandpass filtered to the beta range. This is caused by small amounts of power leaking from high power to low power regions of the spectra, an inevitable part of spectral estimation (e.g., Percival & Walden, 1993), for example giving rise to small non-zero denominator terms in (2), which then inflate the small numerator. Contrariwise, we see that when ζ = 5 · 10 −5 or 10 −4 , the partial coherence is confined to a band of frequencies much closer to the nominal pass-band, and the partial coherence values within the pass-band are virtually unchanged.…”

Section: The Up-weighting Parameter ζmentioning

confidence: 99%

Graphical modelling for brain connectivity via partial coherence

Medkour

Walden

Burgess

2009

Journal of Neuroscience Methods

View full text Add to dashboard Cite

Spectral and coherence methodologies are ubiquitous for the analysis of multiple time series. Partial coherence analysis may be used to try to determine graphical models for brain functional connectivity. The outcome of such an analysis may be considerably influenced by factors such as the degree of spectral smoothing, line and interference removal, matrix inversion stabilization and the suppression of effects caused by side-lobe leakage , the combination of results from different epochs and people, and multiple hypothesis testing. This paper examines each of these steps in turn and provides a possible path which produces relatively 'clean' connectivity plots. In particular we show how spectral matrix diagonal upweighting can simultaneously stabilize spectral matrix inversion and reduce effects caused by side-lobe leakage, and use the stepdown multiple hypothesis test procedure to help formulate an interaction strength.

show abstract

Spectral Analysis for Physical Applications

Cited by 1,810 publications

References 0 publications

Temporal structure in neuronal activity during working memory in macaque parietal cortex

Temporal structure in neuronal activity during working memory in macaque parietal cortex

EEG alpha distinguishes between cuneal and precuneal activation in working memory

Graphical modelling for brain connectivity via partial coherence

Contact Info

Product

Resources

About