EEG, Evoked Potential, and Extracellular Single-Unit Recordings In Vivo - Animal Models of Acute Neurological Injuries: Injury and Mechanistic Assessments - page 96

4. Data Analysis

1. EEG: The EEG is typically described in terms of (1) rhythmic
activity and (2) transients. The rhythmic activity is divided
into bands by frequency. For example, any rhythmic activity
between 4 and 7 Hz can be described as “theta,” 8 and 12 Hz
as “alpha,” 12 and 30 Hz as “beta,” and 30 and 100 Hz as
“gamma.” A given rhythmical activity within the EEG is
thought to refl ect a particular distribution of brain networks
or functional conditions and exhibits a specifi c biological
signifi cance. Frequency bands are usually extracted using spec-
tral methods (for instance, Welch) that are integrated into EEG
software, such as EEGLAB.
2. EP: Two main features of an EP response are the amplitude
and the latency of the peaks. The most pronounced peaks have
been studied and given designations in various laboratories.
Each peak is given a letter and a number in its name. For example,
N20 refers to a negative peak (N) at 20 ms. This peak is
recorded from the cortex by stimulating median nerve. It most
likely corresponds to the signal reaching the somatosensory
cortex. When used in intraoperative monitoring, the latency
and amplitude of the peak relative to the patient's postintuba-
tion baseline is a crucial piece of information. Prominent
increases in latency or decreases in amplitude are the common
indicators of neurological dysfunction.
3. Extracellular single-unit recording: For single-neuron recordings,
neuronal encoded signals are presented as fi ring frequencies
and patterns. Recording software, such as pClampfi t 9 (Axon
Instrument) or Mini Analysis 6 (Synaptosoft, Leonia, NJ), can
be used to quantify these parameters. Bursting was identifi ed
according to the “80/160 ms” defi nition proposed by Grace
and Bunney ( 20, 21 ). Thus, the onset of a burst was identifi ed
as the concurrence of two spikes with an ISI of less than 80 ms
and the termination of a burst was defi ned as an ISI greater
than 160 ms ( 20, 21 ). Firing periodicity is presented using fi ring
rate histograms (bandwidth: 50 ms). The Fast Fourier Transform
(FFT) is usually applied on the rate histograms yielding a spec-
trum with a resolution of 0.078 Hz. Spectra obtained from
rate histograms are qualitatively identical.

5. Advantages
and Pitfalls

5.1. Electro-
encephalography

The EEG has several positive features as a tool for exploring brain
activity. Firstly, an EEG exhibits excellent temporal resolution (can
detect changes within a millisecond time frame). Secondly, it can