Spectral analysis of event related potentials

Event-related slow-wave activity in two subtypes of attention-deficit/hyperactivity disorder

OBJECTIVE

Previous time-frequency studies have indicated that event-related low-frequency activity has important effects on component topography and developmental effects in auditory event-related potentials (ERPs) of children and adolescents. This study investigated the influence of event-related slow-wave (SW) (0.01-2 Hz) activity in the group differences seen between children with and without attention-deficit/hyperactivity disorder (AD/HD) of different subtypes.

METHODS

Time-frequency analysis techniques were applied to auditory ERP data derived from children with AD/HD predominantly inattentive type (n = 30), AD/HD combined type (n = 30) and age-matched control subjects (n = 30).

RESULTS

Event-related early frontal negative and late posterior positive SW components were reduced in the AD/HD combined type group, but not AD/HD inattentive type group, relative to controls. The RESIDUAL ERPs, which represented event-related 2-12 Hz activity, showed clinical vs. control group differences in components that were similar in both AD/HD subtype groups.

CONCLUSIONS

The time-frequency results showed that event-related SW (0.1-2 Hz) activity contributes importantly to group differences between AD/HD and control children, and the pattern of group differences from controls for each of the AD/HD subtype groups, which are evident in raw ERPs. These results emphasise both the clinical and developmental importance of this form of analysis.

SIGNIFICANCE

This novel approach revealed additional specific information about stimulus processing and regional inhibition/activation in two AD/HD subtypes, relative to control subjects.

An investigation of the event-related slow-wave potential (0.01-2 HZ) in normal children

Previous research has indicated that the auditory ERPs of children are dominated by a frontal negative slow-wave. This paper outlines a preliminary attempt to separate event-related low-frequency activity from traditional ERP components as found in older subjects. An active auditory oddball task was completed by 30 normal children aged from 8 to 18 years, and ERPs to target and standard stimuli were derived. The original ERP files were digitally filtered to calculate separate ERPs containing only 0.01-2 Hz activity (termed the SW-ERP) or 2-12 Hz activity (termed the RESIDUAL ERP) for each subject. The SW-ERP was maximally correlated with the slow-wave factor from a principal components analysis of the original target and standard waveforms. The SW-ERPs to target stimuli contained an early negative component which showed an age-related decrease, and a late positive component which did not. The SW-ERP was found to be partly responsible for the differences in component amplitude and morphology between sites observed in the raw ERPs.

Cadmium induced alterations in somatosensory evoked potentials

Pregnant Swiss albino rats were divided into three groups: control (C), cadmium (Cd) and non-cadmium (NCd) groups. Control animals received tap water; the Cd rats received Cd as CdCl2 in their drinking water during the experimental period, while the NCd group was given Cd during pregnancy, and given tap water after birth. Twenty-two days after birth, 15 rats (for each group) were taken from their mothers and continued to be treated with Cd (Cd group) or tap water (C and NCd groups) for an additional 38 days. After the treatment period, somatosensory evoked potentials (SEPs) of the three groups were recorded from central (Cz) referenced to frontal (Fz) following left posterior tibial nerve (PTN) stimulation. Amplitude spectra of SEPs were computed by fast Fourier transform (FFT) algorithm. There was a significant amplitude decrease in 1-3.5 Hz in the NCd group and 1-3.5 and 14-20 Hz frequency bands of the Cd group compared with the control group.

Event-related potentials in hypertriglyceridemia

Event-related potentials (ERPs) of twenty-three hypertriglyceridemic (HTG) patients and twenty-three age-matched healthy controls were recorded in two different experimental conditions that the target stimulus was counted (Test 1) or uncounted (Test 2). Latencies of each wave were inside normal limits in all patients. No differences were found in the P3b amplitudes of Test 1 case among HTG and control subjects. Amplitude spectra of ERPs were computed by transient response-frequency characteristics (TRFC) method. Their maxima were found to occupy the frequency bands of 1-2, 3-4, 5-7, 8-12, 13-20, 20.5-32 Hz. When comparing Test 1 with Test 2, significant amplitude differences were found in 1-2 and 3-4 frequency bands of controls and 1-2 and 13-20 frequency bands of patient group.

The effect of aging on spectral parameters of event-related potentials

Forty-two healthy subjects ranging in age from 20 to 73 were divided into three groups according to age; a young group (20-33 years), a middle-aged group (34-49 years) and older group (50-73 years). Event-related potentials (ERPs) of three groups were recorded in two different experimental conditions that the infrequent stimulus was counted (Test 1) or uncounted (Test 2). ERPs were elicited using infrequent and frequent stimuli as red and green lights respectively. Spectral analysis of ERPs showed that decibel (dB) values of 1-2 and 3-4 Hz in young and middle-aged groups while dB value of 1-2 Hz in older group were significantly decreased in Test 2 compared with Test 1. When the number of subjects displaying amplitude maximum in each frequency band was considered, significant differences were found in 1-2 and 5-7 Hz frequency bands of young and middle-aged groups, but no significant differences were found for older group.

ERP spectral analysis of cholesterol rich patients

Visual event-related potentials (ERPs) were studied in twenty hypercholesterolemic (HC) patients and twenty age-matched healthy controls. ERPs were recorded in two different experimental conditions that the target stimulus (red light) was counted (Test 1) or uncounted (Test 2). Amplitude spectra of ERPs were computed by transient response-frequency characteristics (TRFC) method. Their maxima were found to occupy the frequency bands of 1-2, 3-4, 5-7, 8-12, 13-20, 20.5-32 Hz. The amplitude mean (decibel) of 1-2 and 3-4 frequency bands were decreased in Test 2 compared to Test 1 in the control group, but no significant amplitude differences were found between Test 1 and Test 2 in the HC group.