Methods for improving the signal-to-noise ratio of endogenous-evoked potentials

A stimulus leads to a cortical response (i.e., evoked potential [EP]) which may be recorded from electrodes attached to the scalp. However, background cortical activity, considered as noise (n), is typically of equal or greater magnitude than the response, which is considered as signal (s). This situation leads to the masking of the presence of the electrocortical signal. Two methods are described which enable the enhancement of the signal with respect to the noise. The first method outlined is time-domain averaging. Its relation to Fourier averaging is also presented. Time averaging can lead to an enhancement of the signal with respect to the noise, known as the signal-to-noise ratio (SNR), by a factor of root N; N being the number of trials recorded. However, latency variability (i.e., jitter) present in the signal leads to a decrement in this maximal potential enhancement. The second technique is an adaptive filter method of averaged cross-correlations, developed by Woody (1967), which deals with the variable latency problem. The development of a latency corrected average developed by McGillem and Aunon (1977) is also presented. The final section describes methods for data handling once the electrocortical signal has been enhanced. It is then necessary to describe the EP quantitatively. Principal Components Analysis (PCA) allows for the quantitative compact representation of the evoked potential waveform. The method also allows for the testing of the effects of explanatory variables on the EP.     

Computer signal detection by monitoring auditory evoked potentials

A digital computer, using a simple decision algorithm, attempted to determine when an acoustical signal had been presented to a cat by monitoring the amplitude of the evoked potentials (EP) at brainstem auditory nuclei. The signal-to-noise level at threshold and the shape and range of the decision model “psychometric functions” were similar to those obtained from humans in the same task. In addition, the detection performance obeys Weber’s law, the mean amplitude of the EP increases monotmonically with signal level, the variance of the amplitude is independent of both signal and noise level, and both the mean latency and the variance of the latency of the peaks decrease with increasing signal level. These findings suggest that the synchronization in firing of a population of single units plays a part in determining the amplitude of the EP. Interaural effects in detection performance were found at the inferior colliculus and, to a lesser extent, at the superior olive.     

The psychological refractory period and vertex evoked potentials

The averaged sensory evoked potential (EP) was recorded from the scalp (vertex to mastoid) in a psychological refractory period experiment in which 12 young adults participated. Reaction times (RTs) were measured to either both or only the second of pairs of stimuli, in different trial blocks, with inter-stimulus intervals (ISIs) of 100, 200, 300 and 400 msec occurring in random sequence. EPs were recorded at each ISI. No latency changes could be found in the prominent non-specific components (P1-N1-P2) of the EP to stimulus 2 even at ISIs where the RT was substantially delayed. Thus the notions that the RT2 delay is due to occupation of a single channel central processor by S1 and that non-specific EP components reflect the time course of information processing in underlying neural tissue, do not lend each other mutual support. Furthermore, as profound amplitude refractoriness in components P1-N1 and N1-P2 persisted at ISIs where RT was as fast or faster than simple RT, there appears to be a dissociation between “psychological refractoriness” and “physiological refractoriness”. The implications of these results are discussed.     

A somatosensory latency between the thalamus and cortex also correlates with level of intelligence

As part of a study on speed of information processing and intelligence, 205 young adult postsecondary students were tested for somatosensory evoked potential (SEP) latencies and intelligence. Following stimulation at the wrist, latencies of three SEPs were determined: N13, generated in the cervical spinal cord/medulla region; N19, generated in the thalamus; P22, generated in the arm region of the somatosensory (parietal) cortex. These latencies and two latency differences, N19 - N13 and P22 - N19, were tested for correlation with a nonverbal measure of intelligence; only P22 - N19 significantly correlated (r = −.217; P = .013, two-tailed). Comparing this latency difference in students in the first IQ quartile (mean IQ = 103.4) with that of students in the fourth quartile (M = 131.2) showed mean differences of 4.13 ms versus 3.21 ms, respectively (p = .0034, two-tailed).P22 - N19 measures time for signal transmission from the thalamus to the sensory cortex. These results agree with considerably more extensive data on visually evoked potentials showing a negative correlation between IQ and the latency for a visual stimulus of the retina to produce a signal at the visual cortex (most of this latency is between the thalamus and the cortex; Reed & Jensen, 1992). The findings here agree with the visual results and strongly suggest that the IQ-latency correlation occurs because the latency indexes cortical nerve conduction velocity, an important component of information processing speed.     

Median method for eliminating infrequent artifacts and identifying the signals blurred by latency jitter and uncertain occurrence

Most researchers have employed the average method in the identification of event-related brain potentials (ERPs) because they have assumed that the averaging process results in the cancellation of any type of noise. However, the average method is less effective than is the median method for unexpected, infrequent artifacts. Furthermore (and significantly), the average method does not work well for detecting the endogenous, psychological signals that are noninvariant across every trial. Ruchkin (1988) described two types of possible signal variation in amplitude and in latency. The median method is effective for extreme cases of these variations (i.e., the lack of some signals), even if the amount of data is small and their distribution is non-Gaussian. Importantly, in the latter situation, the trial-to-trial latency jitter of the signal is very difficult to eliminate, which introduces obscure errors into the average ERP measurements. We found that the waveform that was generated by the median method was less affected by this jittering than was that generated by the average method. This effect was demonstrated by simulating the signals of artificial and actual ERP data, when the distribution of latency variation was Gaussian.     

Evoked potentials in reaction time with a variable foreperiod

The reaction times (RTs) of 12 subjects were recorded in a design where a visual or auditory warning signal preceded an auditory RT signal by one of four short foreperiods 500, 750, 1000 or 1250 ms long, which occured in a random sequence. For the 16 trials at each foreperiod, with each modality of warning signal, the average of the 2-s long EEG samples following the warning signal was computed so that the record showed the scalp recorded (vertex--left mastoid) evoked potentials (EPs) to both warning and RT signals, and also the contingent negative variation or expectancy wave occurring during the foreperiod.

Differences between RTs with different foreperiods were not reflected in negatively correlated differences in the amplitude of the RT signal EPs, taking the major positive going deflection between peaks N1 and P2 at mean latencies of 126 and 231 msec after the RT signal. Furthermore RT signal EPs preceded by a warning signal were highly attenuated in amplitude relative to control EPs which were not preceded by a warning signal, whether or not an RT response was required. This was despite the fact that alerted RTs were slightly faster than non-alerted RTs, so that these findings contradict previous findings associating augmented EPs with responding versus not responding and with speeded RTs.

However, it was also found that RT signal EP amplitudes were greater with the more effective modality of warning signal than the less effective, which was consistent with previous findings. The divergence from previous findings when comparing EPs preceded by a warning with those having no prior warning is tentatively accounted for in terms of persisting physiological refractoriness following the warning signal EP.     

APE: average potential evaluation software for the LAB-8 system

A paper-tape software system is described which allows the addition and/or subtraction of signal averaging data from the DEC advanced averager program. This pooled averaging data may be plotted, printed, or punched. A DEC LAB-8 system with 4K memory and high-speed reader/punch are required. Several examples of the application of the system to human evoked brain response data are presented, and the application of the system to other problems is discussed.     

Difference in Cortical Relay Time Between Intrinsic Muscles of Dominant and Nondominant Hands

The authors aimed to calculate and compare cortical relay time (CRT) between intrinsic hand muscles and between homonymous muscles of dominant and nondominant hands. The participants comprised 22 healthy volunteers. The CRT for long-latency reflexes (LLRs) was calculated by subtracting the peak latency of somatosensory evoked potentials of component N20 and the onset latency of motor evoked potentials from the onset latency of LLRs. CRT was significantly shorter for the first dorsal interosseous muscle than for the abductor pollicis brevis muscle, regardless of hand dominance. CRT for the abductor pollicis brevis muscle was significantly shorter in the dominant hand than in the nondominant hand. Evaluation of CRT for intrinsic muscles might be beneficial in the understanding of individuated finger functions.     

Decision-related cortical potentials during an auditory signal detection task with cued observation intervals.

Cortical-evoked potentials were recorded from human subjects performing an auditory detection task with confidence rating responses. Unlike earlier studies that used similar procedures, the observation interval during which the auditory signal could occur was clearly marked by a visual cue light. By precisely defining the observation interval and, hence, syncrhonizing all perceptual decisions to the evoked potential averaging epoch, it was possible to demonstrate that high-confidence false alarms for accompanied by late-positive P3 components equivalent to those for equally confident hits. Moreover the hit and false alarm evoked potentials were found to covary similarily with variations in confidence rating and to have similar amplitude distributions over the scalp. In a second experiment wherein the signal intensity was increased to make signal presence and signal absence clearly discriminable and the a priori probability of signal presentation was varied from .5 to .9, it was demonstrated that correct rejections can be associated with a P3 component larger than that for hits. Thus it was possible to show, within the signal detection paradigm, how the two major factors of decision confidence and expectancy are reflected in the P3 component of the cortical-evoked potential.     

Visual evoked potentials to the geometric forms in the randomized presentation

Two outlined geometric figures, an equilateral triangle and a circle, of equal contour length, were randomly presented at a fixed position in the lower or upper part of the visual field (LVF and UVF). Transient visual evoked potentials (VEPs) were recorded monopolarly from the inion (I), 5, 10, 15 cm above it (I5, I10, I15) and Fz, for 12 subjects. Grand-averaged VEPs were computed. A negative (N) wave (averaged peak latency 155 ms) was identified in the LVF and a positive (P) wave (130 ms) in the UVF. In the LVF, the N amplitude of the triangle was significantly larger than that of the circle at I5 and I. Regarding the P wave in the UVF, the triangle was of a significantly longer latency than the circle at I15, I10 and I5. The enhancement of the N amplitude for the triangle in the LVF is not attributable to the arousal caused by the preparatory state for the figure, since the subject could not predict the figure to be presented next or its position.     

Visual Signal Detection Measured by Event-Related Potentials

Signal Detection tasks typically involve within-subject signal changes. Such a procedure does not lend itself to event-related potential (ERP) experiments where the need for averaging necessitates the maintenance of consistent stimulus parameters. In the present ERP study we adopt a novel approach to thresholding that allows within-subject signal manipulation. The Signal Detection task required the identification of letter targets, formed from dots, in a random dot field. ERP waveforms were segmented into three windows corresponding to N1, N2, and P300 components. Analysis shows that ERP variations are dependent on both task demands and response characteristics for N1, N2, and P300 components.     

Effects of angularity of the figures with sharp and round corners on visual evoked potentials

Following a study in which equilateral triangles elicited larger visual evoked potentials (VEPs) than either squares or circles, we examined the effect of single-line angular figures with a sharp or a round corner at angles of 45°, 90°, 135° or 180°. VEPs were recorded monopolarly at four locations on the midline of scalp for 10 subjects, while the figure was tachiscopically presented to the lower visual field. Subtracted waves were obtained between figure and control (blank) conditions. N1 (peak latency 135–142 ms) and P2 (235–237 ms) waves were identifiable. N1 amplitude tended to decrease as a function of angularity, irrespective of orientation (angle pointing up or down). The effect of the sharpness/roundness of the corner was much smaller than that of its angularity. These findings suggest that the greater VEP response with triangles than with squares and circles may be attributable to the acute angularity of triangles. Related cortical processes and VEP components are also discussed.     

Different Anaphoric Expressions Are Investigated by Event-Related Brain Potentials

Event-related potentials were recorded to substantiate the claim of a distinct psycholinguistic status of (a) pronouns vs. proper names and (b) ellipses vs. proper names. In two studies 41 students read sentences in which the number of intervening words between the anaphor and its antecedent was either small or large. Comparing the far with the near distance condition revealed anaphor resolution specific effects: Ellipses triggered a potential shift with a short latency (approximately 120-200 ms) and with a fronto-central scalp distribution while pronouns and proper names triggered one with a longer latency (approximately 360-440 ms) and a parietal to right-occipital distribution. The early effect resembled the left-anterior negativity which has been related to syntax processing, while the latter resembled an N400 which is assumed to reflect semantic integration processes. These findings support the idea that ellipses and pronouns/proper names are processed by distinct mechanisms being implemented in distinct cortical cell assemblies.     

Sequence-related changes in sensory-evoked potentials in the dentate gyrus: a mechanism for item-specific short-term information storage in the hippocampus

Sensory-evoked potentials were recorded from the dentate gyrus of the rat hippocampus during performance of a differential auditory discrimination task. The short latency (20 ms) component (N1) of the sensory-evoked potential showed systematic amplitude fluctuations dependent upon the sequence of positive and negative trials preceding the presentation of a given trial and did not depend on the associated reward values of the individual tone stimuli which evoked the potential. The amplitude fluctuations could be accurately depicted by a model which retained the sequence for the five preceding trials in a "buffer" with exponentially decaying influence as a function of time of trial occurrence within the sequence. The results provide evidence that the hippocampus encodes accurate short-lasting representations of sensory events which can provide the basis for storage of information pertaining to past experiences.     

Event-related brain potentials dissociate repetition effects of high-and low-frequency words

Event-related brain potentials (ERPs) were recorded while subjects detected nonwords interspersed among sequences of words of high or low frequency of occurrence. In Phase 1, a proportion of the words were repeated after six intervening items. In Phase 2, which followed after a break of approximately 15 min, the words were either repeats of items presented in the previous phase or new. Unrepeated low-frequency words evoked larger N400 components than did high-frequency items. In Phase 1, this effect interacted with repetition, such that no frequency effects were observed on N400s evoked by repeated words. In addition, the post-500-msec latency region of the ERPs exhibited a substantial repetition effect for low-frequency words, but did not differentiate unrepeated and repeated high-frequency words. In Phase 2, ERPs evoked by "old" and "new" high-frequency words did not differ in any latency region, while those evoked by old and new low-frequency words differed only after 500 msec. The interactive effects of frequency and repetition suggest that these variables act jointly at multiple loci during the processing of a word. The specificity of the post-500-msec repetition effect for low-frequency words may reflect a process responsive to a discrepancy between words' intra and extraexperimental familiarity.     

Long latency auditory evoked potentials

Experiment 1 elicited the P1, N1, P2, and N2 components of the long latency auditory evoked potential (AEP) using a 1000 Hz tone presented at 30, 50, or 70 dB SPL and 1-, 3-, or 5- second inter-stimulus intervals to assess the relative effects of the combination of these variables on component amplitude and latency. Four blocks of 16 tone presentations each were recorded from each subject to determine if changes in the AEP would occur because of short-term habituation. Both stimulus factors interacted significantly in a systematic fashion for the amplitude measures, with increases in latency also associated with increases in intensity and inter-stimulus interval. Only minor changes across the four trial blocks for either the amplitude or latency measures were observed over the various stimulus presentation conditions. Experiment 2 employed the same tone stimulus presented at 50 dB SPL and a 3-second inter-stimulus interval. Eight blocks of 64 trials were recorded from each subject on each day for four days to investigate long-term habituation effects. No substantial changes in any of the component amplitudes or latencies were obtained across the 32 trial blocks. It was concluded that intensity and inter-stimulus interval interact to determine AEP amplitude as well as latency values and that the constituent components do not change appreciably with repeated stimulus presentations, even after several days.     

Electrophysiologic Manifestations of Impaired Temporal Lobe Auditory Processing in Verbal Auditory Agnosia

The present study examined the extent to which verbal auditory agnosia (VAA) is primarily a phonemic decoding disorder, as contrasted to a more global defect in acoustic processing. Subjects were six young adults who presented with VAA in childhood and who, at the time of testing, showed varying degrees of residual auditory discrimination impairment. They were compared to a group of young adults with normal language development matched for age and gender. Cortical event-related potentials (ERPs) were recorded to tones and to consonant-vowel stimuli presented in an "oddball" discrimination paradigm. In addition to cortical ERPs, auditory brainstem responses (ABRs) and middle latency responses (MLRs) were recorded. Cognitive and language assessments were obtained for the VAA subjects. ABRs and MLRs were normal. In comparison with the control group, the cortical ERPs of the VAA subjects showed a delay in the N1 component recorded over lateral temporal cortex both to tones and to speech sounds, despite an N1 of normal latency overlying the frontocentral region of the scalp. These electrophysiologic findings indicate a slowing of processing of both speech and nonspeech auditory stimuli and suggest that the locus of this abnormality is within the secondary auditory cortex in the lateral surface of the temporal lobes.     

A wavelet packet model of evoked potentials

The standard methods for decomposition and analysis of evoked potentials are bandpass filtering, identification of peak amplitudes and latencies, and principal component analysis (PCA). We discuss the limitations of these and other approaches and introduce wavelet packet analysis. Then we propose the "single-channel wavelet packet model," a new approach in which a unique decomposition is achieved using prior time-frequency information and differences in the responses of the components to changes in experimental conditions. Orthogonal sets of wavelet packets allow a parsimonious time-frequency representation of the components. The method allows energy in some wavelet packets to be shared among two or more components, so the components are not necessarily orthogonal. The single-channel wavelet packet model and PCA both require constraints to achieve a unique decomposition. In PCA, however, the constraints are defined by mathematical convenience and may be unrealistic. In the single-channel wavelet packet model, the constraints are based on prior scientific knowledge. We give an application of the method to auditory evoked potentials recorded from cats. The good frequency resolution of wavelet packets allows us to separate superimposed components in these data. Our present approach yields estimates of component waveforms and the effects of experiment conditions on the amplitude of the components. We discuss future extensions that will provide confidence intervals and p values, allow for latency changes, and represent multichannel data.     

Symmetric distribution in latencies of cortical somatosensory potentials evoked by right and left posterior tibial nerve stimulation in right-, left-, and mixed-handed men and women

The latencies of the P1, N1, P2, and N2 components of the somatosensory potentials evoked by stimulation of the right and left posterior tibial nerves in the right-, left-, and mixed-handed men and women were subjected to a statistical analysis. The mean latencies of the primary and secondary cortical responses are symmetrically distributed between the right and left sides for all the subjects. The mean latencies obtained of women were significantly shorter than those of men. This difference was accounted for by the fact that the mean body height of women was significantly less than that of men, since there was a significant positive correlation between body height and latency. It was concluded that the conduction times for the cortical input as well as the early cortical information processing, despite synaptic interventions, cannot play a role in the cerebral lateralization concerning the perceptual and manipulospatial tasks.