Auditory evoked potentials as probes of hemispheric differences in cognitive processing

A novel approach to validate the efficacy of single task ERP paradigms to measure cognitive workload

The present study examined the utility of a single-task paradigm to evaluate cognitive workload. The cognitive workload from twenty-five healthy participants was measured during a tilt-ball game while tones were presented in the background to generate event-related potentials (ERPs) in electroencephalographic (EEG) data. In the game, participants were instructed to move the ball to highlighted targets and avoid moving obstacles. The game's difficulty level was manipulated (easy, medium, hard) by adjusting the number and speed of the moving obstacles. The difficulty levels were presented in a random order during multiple short runs to minimize the effects of habituation, fatigue, and boredom. The behavioral results showed that greater task difficulty resulted in a significant decrease (p < 0.001) in game performance, i.e., participants achieved few targets with a high collision rate. To evaluate cognitive workload, we measured the amplitude of early ERP components (N1, P1, and P2) corresponding to the involuntary attention orienting response. The amplitude of the N1 component decreased significantly (p = 0.029) with an increase in cognitive workload. These findings suggest that the early ERP component, specifically the N1, corresponds to attention orienting response, and that the task difficulty modulates it. This study provided evidence that the inverse relationship between ERP components and cognitive workload can be reliably assessed by controlling for other factors such as habituation or boredom during a single task paradigm.

Neural Indices of Vowel Discrimination in Monolingual and Bilingual Infants and Children

OBJECTIVES

To examine maturation of neural discriminative responses to an English vowel contrast from infancy to 4 years of age and to determine how biological factors (age and sex) and an experiential factor (amount of Spanish versus English input) modulate neural discrimination of speech.

DESIGN

Event-related potential (ERP) mismatch responses (MMRs) were used as indices of discrimination of the American English vowels [ε] versus [I] in infants and children between 3 months and 47 months of age. A total of 168 longitudinal and cross-sectional data sets were collected from 98 children (Bilingual Spanish-English: 47 male and 31 female sessions; Monolingual English: 48 male and 42 female sessions). Language exposure and other language measures were collected. ERP responses were examined in an early time window (160 to 360 msec, early MMR [eMMR]) and late time window (400 to 600 msec, late MMR).

RESULTS

The eMMR became more negative with increasing age. Language experience and sex also influenced the amplitude of the eMMR. Specifically, bilingual children, especially bilingual females, showed more negative eMMR compared with monolingual children and with males. However, the subset of bilingual children with more exposure to English than Spanish compared with those with more exposure to Spanish than English (as reported by caretakers) showed similar amplitude of the eMMR to their monolingual peers. Age was the only factor that influenced the amplitude of the late MMR. More negative late MMR was observed in older children with no difference found between bilingual and monolingual groups.

CONCLUSIONS

Consistent with previous studies, our findings revealed that biological factors (age and sex) and language experience modulated the amplitude of the eMMR in young children. The early negative MMR is likely to be the mismatch negativity found in older children and adults. In contrast, the late MMR amplitude was influenced only by age and may be equivalent to the Nc in infants and to the late negativity observed in some auditory passive oddball designs.

Socio-emotionally significant experience and children's processing of irrelevant auditory stimuli

Theory and research indicate considerable influence of socio-emotionally significant experiences on children's functioning and adaptation. In the current study, we examined neurophysiological correlates of children's allocation of information processing resources to socio-emotionally significant events, specifically, simulated marital interactions. We presented 9- to 11-year-old children (n=24; 11 females) with 15 videos of interactions between two actors posing as a married couple. Task-irrelevant brief auditory probes were presented during the videos, and event-related potentials (ERPs) elicited to the auditory probes were measured. As hypothesized, exposure to higher levels of interparental conflict was associated with smaller P1, P2, and N2 ERPs to the probes. This finding is consistent with the idea that children who had been exposed to more interparental conflict attended more to the videos and diverted fewer cognitive resources to processing the probes, thereby producing smaller ERPs to the probes. In addition, smaller N2s were associated with more child behavior problems, suggesting that allocating fewer processing resources to the probes was associated with more problem behavior. Results are discussed in terms of implications of socio-emotionally significant experiences for children's processing of interpersonal interactions.

Sex-related differences in event-related potentials, face recognition, and facial affect processing in prepubertal children

Thirty-five prepubertal children, 17 boys and 18 girls, between the ages of 8 and 11 years, were studied to examine electrophysiological and cognitive sex differences during a face-recognition-memory (FRM) task and a facial-affect-identification task (FAIT). All participants were prepubertal, as determined by J. M. Tanner's (1962) staging and endocrine evaluation. Sex-dependent event-related potential (ERP) amplitude asymmetries were found during FRM. Boys displayed greater right versus left ERP amplitude to auditory tone probes during the task, whereas girls displayed the opposite pattern. In addition, positive correlations were obtained between ERP amplitude during FRM and FAIT accuracy scores for boys, but not for girls. Results suggest that girls and boys may use different neuronal systems in the processing of faces and facial affect. Findings are consistent with developmental theories regarding sex differences in visuospatial processing.

An electrophysiological study of infants' sensitivity to the sound patterns of English speech

The study explores 10- to 11-month-old infants' sensitivity to the phonological characteristics of their native language. Event-related potentials (ERPs) were obtained for tones that were superimposed on two versions of a story: an Unmodified version containing normal English function morphemes, and a Modified version in which the prosodic and segmental properties of a subset of function morphemes were changed to make them atypical. The 11-month-olds exhibited significantly lower amplitude ERPs to the tones during the Modified story than to the Unmodified story, whereas the 10-month-olds showed no differences. These results suggest that the 11-month-olds discriminated the two versions of the story based on their representations of the phonological properties of English. Further, the tone-probe ERP method can successfully be used to study the development of speech perception in the pre-linguistic infant.

Sex differences in visual recognition memory: support for a sex-related difference in attention in adults and children

The selectivity hypothesis of Meyers-Levy (1989) proposes that cognitive sex differences reflect underlying differences in information processing between males and females. Males are considered to be more likely to organize information in a self-related manner, whereas females are more likely to adopt a comprehensive approach to information processing. We tested this hypothesis in children (10-15 years) and adults using recognition memory tasks. Tests were devised which employed male-oriented objects, female oriented objects, or random objects. In both the child and adult samples, females performed significantly better than males on tests using random and female-oriented objects. Males performed at the level of females only when tested for recognition of male-oriented objects. These results demonstrate that this sex difference is present prior to puberty and support the concept of sex differences in information processing.

CNV, PINV and probe-evoked potentials in schizophrenics

Distractibility and temporal modulation of attention in schizophrenics were studied using a visual reaction time task with additional auditory probe stimuli during the forewarning period or between trials. The probes were thought to exert a distracting influence, especially on schizophrenics, and at the same time they generated auditory EPs which allowed to track the modulation of cortical excitability during response preparation. The midline distribution of the terminal contingent negative variation (tCNV) and the amplitude of the postimperative negative variation (PINV) were clearly different in 20 DSM III-R schizophrenics, as compared with 20 alcoholics and 20 normal controls. In schizophrenics, the more frontal distribution of the tCNV was associated with a higher degree of psychopathology (measured with the Brief Psychiatric Rating Scale) and with delayed reactions. Probes between trials reduced tCNV and PINV in all subjects alike. However, this effect could not be attributed to distraction, because reaction times were faster in these trials, possibly due to an alerting effect of the auditory probes. The N100 and P300 amplitudes to probes in the forewarning period, i.e., during the negative potential shift of the CNV, were significantly enhanced in all groups. Apparently there is a state of increased cortical excitability during the CNV which is not selectively "tuned" toward relevant stimuli. In schizophrenia, the temporal and topographical regulation of this excitability is disturbed.

Fast habituation of the long-latency auditory evoked potential in the awake albino rat

Fast habituation of the long-latency, vertex-recorded auditory evoked potential (AEP) peaks in humans was first described by Callaway (1973) as a reduction in AEP amplitude that occurs to the second of a pair of acoustic stimuli when both stimuli are presented with an interstimulus interval (ISI) of no more than 10 sec. When acoustic stimuli are presented in pairs with an ISI of 2 sec and an interpair interval (IPI) of approximately 10 sec, reduction in amplitude to the second tone occurs by as much as 30-50%. Fast habituation may depend somewhat on a subject's anticipation of the stimulus and on other factors related to attention and orienting. Studies in our laboratory have demonstrated this amplitude decrement to the second tone of a pair in human infants, children and adults and have explored the implications of this finding with respect to attentional processes and the allocation of cerebral resources. In the present investigation we describe an animal model of fast habituation. Here, vertex-recorded AEPs were obtained to paired tone stimuli delivered to awake adult male Sprague-Dawley rats chronically implanted with skull electrodes. Findings showed: (a) an AEP wave form with 8 distinct peaks, (b) for one component there was a marked decrement in amplitude from tone 1 to tone 2 in recordings obtained from an electrode placed slightly to the right of midline, and (c) that there were no significant differences in peak latencies across tones. This methodology may further our understanding of fast habituation in humans and may prove useful for studies of attention, orienting, and resource allocation using techniques that are not possible for use with human subjects.

Screening of dominantly inherited Charcot-Marie-Tooth neuropathies

Sixty-three families with dominantly inherited Charcot-Marie-Tooth (CMT) neuropathies including 730 subjects (total) from which 356 affected were studied clinically, electrophysiologically (MNCVs and EMGs), by genetic linkage, and screened for DNA duplication. Thirty-eight families (60.3%) were type 1A (demyelinating CMT mapped on chromosome 17). DNA duplication was present in 36 families (94.8% of CMT1A families). One CMT1A family (2.6%) showed no duplication but suggested genetic linkage with markers of chromosome 17. One CMT1A family (2.6%) revealed nonduplication in some affected members and duplication in other affected members. The disease in that family segregated with the same chromosome 17 markers regardless of duplication status. The other CMT families with dominant inheritance but without duplication included one family with CMT1B (demyelinating CMT mapped on chromosome 1) (1.6%), 14 families with CMT2 axonal neuropathy (22.2%), and 10 families with X-linked dominant CMT (15.9%).

Effects of stimulation intensity, gender and handedness upon auditory evoked potentials

Left handers and women show less anatomical brain asymmetry, larger corpus callosum and more bilateral representation of specific functions. Sensory and cognitive components of cortical auditory evoked potentials (AEP) have been shown to be asymmetric in right handed males and to be influenced by stimulus intensity. In this study the influence of sex, handedness and stimulus intensity upon AEP components is investigated under basal conditions of passive attention. 14 right handed males, 14 right handed females, 14 left handed males, and 14 left handed females were studied while lying awake and paying passive attention to auditory stimulation (series of 100 binaural clicks, duration 1 msec, rate 1/sec, at four intensities). Cz, C3 and C4 referenced to linked mastoids and right EOG were recorded. Analysis time was 400 msec, average evoked potentials were based on 100 clicks. Stimulus intensity and gender affect early sensory components (P1N1 and N1P2) at central leads, asymmetry is influenced only by handedness, right handers showing larger P1N1 amplitudes over the right hemisphere.

Right-sided facial asymmetry in infant emotional expression

In normal adults, emotional expressions tend to be more intense on the left side of the face, while in preschool-age children, no significant bias in facial asymmetry has been found. We examined facial asymmetries during smiling and distress in 59 infants studied longitudinally at 6.5, 10 and 13.5 months of age. In these infants, asymmetric expressions showed a bias toward greater intensity on the right side of the face. The study of infant facial expression may provide clues to the maturation of the cortical control of emotional responses.

Hemispheric asymmetries in phonological processing assessed with probe evoked magnetic fields

Auditory Evoked Magnetic Fields (EFs) to tonal stimuli were recorded at homotopic maxima over the left and right auditory areas in nine subjects. Recordings were made during two conditions, both involving simultaneous presentation of the probe tone stimuli and a set of tape-recorded verbal material. During the control condition subjects were instructed to attend to the tones and ignore the verbal material. In the phonological processing condition they were instructed to ignore the tones and attempt to identify a phonological target item which was embedded in the verbal material. EFs obtained during both conditions were characterized by an early N1m and a later P2m component corresponding to the N1 and P2 components of auditory evoked potentials (EPs). During the phonological condition, the amplitude of the N1m was significantly reduced in both hemispheres symmetrically whereas the amplitude of the P2m was attenuated to a significantly greater degree in the left hemisphere. These data are in agreement with previous EP evidence of greater interference of linguistic processing with processing of irrelevant probe stimuli in the left hemisphere, indicative of greater left hemisphere involvement in language tasks.

Event-related potentials as asymmetry indices of lateralized cognitive processes during music and verbal tasks

The present study replicates and extends earlier findings by using event-related potentials (ERPs) as a means of providing information about differential hemispheric engagement in cognitive processing. ERPs were recorded from T3, T4, T5 and T6 in response to task-irrelevant tone pips superimposed on white noise, on music and on verbal passages, which are thought to differentially activate the two hemispheres. The latencies and amplitudes of three peaks (N1, P2 and N2) were obtained in 37 women. Task-related ERP asymmetries were found only for the peak amplitude of N2 recorded at posterior-temporal scalp areas, whereas at the anterior-temporal recording, there were no interactional effects. The hemisphere assumed to be most engaged in task processing showed a smaller N2 peak amplitude to tone pips at the posterior-temporal scalp area than the hypothesized unengaged hemisphere. A result common to the three peaks in the ERP was that peak latencies were longer in the music than in the noise condition. Finally, the posterior-temporal areas produced longer N2 peak latencies in verbal and even more in music conditions compared to white noise conditions. No differences in latencies were found between hemispheres during the ongoing cognitive tasks.

Reading comprehension performance and laterality: evidence for concurrent validity of dichotic, dichhaptic and EEG laterality measures

EEGs were recorded while 55 boys read and answered questions about passages. Concurrent validity between parietal alpha ratios obtained from these recordings and dichotic/dichhaptic scores supports their utility as measures of lateral processing. The evidence was particularly strong between right hand advantage for dichhaptically presented sequential patterns and left hemispheric activity while answering. Results also indicate that coefficients for dichotically presented digits and alpha ratios during silent reading are indices of receptive laterality and measure the deployment of lateral structures rather than degree of structural laterality. This supports a dynamic rather than static model of lateral processing during receptive language.

A method for chronically recording brain-stem and cortical auditory evoked potentials from unanesthetized mice

Anesthetizing or sedating animals affords control over movement artifact during electrophysiological recording. However, the use of chemical restraint leads to unwanted side effects such as drug-induced hypothermia. Hypothermia is problematic because BAEP amplitudes and latencies are affected significantly by core temperature changes. Moreover, several recent studies have indicated that anesthetics and sedatives may significantly alter the BAEP and CAEP even in thermoregulated subjects. There is a need, therefore, to develop a practical technique for obtaining chronic BAEPs and CAEPs in restrained, unanesthetized animals. The present report describes a technique that permits the gathering of consistent, reliable evoked potential recordings over time in unanesthetized mice. The preparation is useful for studying the CNS effects of pharmacological agents, pathological conditions, aging and development in the mouse and could be adapted for use with other small animals as well.

Probe evoked potentials: theory, method and applications

The probe evoked potentials method which involves recording brain responses to irrelevant stimuli during performance of a wide range of cognitive tasks, is used with increasing frequency to assess patterns of regional cerebral activation mediating distinct cognitive operations. This paper delineates the evolution of the method, describes its procedural and theoretical features, reviews its applications with normal adults, infants and children, and with neuropsychologically compromised individuals. The outcome of these studies points to the advantages of this method over traditional evoked potential procedures in the assessment of regional cerebral engagement in cognitive operations, free of confoundings due to stimulus and response-specific activity, in ecologically valid experimental situations.

Regional brain activity in dyslexic and control children during reading tasks: visual probe event-related potentials

Event-related potentials (ERPs) elicited by task-irrelevant visual stimuli were recorded from 34 control and 32 dyslexic 10- to 12-year-old boys while they performed silent and oral reading at two levels of difficulty. All subjects were extensively screened for neurological problems, IQ, and sensory acuity. Specific features of the ERPs were affected in amplitude and latency by the experimental variables: group membership, oral vs. silent reading, difficulty level, and recording site (within and between hemispheres). A specific effect of difficulty on the central and parietal ERP was seen in the dyslexics but not the controls. Different patterns of asymmetry were found for the two groups in silent vs. oral reading at midtemporal placements. A marked asymmetry (R greater than L) was found at the midtemporal region for both groups.

Event-related brain potentials differentiate normal and disabled readers

Auditory event-related potentials (AERPs) recorded to irrelevant tone pairs while subjects performed visual, reading-related cognitive tasks differed significantly between normal and disabled readers. Disabled readers as compared with normal readers showed significantly lower amplitude right hemisphere AERP responses during tasks that involved visual-phonemic transfer of information and simple pattern recognition. Disabled readers as compared with normal readers also showed significantly higher amplitude left hemisphere responses during the visual-phonemic task. In both experimental conditions the reading-disabled subjects showed significantly lower amplitude right than left hemisphere AERP responses. Task-related strategies did not differ between groups. The pattern of AERP amplitude asymmetry found for disabled readers, which was opposite to that found for normal readers, suggests that the same reading-related tasks activated different cerebral processes in the two groups studied.

The use of a control or baseline condition in electrophysiological studies of hemispheric specialization of function

Asymmetries in event-related potential (ERP) measures between the two cerebral hemispheres in humans may not be apparent when the absolute values of left and right hemisphere recordings are only considered within the context of one experimental condition. ERP asymmetries may only emerge when results obtained during experimental conditions designed to differentially activate the two cerebral hemispheres are compared to a baseline or control condition involving low-level cognitive demands. Both theoretical and empirical considerations of such a baseline condition are discussed.

Changes in patterns of hemispheric electrophysiological activity as a function of instructional set

Investigations of differential hemispheric functioning which have used auditory event-related potentials (AERPs) typically employ paradigms in which two or more conditions are used. In these paradigms, conditions typically differ from each other with respect to both the physical aspects of the stimuli and the operations the subject must perform on these stimuli. Because both the subject's task and the physical properties of the stimuli change from one condition to another, it is unclear whether different patterns of AERP cerebral asymmetries obtained between conditions are primarily a function of differences in the stimuli (e.g., verbal vs. musical stimuli) or of endogenous factors such as the subject's cognitive approach to the processing of the stimuli. In the present study, two conditions with identical stimuli were employed but different instructions were given about how to process these stimuli. Music and paired tones were simultaneously presented in both conditions. In the Music Condition, subjects performed a task which involved attending to the music and ignoring the tones. In the Tones Sequence Condition, subjects counted specified sequences of tones and ignored the music. The results indicated that there were different patterns of left-right hemispheric AERP relationships between the two conditions. These findings support the hypothesis that endogeneous factors play a significant role in differential hemispheric functioning and that the AERP technique employed in this paradigm is a sensitive index of changes in electrophysiological activity associated with these endogeneous factors.

The lateral distribution of event-related potentials during sentence processing

Event-related brain potentials (ERPs) were recorded from adults as they read 160 different sentences, half of which ended with a semantically anomalous word. These deviant words elicited a broad, negative component (N400). Measured in the difference wave between ERPs to incongruous and congruous endings, the N400 was slightly larger and more prolonged over the right than the left hemisphere and diminished in amplitude over the course of the experiment. A left-greater-than-right asymmetry was again observed in the slow, positive ERP elicited by the first six words in the sentences, being most pronounced for subjects having no left-handers in their immediate family.

Evoked potentials to auditory probes as indices of cerebral specialization of function--replication and extension

This study was an attempt to replicate findings from previous work in our laboratory as well as to study the relationship between vertex-referenced and linked mastoid-referenced responses under conditions purported to produce differential hemispheric activation. Auditory evoked potentials were recorded between T4-Cz, T3-Cz, T4-A1A2, and T3-A1A2 to task-irrelevant tone pips superimposed on a baseline condition and on two experimental conditions (verbal and music) hypothesized to differentially activate areas of the left and right cerebral hemispheres. Significant AEP amplitude asymmetries were obtained as a function of the ongoing tasks, replicating our previous findings. Higher amplitude AEPs were recorded from temporal-to-vertex leads from the hemisphere thought to be most involved in the ongoing task. In addition, temporal-linked mastoids placements showed an effect opposite to that seen for the vertex-referenced leads with lower amplitude AEPs occurring from the hemisphere purported to be most activated by the task. Mastoid-referenced placements were also less sensitive to task effects than vertex-referenced placements.