Can contrast sensitivity functions in dyslexia be explained by inattention rather than a magnocellular deficit?

We examined whether data demonstrating contrast sensitivity losses in dyslexia that have been interpreted as evidence for loss of magnocellular visual function could be explained by inattention. Computer simulations of observers with poor concentration yielded inflated estimates of threshold that were a constant proportion of the true threshold across spatial frequencies. Data from many, but not all, studies supporting the magnocellular deficit theory are well described by these simulations, which predicted no interaction between observer group and spatial frequency. Some studies have reported significant interactions, but suffer from statistical deficiencies. This compromises some of the evidence for a magnocellular deficit in dyslexia derived from studies of threshold contrast sensitivity.

Effect of normobaric hypoxia on sound localization.

BACKGROUND: Three-dimensional (3D) audio displays have considerable potential for conveying spatial information in the aviation environment. How reliably these displays will function in that environment will depend in part on the extent to which sound localization is affected by hypoxia. Many aircraft systems operators are routinely exposed to mild hypoxia, and all are at risk of exposure to more severe hypoxia. METHODS: We have examined hypoxia's effect on localization by measuring the localization performance of four subjects exposed to simulated altitudes of 0, 1200, 2400 and 3700 m above sea level for about 30 min. Sounds to be localized were presented from a free-field source at locations covering 360 degrees of azimuth and extending from -47.6 to + 80 degrees elevation. RESULTS: Localization performance was statistically indistinguishable across the altitudes tested. Average localization errors ranged from 12.6 degrees +/- 0.7 SE at 2400 m to 14 degrees +/- 0.84 SE at 0 m. CONCLUSION: The finding that hypoxia induced by exposure to simulated altitudes as high as 3700 m has no effect of sound localization is encouraging with respect to the continued development of 3D audio displays for use in the aviation environment.

Effect of normobaric hypoxia on auditory sensitivity.

Previous psychophysical studies of hypoxia's effects on auditory sensitivity have provided mixed results but the weight of evidence supports the conclusion that sensitivity is unaffected by hypoxia. This conclusion is discrepant with that drawn from physiological studies in which hypoxia has been found to affect auditory-evoked response (AER) latency. One possible explanation of this discrepancy concerns the relatively low maximum frequency (8 kHz) for which hypoxia's effects were assessed in the psychophysical studies. We have extended the range of frequencies over which hypoxia's effects have been examined to include frequencies up to 16 kHz. Thresholds for 1-, 8-, 10-, 12-, 14- and 16-kHz tones were measured at levels of hypoxia equivalent to altitudes of 0, 1,200, 2,400 and 3,700 m. Our results indicate that sensitivity for frequencies up to 16 kHz is unaffected by hypoxia. We suggest that AER latency does not provide a valid measure of auditory sensitivity.

Psychophysical sensitivity and physiological response to amplitude modulation in adult dyslexic listeners.

This study reports two experiments conducted to assess the sensitivity of dyslexic listeners to amplitude modulation (AM) of acoustic stimuli. The smallest detectable depth of AM of white noise was measured as a function of modulation frequency. Dyslexic listeners had significantly higher thresholds of AM depth than did matched control listeners. We also recorded the scalp potential evoked by AM of white noise (the amplitude modulation following response, AMFR). Dyslexic listeners had significantly smaller AMFRs than did matched control listeners. The reduced AMFR is consistent with reduced sensitivity to AM, and there was a strong association between these psychophysical and physiological measures. This deficit in AM sensitivity may result in impaired perception of the AM present in speech.

Aurally and visually guided visual search in a virtual environment.

We investigated the time participants took to perform a visual search task for targets outside the visual field of view using a helmet-mounted display. We also measured the effectiveness of visual and auditory cues to target location. The auditory stimuli used to cue location were noise bursts previously recorded from the ear canals of the participants and were either presented briefly at the beginning of a trial or continually updated to compensate for head movements. The visual cue was a dynamic arrow that indicated the direction and angular distance from the instantaneous head position to the target. Both visual and auditory spatial cues reduced search time dramatically, compared with unaided search. The updating audio cue was more effective than the transient audio cue and was as effective as the visual cue in reducing search time. These data show that both spatial auditory and visual cues can markedly improve visual search performance. Potential applications for this research include highly visual environments, such as aviation, where there is risk of overloading the visual modality with information.

Effect of time and frequency manipulation on syllable perception in developmental dyslexics.

Many people with developmental dyslexia have difficulty perceiving stop consonant contrasts as effectively as other people and it has been suggested that this may be due to perceptual limitations of a temporal nature. Accordingly, we predicted that perception of such stimuli by listeners with dyslexia might be improved by stretching them in time-equivalent to speaking slowly. Conversely, their perception of the same stimuli ought to be made even worse by compressing them in time-equivalent to speaking quickly. We tested 15 children with dyslexia on their ability to identify correctly consonant-vowel-consonant (CVC) stimuli that had been stretched or compressed in the time domain. We also tested their perception of the same CVC stimuli after the formant transitions had been stretched or compressed in the frequency domain. Contrary to our predictions, we failed to find any systematic improvement in their performance with either manipulation. We conclude that simple manipulations in the time and frequency domains are unlikely to benefit the ability of people with dyslexia to discriminate between CVCs containing stop consonants.

Scalp potentials evoked by amplitude-modulated tones in dyslexia.

We recorded the far-field EEG potential evoked by amplitude modulation of acoustic stimuli (the amplitude modulation following response, AMFR) in adults with developmental dyslexia and in a matched control group of adults with no history of reading problems. The mean AMFR recorded from participants with dyslexia was significantly smaller than that recorded from members of the control group. In contrast, the amplitude of the click-evoked auditory brainstem response (ABR) was not significantly different between participant groups. Also, there was no difference between participant groups in the latency of the AMFR or ABR. The reduced AMFR in listeners with dyslexia may reflect impaired ability of the auditory system to follow rapid changes in stimulus energy, a cue believed to be important in the perception of speech.

Acoustic and electric forward-masking of the auditory nerve compound action potential: evidence for linearity of electro-mechanical transduction.

We investigated electro-mechanical transduction within the cochlea by comparing masking of the auditory nerve compound action potential (CAP) by acoustical and electrical maskers. Forward-masking of the CAP reflects the response to the masker of the cochlear location tuned to the probe. Electrical stimulation was delivered through bipolar stimulating electrodes within the basal turn of the scala tympani. The growth of masking of high-frequency probes which excite cochlear locations close to the stimulating electrodes was similar for both acoustic and electrical maskers, suggesting a linear transduction of electrical energy to mechanical energy. Exposure to intense acoustic stimulation caused an equal loss of sensitivity to acoustic and electrical maskers. Masking of lower-frequency probes by electrical maskers increased rapidly with masker current, suggesting the direct electrical stimulation of neural elements. This masking was reduced by the administration of strychnine suggesting a contribution by the efferents towards masking of these low-frequency probes.

Estimating mechanical responses to pulsatile electrical stimulation of the cochlea.

This study estimated the mechanical response of the cochlea to pulsatile electrical stimulation of the scala tympani of the cat. The auditory nerve compound action potential evoked by an acoustic probe was forward-masked by a train of charge-balanced biphasic current pulses. Masking as a function of probe frequency reflected the excitation pattern of the response to the masker and resembled the spectrum of the electrical stimulus. Both pulse rate and pulse width influenced the degree of masking. The vibration of a region of the basilar membrane was estimated by recording the local cochlear microphonic evoked by biphasic pulses. The amplitude of the cochlear microphonic was proportional to the amplitude of the spectral component of the electrical stimulus to which the local cochlear microphonic was tuned. These results are consistent with the generation of a mechanical response to the electrical stimulus.

Comparison of current waveforms for the electrical stimulation of residual low frequency hearing.

Many cochlear prostheses employ charge-balanced biphasic current pulses. These pulses have little energy at low frequencies resulting in limited stimulation of low frequency hearing by mechanical responses to the electrical stimulus. However, if electro-mechanical transduction within the cochlea is nonlinear, electrical stimulation with asymmetric, charge-balanced current pulses may result in a mechanical response with significantly more low frequency energy. We estimated the mechanical response at low frequencies to pulsatile electrical stimulation of the cochlea. The auditory nerve compound action potential evoked by low frequency tones was forward-masked by a train of symmetric or asymmetric current pulses. Masking by asymmetric current pulses was not significantly different from masking by symmetric pulses matched for pulse duration and charge. In conclusion, there appears to be no advantage to using asymmetric current pulses for the mechanical stimulation of residual low frequency hearing by electrical stimulation of the cochlea.

Variability of amplitude and area of the auditory nerve compound action potential.

The strength of neural response to sensory stimuli is often estimated by measurement of the amplitude of gross neural potentials. These gross potentials reflect the summed activity of a population of neurons. The amplitude of these potentials is dependent upon the synchrony of the contributing neural responses. We compared the variability of the peak-to-peak amplitude of the auditory nerve compound action potential (CAP) with that of the area under the peaks. The area under the peaks was significantly less variable than the amplitude for responses to low frequency stimuli. Responses to other stimuli showed differences in the same direction, but these were not significant. We conclude that the area under these peaks is a more precise measure of neural response than measurement of waveform amplitude at least for responses to low frequency stimuli.

Auditory temporal coding in dyslexia.

Developmental dyslexia is generally believed to result from impaired linguistic processing rather than from deficits in low-level sensory function. Challenging this view, we studied the perception of non-verbal acoustic stimuli and low-level auditory evoked potentials in dyslexic adults. Compared with matched controls, dyslexics were selectively impaired in tasks (frequency discrimination and binaural unmasking) which rely on decoding neural discharges phase-locked to the fine structure of the stimulus. Furthermore, this ability to use phase-locking was related to reading ability. In addition, the evoked potential reflecting phase-locked discharges was significantly smaller in dyslexics. These results demonstrate a low-level auditory impairment in dyslexia traceable to the brainstem nuclei.

Phase effects in forward masking of the compound action potential: a comparison of responses to stimulus and distortion frequencies.

When a complex stimulus is presented, new frequencies (distortion products, DPs) are generated within the cochlea. The most intense DPs are lower in frequency than the stimulus tones (primaries). It is not clear whether the relative phase of stimuli is encoded by neural channels tuned to the primaries or by channels tuned to the DPs. We estimated the response of auditory nerve fibres tuned to each of these channels as a function of the relative phase of harmonic stimuli. The compound action potential (CAP) evoked by probes at the primary or distortion frequencies was masked by harmonic 2-tone maskers and cochlear generated DPs. The degree of masking reflected the response to the masker of fibres tuned to the probe. Changes in relative phase of the primaries resulted in a large modulation of the response of fibres tuned to the DPs. Except for a primary frequency ratio of 1:2, the response of fibres tuned to the primaries was only shallowly modulated by changes in relative phase. However, the level of response to the DPs was much lower than the response to the stimulus tones.

The perception of frequency peaks and troughs in wide frequency modulations.

This work was concerned with the perception of "instantaneous pitch" in continuously frequency modulated sounds. In experiment 1, a 70-dB sinusoidal carrier, close to 1 kHz, was modulated by the exponential of periodic functions corresponding to the sum of a few sinusoids [e.g., sin(at)+sin(3at)]. Each modulation had a fundamental frequency (a/2 pi) of 1.5 Hz and was symmetric on the dimensions of time and log frequency. Thirty listeners identified discrete melodic motifs within these stimuli. The pitches of the identified notes mainly corresponded to the local frequency maxima; generally, the local minima were not heard as auditory "events" (pitch singularities). A similar perceptual asymmetry was not observed for comparable sequences of discrete tones. In experiments 2-4, frequency difference limens were measured for the maxima and minima of continuous frequency modulations, using an adaptive forced-choice method. Sinusoidal carriers were modulated by the exponential of one cycle of a 5-Hz cosine function, starting at phase pi or phase 0 and giving an overall frequency swing of about 0.5 oct. For maxima and minima around 1 kHz, frequency shifts of maxima were better detected than frequency shifts of minima, by an average factor of 2. Generally, this asymmetry did not decrease as a function of subjects' training in the discrimination task, and was still present when frequency minima were given a 6-dB intensity advantage over frequency maxima. No explanation was found for the advantage of frequency maxima with respect to perceptual salience (experiment 1) or discriminability (experiments 2-4).

Stimulation of residual hearing in the cat by pulsatile electrical stimulation of the cochlea.

Electrical stimulation of the cochlea may excite residual inner hair cells, either by direct electrical stimulation or through a mechanical event. Hair cell mediated responses of the auditory nerve to electrical stimulation were estimated from forward masking of the compound action potential evoked by an acoustic probe. Masking by a fixed electrical masker peaked for probes equal in frequency to the pulse repetition rate and its second harmonic, suggesting a spatially tuned profile of excitation within the cochlea. Furthermore, the tuning curves for masking of a fixed acoustic probe peaked for masker pulse rates close to the frequency of the probe. A secondary peak of masking was commonly seen for electrical stimulation at one half of the probe frequency, suggesting masking of the probe by the second harmonic of the electrical stimulus. These results suggest that pulsatile stimulation at the base of the cochlea generates a spectrally rich mechanical disturbance in which each component propagates to its place of resonance in the cochlea.

Comparison of half-band and full-band electrodes for intracochlear electrical stimulation.

It was hypothesized that intracochlear stimulating electrodes oriented toward the modiolus would require a lower stimulus current to elicit a threshold neural response than longitudinal band electrodes. Electrically evoked auditory brain stem responses (EABRs) and electrode impedances were recorded with full-band and oriented half-band scala tympani electrodes in anesthetized, deafened cats. To elicit a threshold EABR, the stimulus current required for stimulation through half-band electrodes oriented toward the modiolus was not significantly different from the current required for stimulation through full-band electrodes. The impedances of full-band electrodes were significantly lower than those of half-band electrodes. Considering the significantly higher impedance and current density of half-band electrodes in comparison to full-band electrodes, as well as the critical orientation of half-band electrodes during implantation, we believe that full-band electrodes have advantages over half-band electrodes for scala tympani implantation.

Hair cell mediated responses of the auditory nerve to sinusoidal electrical stimulation of the cochlea in the cat.

Electrical stimulation of the cochlea elicits discharges of auditory nerve fibres which are mediated by the electrical or mechanical stimulation of inner hair cells (electrophonic responses). In order to isolate hair-cell mediated responses from those elicited by electrical stimulation of the nerve, the compound action potential (CAP) evoked by an acoustic probe was forward-masked by sinusoidal monopolar, or localized bipolar electrical stimulation of the base of the cochlea. The degree of masking of a given probe estimated the synaptically mediated response to the masker of the population of auditory nerve fibres innervating the cochlear location tuned to the probe. There was a peak of masking for probes close to the frequency of the electrical stimulus, suggesting a spatial tuning of the hair cell mediated response along the cochlea. This is consistent with excitation of the inner hair cells by a propagating mechanical response which is generated within the electrical field at the base of the cochlea. Furthermore, tuning curves for masking of a given probe were sharply tuned to electrical stimulation close to the probe frequency. This masking was not dependent upon the presence of functional outer hair cells close to the electrodes, suggesting an alternate transduction of electrical to mechanical energy.

A gated differential amplifier for recording physiological responses to electrical stimulation.

Artifact from electrical stimulation imposes a problem for the recording of physiological responses to electrical stimulation. Here we describe a simple, low-cost, gated differential amplifier for the recording of physiological responses to electrical stimulation. The gain of the amplifier is set to 1 during electrical stimulation by setting the gate input to a high logic state to avoid overloading of the amplifier by the artifact. Following electrical stimulation, the gate input is set to a low logic state, resulting in a gain of 1000 for frequencies between 300 Hz and 25 kHz (-3 dB points). The gain at low frequencies (0-0.2 Hz) is held constant at 1 to avoid transients in the output signal arising from changes in gain at these frequencies. The gain of the amplifier following stimulation (gate low) was independent of the magnitude of the artifact and was therefore suitable for the measurement of neural field potentials with low impedance electrodes.