Continuous White Noise Research Articles

Protection from Noise-Induced Temporary Threshold Shift by d-Methionine is Associated with Preservation of ATPase Activities

The present study was designed to test whether noise-induced temporary threshold shift (TTS) could be attenuated by D-methionine and its possible relation to the biochemical changes of cochlear lateral walls such as ATPase activities and oxidative stress in guinea pigs. Thirty-two normal-hearing male guinea pigs were randomly divided into saline-treated and D-methionine-treated (300 mg/kg) experimental groups. One hour after treatment, they were exposed to a continuous broadband white noise at 105 +/- 2 dB sound pressure level for 10 min, causing TTS. Each group was then divided into four subgroups based on the number of survival days after noise exposure (0, 1, 2, and 7 days). Each subgroup had four animals and eight ears included. By means of click-evoked auditory brain stem responses (ABR), auditory thresholds of guinea pigs were measured before noise exposure, immediately after noise exposure, and before killing. After animals were killed, cochlear lateral walls were immediately harvested and assayed for enzyme-specific activities of Na+, K+-ATPase and Ca2+-ATPase, lipid peroxidation, and nitric oxide. A 15.31 +/- 3.80 dB threshold shift was found immediately after noise exposure in saline-pretreated guinea pigs. In contrast, ABR threshold shift was significantly attenuated to 4.06 +/- 2.35 dB in D-methionine-treated animals. Furthermore, D-methionine enhanced the restoration of ABR threshold to baseline level by 1 day. In addition, noise significantly decreased Na+, K+-ATPase, and Ca2+-ATPase activities and increased lipid peroxidation and nitric oxide levels of the cochlear lateral walls. D-methionine significantly protected against all of these changes. Noise not only induced TTS but also inhibited ATPase activities as well as increased oxidative stress in guinea-pig cochlear lateral walls; all of these changes could be attenuated by d-methionine through its antioxidative property. These results suggest the potential usefulness of d-methionine in protecting from noise-induced ototoxicity.

The auditory P 50 component to onset and offset of sound

Objective The auditory Event-Related Potentials (ERP) of component P 50 to sound onset and offset have been reported to be similar, but their magnetic homologue has been reported absent to sound offset. We compared the spatio-temporal distribution of cortical activity during P 50 to sound onset and offset, without confounds of spectral change. Methods ERPs were recorded in response to onsets and offsets of silent intervals of 0.5 s (gaps) appearing randomly in otherwise continuous white noise and compared to ERPs to randomly distributed click pairs with half second separation presented in silence. Subjects were awake and distracted from the stimuli by reading a complicated text. Measures of P 50 included peak latency and amplitude, as well as source current density estimates to the clicks and sound onsets and offsets. Results P 50 occurred in response to noise onsets and to clicks, while to noise offset it was absent. Latency of P 50 was similar to noise onset (56 ms) and to clicks (53 ms). Sources of P 50 to noise onsets and clicks included bilateral superior parietal areas. In contrast, noise offsets activated left inferior temporal and occipital areas at the time of P 50. Source current density was significantly higher to noise onset than offset in the vicinity of the temporo-parietal junction. Conclusions P 50 to sound offset is absent compared to the distinct P 50 to sound onset and to clicks, at different intracranial sources. P 50 to stimulus onset and to clicks appears to reflect preattentive arousal by a new sound in the scene. Sound offset does not involve a new sound and hence the absent P 50. Significance Stimulus onset activates distinct early cortical processes that are absent to offset.

Continuous white noise exposure during and after auditory critical period differentially alters bidirectional thalamocortical plasticity in rat auditory cortex in vivo

Long-term potentiation (LTP) and long-term depression are thought to mediate activity-dependent brain plasticity but their role in the development of the thalamocortical auditory system in vivo has not been investigated. In adult urethane-anaesthetized rats, theta-burst stimulation of the medial geniculate nucleus produced robust LTP (40% amplitude enhancement) of field post-synaptic evoked potentials recorded in the superficial layers of the primary auditory cortex. Low-frequency (1-Hz) stimulation resulted in transient depression ( approximately 40%) of field post-synaptic evoked potential amplitude. Both LTP and synaptic depression were found to be dependent on cortical N-methyl-d-aspartate receptors. Thalamocortical plasticity was also assessed after continuous white noise exposure, thought to arrest auditory cortex maturation when applied during the critical period of post-natal primary auditory cortex development. Rats housed in continuous white noise for the first 50 days of post-natal life exhibited greater LTP ( approximately 80%) than controls reared in unaltered acoustic environments. The protocol used to elicit depression also resulted in substantial LTP ( approximately 50%) in white noise-reared animals. Adults housed in white noise for the same length of time exhibited normal LTP but displayed greater and persistent levels of synaptic depression ( approximately 70%). Thus, the absence of patterned auditory stimulation during early post-natal life appears to retard sensory-dependent thalamocortical synaptic strengthening, as indicated by the preferential readiness for synaptic potentiation over depression. The fact that the same auditory manipulation in adults results in synapses favouring depression demonstrates the critical role of developmental stage in determining the direction of synaptic modification in the thalamocortical auditory system.

Effects of continuous masking noise on tone-evoked magnetic fields in humans

Two different types of steep loudness growth have been reported in detail in psychoacoustical studies but have rarely been evaluated by objective methods in humans. One occurs in inner-ear hearing-impaired patients and is known as loudness recruitment. Another similar phenomenon is observed in healthy subjects with concurrent presence of background noise. Concerning the first type, our previous study using magnetoencephalography (MEG) showed that enhancement of the dipole moment of N100m with increase in stimulus intensity was greater in patients than in normal individuals. However, it is unclear whether the enhancement of activity in auditory cortex will also be detected with background noise in healthy subjects. To elucidate the effects of continuous background noise on tone-evoked cortical activity, we measured auditory-evoked magnetic fields (AEFs) from 7 normal-hearing subjects in two different conditions, with and without 55 dB SPL continuous masking white noise (noise/quiet conditions). The stimuli were 200 ms 1-kHz tones delivered monaurally and randomly at 4 different intensities (40–70 dB SPL) with constant 1-s interstimulus intervals. The N100m increased in amplitude and decreased in latency as a function of stimulus intensity in both noise and quiet conditions. The dipole moment of N100m was significantly smaller in the noise than in the quiet condition, showing that continuous background noise suppresses the strength of tone-evoked cortical responses. The mechanisms underlying these two psychoacoustically similar phenomena of rapid loudness growth thus differ.

Cutting Process Monitoring by Means of Acoustic Emission Method; Part I - New Approach of Acoustic Emission Sensor; Part II - Transformation of Acoustic Emission into Audible Sound

Development of modern society is converging to a status where many human actions can be performed by machines. To achieve production without human intervention, machines require artificial receptors. Data gathering for processing and analysis of signals, together with determination of feedback reactions can be achieved by a suitable decision maker unit. A sensed value suited to this so-called intelligent sensing process would be the acoustic emission signal. In the case of intelligent cutting tools this would require miniature highly sensitive sensors integrated into the cutting tool body. Part I of this paper deals with the possibility of practical usage of the piezoelectric properties of copolymer foils for the acoustic emission sensor as a transducer of a mechanical surface wave into electrical signal. Part II of the paper deals with the most fundamental requirement for monitoring of cutting conditions during machining, i.e. excellent processing of measured data. Data obtained from machining process obtained by means of acoustic emission sensors, as discussed in the first part of this article, have high-frequency and continuous character of a white noise. These data are very difficult to process. New apparatus for transformation of acoustic emission into audible sound in the workplace is presented. The first stage of processing is by listening to transformed data it is subjectively possible to recognize differences in audible spectrum, corresponding to different states of the cutting tool. The second step is visualization of the differences via the fast Fourier transform (FFT) in the spectrum graphic chart.

Effects of continuous conditioning noise and light on the auditory- and visual-evoked potentials of the guinea pig

Neurophysiological studies aiming to explore how the brain integrates information from different brain regions are increasing in the literature. The aim of the present study is to explore intramodal (binaural, binocular) and intermodal (audio-visual) interactions in the guinea pig brain through the observation of changes in evoked potentials by generalized continuous background activity. Seven chronically prepared animals were used in the study and the recordings were made as they were awake. Epidural electrodes were implanted to the skulls by using stereotaxic methods. Continuous light for retinal or continuous white noise for cochlear receptors were used as continuous conditioning stimuli for generalized stimulation. To evoke auditory or visual potentials, click or flash were used as transient imperative stimuli. The study data suggest that (a) white noise applied to one ear modifies the response to click in the contralateral ear which is a binaural interaction; (b) continuous light applied to one eye modifies the response to flash applied to the contralateral eye which is interpreted as a binocular interaction; (c) regardless of the application side, white noise similarly modified the response to flash applied to the either eye connoting a nonspecific effect of white noise on vision, independent from spatial hearing mechanisms; (d) on the other hand, continuous light, in either eye, did not affect the response to click applied to any ear, reminding a ‘one-way’ interaction that continuous aural stimulation affects visual response.

Detrimental Effects of Irrelevant Speech on Serial Recall of Visual Items are Reflected in Reduced Visual N1 and Reduced Theta Activity

The term irrelevant sound effect (ISE) describes an empirically robust finding in which serial recall performance of visual items is reduced by irrelevant speech. At present little is known about its neurophysiological basis. Although some previous neuroelectric studies have concentrated on responses elicited by irrelevant background sound, whether the processing of visually presented to-be-remembered digits itself is affected by irrelevant speech has yet to be studied. An experiment (n = 20) was conducted in which serial recall performance for visually presented digits was tested during exposure to either irrelevant speech, continuous white noise or silence while measuring EEG activity. White noise was chosen as a control condition, because it constitutes auditory stimulation while leaving serial recall performance unimpaired. In addition to replicating the detrimental behavioural effect of irrelevant speech, an analogous speech-specific early decrease ( approximately 160 ms) in the visual ERP (N1) and subsequently a reduced theta response (4-6 Hz; 400-800 ms) at right prefrontal electrodes were observed. Although irrelevant sound presentation was restricted to the visual presentation phase, power spectra reveal that the weaker theta response for speech persisted in a silent retention phase before serial recall. Based on such data we propose re-evaluating the role attention plays in explaining the ISE.

The composite N 1 component to gaps in noise

Objective To indicate whether the double peaked N 1 to gaps in continuous white noise is a composite of onset and offset responses to transients or whether it reflects higher processing such as change or mismatch detection and to assess the role of attention in this process. Methods Evoked potentials were recorded to two binaural stimulus types: (1) gaps of different durations randomly distributed in continuous white noise; and (2) click pairs at intervals identical to those between gap onsets and offsets in the continuous noise stimulus. Potentials to these stimuli were recorded while subjects read a text and while detecting gaps in noise or click pairs. Results Potentials were detected to all click pairs and to gaps of 5 ms or longer, corresponding to the subjects' psychoacoustic gap detection threshold. With long gap durations of 200–800 ms, distinct potentials to gap onset and gap offset were observed. The waveforms to all click pairs and to offsets of long gaps were similar and single-peaked, while potentials to gaps of 10 ms and longer, and potentials to onsets of long gaps were double-peaked, consisting of two N 1 negativities, 60 ms apart, irrespective of gap duration. The first (N 1a), was more frontal in its distribution and similar to that of clicks. The second (N 1b) peak's distribution was more central/temporal and its source locations and time course of activity were distinct. No effects of attention on any of the varieties and constituents of N 1 were observed. Conclusions Comparing potentials to gap onsets, to click pairs and to gap offsets, suggests that potentials to gap onsets involve not only sound onset/offset responses (N 1, N 1a) but also the subsequent pre-attentive perception of the cessation of an ongoing sound (N 1b). We propose that N 1b is distinct from change or mismatch detection and is associated with termination of an ongoing continuous stimulus. We propose to call it the N(egation)-process. Significance A constituent of the N 1 complex is shown to be associated with the pre-attentive perception of termination of an ongoing stimulus and to have distinct scalp distribution and intracranial sources.

Hearing in Fishes under Noise Conditions

Our current knowledge on sound detection in fishes is mainly based on data acquired under quiet laboratory conditions. However, it is important to relate auditory thresholds to background noise in order to determine the signal-detecting abilities of animals in the natural environment. We investigated the influence of two noise levels within the naturally occurring range on the auditory sensitivity of two hearing specialists (otophysines) and a hearing generalist. Audiograms of the goldfish Carassius auratus, the lined Raphael catfish Platydoras costatus and the pumpkinseed sunfish Lepomis gibbosus (hearing generalist) were determined between 200 and 4000 Hz (100-800 Hz for L. gibbosus) under laboratory conditions and under continuous white noise by recording auditory evoked potentials (AEPs). Baseline thresholds showed greatest hearing sensitivity around 500 Hz in goldfish and catfish and at 100 Hz in the sunfish. Continuous white noise of 110 dB RMS elevated the thresholds by 15-20 dB in C. auratus and by 4-22 dB in P. costatus. White noise of 130 dB RMS elevated overall hearing thresholds significantly in the otophysines by 23-44 dB. In the goldfish, threshold did not shift at 4 kHz. In contrast, auditory thresholds in the sunfish declined only at the higher noise level by 7-11 dB. Our data show that the AEP recording technique is suitable for studying masking in fishes, and that the occurrence and degree of the threshold shift (masking) depend on the hearing sensitivity of fishes, the frequency, and noise levels tested. The results indicate that acoustic communication and orientation of fishes, in particular of hearing specialists, are limited by noise regimes in their environment.

Early ERP components to gaps in white noise: onset and offset effects

Evoked potentials to two stimulus types were compared: (1) binaural continuous white noise with randomly distributed gaps of different durations and (2) click pairs presented with intervals identical to the gap durations. Potentials were detected in response to gaps of 5 ms or longer, corresponding to the subjects' psychoacoustic gap detection threshold. The waveforms to all click pairs and to gap offsets were similar and single-peaked, while response to gap onset included two N1 negativities about 60 ms apart, regardless of gap duration. The first peak (N1a) was more frontal in its distribution and similar to that of clicks. The second (N1b) peak's scalp distribution was more central/temporal. The differences between waveforms to click pairs and gap offsets, compared with gap onsets, suggest that potentials to gap onsets involve more than just onset/offset responses (N1, N1a) but also the subsequent perception of the cessation of an ongoing sound (N1b). In previous studies, two processes were defined: (1) M-process to deviation from the preceding auditory pattern and (2) C-process in response to acoustic change within an ongoing stream. We propose that N1b is different from these processes because it relates to the negation of an ongoing stimulus, and we thus call it an N-process.

Hearing in Fishes under Noise Conditions

Our current knowledge on sound detection in fishes is mainly based on data acquired under quiet laboratory conditions. However, it is important to relate auditory thresholds to background noise in order to determine the signal-detecting abilities of animals in the natural environment. We investigated the influence of two noise levels within the naturally occurring range on the auditory sensitivity of two hearing specialists (otophysines) and a hearing generalist. Audiograms of the goldfish Carassius auratus, the lined Raphael catfish Platydoras costatus and the pumpkinseed sunfish Lepomis gibbosus (hearing generalist) were determined between 200 and 4000 Hz (100–800 Hz for L. gibbosus) under laboratory conditions and under continuous white noise by recording auditory evoked potentials (AEPs). Baseline thresholds showed greatest hearing sensitivity around 500 Hz in goldfish and catfish and at 100 Hz in the sunfish. Continuous white noise of 110 dB RMS elevated the thresholds by 15–20 dB in C. auratus and by 4–22 dB in P. costatus. White noise of 130 dB RMS elevated overall hearing thresholds significantly in the otophysines by 23–44 dB. In the goldfish, threshold did not shift at 4 kHz. In contrast, auditory thresholds in the sunfish declined only at the higher noise level by 7–11 dB. Our data show that the AEP recording technique is suitable for studying masking in fishes, and that the occurrence and degree of the threshold shift (masking) depend on the hearing sensitivity of fishes, the frequency, and noise levels tested. The results indicate that acoustic communication and orientation of fishes, in particular of hearing specialists, are limited by noise regimes in their environment.

Dynamics of the contralateral white noise-induced enhancement in the guinea pig's middle latency response

The peak-to-peak amplitude of temporal middle latency response (MLR) of the guinea pig, evoked by a click in the contralateral ear, according to the recording side, is increased with the presence of continuous white noise (CWN) in the ipsilateral ear and this specialty is defined as the white noise enhancement (WNE). This phenomenon is evaluated as an interesting electrophysiological finding from the viewpoint of binaural interaction and in this study, its dynamic specifications were investigated. After the beginning of ipsilateral CWN, significant WNE was observed at 275th ms and it reached to a maximum, with an increase more than 40%, at 350th ms. After a habituation occurred, WNE reached to 20% on the 4th second by gradually decreasing and came to a steady state. In the time window between 2 and 5 ms after CWN started, a surprising amplitude decrease is observed. Therefore, CWN causes an effect, like a click, in the short-term and this on-response type effect originates from low level binaural centers, which decreases the MLR amplitude. However, the same CWN increases the MLR amplitude (WNE) by the effects over the high level binaural centers in the succeeding period, by its continuous characteristic.

Noise robust speech recognition with support vector learning algorithms

We propose a new noise robust speech recognition system using time-frequency domain analysis and radial basis function (RBF) support vector machines (SVM). Here, we ignore the effects of correlative and nonstationary noise and only focus on continuous additive Gaussian white noise. We then develop an isolated digit/command recognizer and compare its performance to two other systems, in which the SVM classifier has been replaced by multilayer perceptron (MLP) and RBF neural networks. All systems are trained under the low signal-to-noise ratio (SNR) condition. We obtained the best correct classification rate of 83% and 52% for digit recognition on the TI-46 corpus for the SVM and MLP systems, respectively under the SNR=0 (dB), while we could not train the RBF network for the same dataset. The newly developed speech recognition system seems to be noise robust for medium size speech recognition problems under continuous, stationary background noise. However, it is still required to test the system under realistic noisy environment to observe whether the system keeps its adaptability and robustness under such conditions. [Work supported in part by grants from DARPA CBS, NASA, and ONR.]

Boundary coiflets for wavelet shrinkage in function estimation

There are standard modifications of certain compactly supported wavelets that yield orthonormal bases on a bounded interval. We extend one such construction to those wavelets, such as ‘coiflets', that may have fewer vanishing moments than had to be assumed previously. Our motivation lies in function estimation in statistics. We use these boundary-modified coiflets to show that the discrete wavelet transform of finite data from sampled regression models asymptotically provides a close approximation to the wavelet transform of the continuous Gaussian white noise model. In particular, estimation errors in the discrete setting of computational practice need not be essentially larger than those expected in the continuous setting of statistical theory.

Boundary coiflets for wavelet shrinkage in function estimation

There are standard modifications of certain compactly supported wavelets that yield orthonormal bases on a bounded interval. We extend one such construction to those wavelets, such as ‘coiflets', that may have fewer vanishing moments than had to be assumed previously. Our motivation lies in function estimation in statistics. We use these boundary-modified coiflets to show that the discrete wavelet transform of finite data from sampled regression models asymptotically provides a close approximation to the wavelet transform of the continuous Gaussian white noise model. In particular, estimation errors in the discrete setting of computational practice need not be essentially larger than those expected in the continuous setting of statistical theory.

Contralateral White Noise Masking Affects Auditory N1 and P2 Waves Differently

Abstract In this study, we examined the effect of contralateral masking on cortical auditory evoked potentials N1 (modal-specific slowly adapting component) and P2 at different masking intensities. N1 and P2 potentials were recorded from 15 subjects with normal hearing using 500Hz tone pips (intensity 65dB HL, duration 100ms, ISI 1s) presented to the right ear. Continuous white noise was delivered to the left ear at the intensities of 35, 50, 65, or 75dB effective masking level (EML), as well as a no-mask condition. The electrodes F3, Fz, F4, C3, Cz, C4, and Pz were used. The results show that N1 amplitude was significantly attenuated and, in contrast, P2 amplitude was significantly increased, with contralateral 75dB EML white noise. N1P2 peak to peak amplitude was not affected by masking, nor were the peak latencies. Thus, contralateral masking affects the exogenous cortical evoked N1 and P2 curves differently. We suggest that the effect is mediated by the efferent hearing system. The effect of ≤ 50dB EML contralateral white noise masking is so small that it should not affect clinical recordings.

Stimulus conditioned to foot-shock stress reinstates alcohol-seeking behavior in an animal model of relapse.

Stress and conditioned responses to drug cues have been implicated as critical factors in relapse to drug use. In the animal literature, both the conditioned effects of drug-related stimuli and the unconditioned effects of foot-shock stress have been well documented to reinstate extinguished drug-seeking behavior. What has remained largely unexplored, however, is the significance of stimuli conditioned to foot-shock stress for the resumption of drug seeking. Additionally, although relapse is often the result of several risk factors acting in combination, the possibility that interactions among risk factors such as conditioned stress and drug cues may intensify drug-seeking behavior has received little experimental attention. The purpose of this study was to examine the individual and interactive effects of a stimulus conditioned to foot-shock stress (STRESS CS) and a stimulus conditioned to ethanol reward (EtOH CS) on the reinstatement of ethanol-seeking behavior following extinction. Male Wistar rats were trained to orally self-administer 10% ethanol on a fixed-ratio 3 schedule of reinforcement. The EtOH CS was established by response-contingently pairing 0.5 s illumination of a white cue light with each reinforced response. The STRESS CS was established by pairing a continuous white noise (70 dB) with intermittent foot shock (10 min; 0.5 mA; 0.5 s on; mean off period of 40 s). Ethanol dependence was induced by an ethanol vapor-inhalation procedure. After ethanol-maintained instrumental responding was extinguished by withholding ethanol and the EtOH CS, reinstatement tests were conducted. Both exposure to the STRESS CS and response-contingent presentation of the EtOH CS reinstated extinguished responding at the previously active, ethanol-paired lever without further ethanol availability. When response-contingent availability of the EtOH CS was preceded by exposure to the STRESS CS, interactive effects of these stimuli on responding were observed. However, both the individual and interactive effects of the STRESS CS and the EtOH CS reached statistical significance only in rats with a history of ethanol dependence but not in ethanol-nondependent rats. The results confirm that both conditioned stress and ethanol cues elicit ethanol-seeking behavior and, more importantly, that these stimuli produce interactive effects resulting in an increased ethanol-seeking response. The findings also indicate that susceptibility to ethanol seeking induced by conditioned stress and alcohol cues depends significantly on the history of prior alcohol exposure.

Temporal modulation transfer functions in the barn owl (Tyto alba).

Barn owls (Tyto alba) have evolved several specializations in their auditory system to achieve the high sensory acuity required for prey capture, including superior processing of interaural time differences and phase coding in the auditory periphery. Here, we tested whether barn owls are capable of high temporal resolution that may be a prerequisite for the accuracy in binaural processing. Temporal resolution was measured psychoacoustically and demonstrated in temporal modulation transfer functions. Four barn owls were trained in an operant task with food reward to detect sinusoidal amplitude modulations within an 800-ms gated white-noise burst or 800-ms periods of modulation in continuous white noise (spectrum levels of -5 dB and 15 dB SPL). Within the range of tested amplitude modulation frequencies from 5 Hz to 1280 Hz, barn owls' detection thresholds were lowest at 10-20 Hz. This sensitivity corresponds to an intensity-difference limen of between 0.9 dB and 1.4 dB. For all conditions, temporal modulation transfer functions showed band-pass characteristics with a high-frequency cutoff in the range of 37 Hz to 92 Hz, corresponding to minimum integration times of 4.3 ms and 1.7 ms, respectively. In summary, these data indicate a temporal resolution in the owl's auditory system that is good, but not unusual, compared to other vertebrates.

Original Article: The effect of blindness on horizontal plane sound source identification: El efecto de la ceguera en la identificatión de la fuente sonora en el piano horizontal

The effect of blindness on sound source identification was studied. Four groups of normally-hearing adults, two sighted and two blind, participated. Subjects were tested using arrays of four and eight loudspeakers, surrounding them in the horizontal plane. One sighted group was tested in quiet. The other groups were tested in continuous 60-dB SPL white noise. Three 75-dB SPL 300-ms stimuli were localized: one-third octave noise bands, centered at 0.5 and 4 kHz, and broadband noise. Broadband noise was easiest to localize (both binaural and spectral cues available), and the 0.5-kHz noise band was the most difficult (primarily interaural temporal difference cue available). Subjects with late-onset blindness achieved significantly higher scores than the early blind and blindfolded sighted subjects. The percentage correct decreased with an increase in the number of speakers, but background noise had no effect. The results attest to the benefit of early visual experience for spatial hearing in adulthood, and demonstrate the negative impact of sudden loss of sight.

CHAOTIC RESONANCE — METHODS AND APPLICATIONS FOR ROBUST CLASSIFICATION OF NOISY AND VARIABLE PATTERNS

A fundamental tenet of the theory of deterministic chaos holds that infinitesimal variation in the initial conditions of a network that is operating in the basin of a low-dimensional chaotic attractor causes the various trajectories to diverge from each other quickly. This "sensitivity to initial conditions" might seem to hold promise for signal detection, owing to an implied capacity for distinguishing small differences in patterns. However, this sensitivity is incompatible with pattern classification, because it amplifies irrelevant differences in incomplete patterns belonging to the same class, and it renders the network easily corrupted by noise. Here a theory of stochastic chaos is developed, in which aperiodic outputs with 1/f2 spectra are formed by the interaction of globally connected nodes that are individually governed by point attractors under perturbation by continuous white noise. The interaction leads to a high-dimensional global chaotic attractor that governs the entire array of nodes. An example is our spatially distributed KIII network that is derived from studies of the olfactory system, and that is stabilized by additive noise modeled on biological noise sources. Systematic parameterization of the interaction strengths corresponding to synaptic gains among nodes representing excitatory and inhibitory neuron populations enables the formation of a robust high-dimensional global chaotic attractor. Reinforcement learning from examples of patterns to be classified using habituation and association creates lower dimensional local basins, which form a global attractor landscape with one basin for each class. Thereafter, presentation of incomplete examples of a test pattern leads to confinement of the KIII network in the basin corresponding to that pattern, which constitutes many-to-one generalization. The capture after learning is expressed by a stereotypical spatial pattern of amplitude modulation of a chaotic carrier wave. Sensitivity to initial conditions is no longer an issue. Scaling of the additive noise as a parameter optimizes the classification of data sets in a manner that is comparable to stochastic resonance. The local basins constitute dynamical memories that solve difficult problems in classifying data sets that are not linearly separable. New local basins can be added quickly from very few examples without loss of existing basins. The attractor landscape enables the KIII set to provide an interface between noisy, unconstrained environments and conventional pattern classifiers. Examples given here of its robust performance include fault detection in small machine parts and the classification of spatiotemporal EEG patterns from rabbits trained to discriminate visual stimuli.