AIRSTREAM: A Neural Network Model of Auditory Scene Analysis and Source Segregation

Multiple sound sources often contain harmonics that overlap and may be degraded by environmental noise. The auditory system is capable of teasing apart these sources into distinct mental objects, or streams. Such an "auditory scene analysis" enables the brain to solve the cocktail party problem. A neural network model of auditory scene analysis, called the AIRSTREAM model, is presented to propose how the brain accomplishes this feat. The model clarifies how the frequency components that correspond to a give acoustic source may be coherently grouped together into distinct streams based on pitch and spatial cues. The model also clarifies how multiple streams may be distinguishes and seperated by the brain. Streams are formed as spectral-pitch resonances that emerge through feedback interactions between frequency-specific spectral representaion of a sound source and its pitch. First, the model transforms a sound into a spatial pattern of frequency-specific activation across a spectral stream layer. The sound has multiple parallel representations at this layer. A sound's spectral representation activates a bottom-up filter that is sensitive to harmonics of the sound's pitch. The filter activates a pitch category which, in turn, activate a top-down expectation that allows one voice or instrument to be tracked through a noisy multiple source environment. Spectral components are suppressed if they do not match harmonics of the top-down expectation that is read-out by the selected pitch, thereby allowing another stream to capture these components, as in the "old-plus-new-heuristic" of Bregman. Multiple simultaneously occuring spectral-pitch resonances can hereby emerge. These resonance and matching mechanisms are specialized versions of Adaptive Resonance Theory, or ART, which clarifies how pitch representations can self-organize durin learning of harmonic bottom-up filters and top-down expectations. The model also clarifies how spatial location cues can help to disambiguate two sources with similar spectral cures. Data are simulated from psychophysical grouping experiments, such as how a tone sweeping upwards in frequency creates a bounce percept by grouping with a downward sweeping tone due to proximity in frequency, even if noise replaces the tones at their interection point. Illusory auditory percepts are also simulated, such as the auditory continuity illusion of a tone continuing through a noise burst even if the tone is not present during the noise, and the scale illusion of Deutsch whereby downward and upward scales presented alternately to the two ears are regrouped based on frequency proximity, leading to a bounce percept. Since related sorts of resonances have been used to quantitatively simulate psychophysical data about speech perception, the model strengthens the hypothesis the ART-like mechanisms are used at multiple levels of the auditory system. Proposals for developing the model to explain more complex streaming data are also provided.

Parallel Auditory Filtering By Sustained and Transient Channels Separates Coarticulated Vowels and Consonants

A neural model of peripheral auditory processing is described and used to separate features of coarticulated vowels and consonants. After preprocessing of speech via a filterbank, the model splits into two parallel channels, a sustained channel and a transient channel. The sustained channel is sensitive to relatively stable parts of the speech waveform, notably synchronous properties of the vocalic portion of the stimulus it extends the dynamic range of eighth nerve filters using coincidence deteectors that combine operations of raising to a power, rectification, delay, multiplication, time averaging, and preemphasis. The transient channel is sensitive to critical features at the onsets and offsets of speech segments. It is built up from fast excitatory neurons that are modulated by slow inhibitory interneurons. These units are combined over high frequency and low frequency ranges using operations of rectification, normalization, multiplicative gating, and opponent processing. Detectors sensitive to frication and to onset or offset of stop consonants and vowels are described. Model properties are characterized by mathematical analysis and computer simulations. Neural analogs of model cells in the cochlear nucleus and inferior colliculus are noted, as are psychophysical data about perception of CV syllables that may be explained by the sustained transient channel hypothesis. The proposed sustained and transient processing seems to be an auditory analog of the sustained and transient processing that is known to occur in vision.

Changes in the McGurk Effect Across Phonetic Contexts

To investigate the process underlying audiovisual speech perception, the McGurk illusion was examined across a range of phonetic contexts. Two major changes were found. First, the frequency of illusory /g/ fusion percepts increased relative to the frequency of illusory /d/ fusion percepts as vowel context was shifted from /i/ to /a/ to /u/. This trend could not be explained by biases present in perception of the unimodal visual stimuli. However, the change found in the McGurk fusion effect across vowel environments did correspond systematically with changes in second format frequency patterns across contexts. Second, the order of consonants in illusory combination percepts was found to depend on syllable type. This may be due to differences occuring across syllable contexts in the timecourses of inputs from the two modalities as delaying the auditory track of a vowel-consonant stimulus resulted in a change in the order of consonants perceived. Taken together, these results suggest that the speech perception system either fuses audiovisual inputs into a visually compatible percept with a similar second formant pattern to that of the acoustic stimulus or interleaves the information from different modalities, at a phonemic or subphonemic level, based on their relative arrival times.

Linking Attention to Learning, Expectation, Competition, and Consciousness

The concept of attention has been used in many senses, often without clarifying how or why attention works as it does. Attention, like consciousness, is often described in a disembodied way. The present article summarizes neural models and supportive data and how attention is linked to processes of learning, expectation, competition, and consciousness. A key them is that attention modulates cortical self-organization and stability. Perceptual and cognitive neocortex is organized into six main cell layers, with characteristic sub-lamina. Attention is part of unified design of bottom-up, horizontal, and top-down interactions among indentified cells in laminar cortical circuits. Neural models clarify how attention may be allocated during processes of visual perception, learning and search; auditory streaming and speech perception; movement target selection during sensory-motor control; mental imagery and fantasy; and hallucination during mental disorders, among other processes.

Neural Modeling and Imaging of the Cortical Interactions Underlying Syllable Production

This paper describes a neural model of speech acquisition and production that accounts for a wide range of acoustic, kinematic, and neuroimaging data concerning the control of speech movements. The model is a neural network whose components correspond to regions of the cerebral cortex and cerebellum, including premotor, motor, auditory, and somatosensory cortical areas. Computer simulations of the model verify its ability to account for compensation to lip and jaw perturbations during speech. Specific anatomical locations of the model's components are estimated, and these estimates are used to simulate fMRI experiments of simple syllable production with and without jaw perturbations.

Neural Dynamics of Saccadic and Smooth Pursuit Eye Movement Coordination during Visual Tracking of Unpredictably Moving Targets

How does the brain use eye movements to track objects that move in unpredictable directions and speeds? Saccadic eye movements rapidly foveate peripheral visual or auditory targets and smooth pursuit eye movements keep the fovea pointed toward an attended moving target. Analyses of tracking data in monkeys and humans reveal systematic deviations from predictions of the simplest model of saccade-pursuit interactions, which would use no interactions other than common target selection and recruitment of shared motoneurons. Instead, saccadic and smooth pursuit movements cooperate to cancel errors of gaze position and velocity, and thus to maximize target visibility through time. How are these two systems coordinated to promote visual localization and identification of moving targets? How are saccades calibrated to correctly foveate a target despite its continued motion during the saccade? A neural model proposes answers to such questions. The modeled interactions encompass motion processing areas MT, MST, FPA, DLPN and NRTP; saccade planning and execution areas FEF and SC; the saccadic generator in the brain stem; and the cerebellum. Simulations illustrate the model’s ability to functionally explain and quantitatively simulate anatomical, neurophysiological and behavioral data about SAC-SPEM tracking.

Speaker Normalization Using Cortical Strip Maps: A Neural Model for Steady State Vowel Identification

Auditory signals of speech are speaker-dependent, but representations of language meaning are speaker-independent. Such a transformation enables speech to be understood from different speakers. A neural model is presented that performs speaker normalization to generate a pitchindependent representation of speech sounds, while also preserving information about speaker identity. This speaker-invariant representation is categorized into unitized speech items, which input to sequential working memories whose distributed patterns can be categorized, or chunked, into syllable and word representations. The proposed model fits into an emerging model of auditory streaming and speech categorization. The auditory streaming and speaker normalization parts of the model both use multiple strip representations and asymmetric competitive circuits, thereby suggesting that these two circuits arose from similar neural designs. The normalized speech items are rapidly categorized and stably remembered by Adaptive Resonance Theory circuits. Simulations use synthesized steady-state vowels from the Peterson and Barney [J. Acoust. Soc. Am. 24, 175-184 (1952)] vowel database and achieve accuracy rates similar to those achieved by human listeners. These results are compared to behavioral data and other speaker normalization models.

A Neural Network Architecture for Autonomous Learning, Recognition, and Prediction in a Nonstationary World

In a constantly changing world, humans are adapted to alternate routinely between attending to familiar objects and testing hypotheses about novel ones. We can rapidly learn to recognize and narne novel objects without unselectively disrupting our memories of familiar ones. We can notice fine details that differentiate nearly identical objects and generalize across broad classes of dissimilar objects. This chapter describes a class of self-organizing neural network architectures--called ARTMAP-- that are capable of fast, yet stable, on-line recognition learning, hypothesis testing, and naming in response to an arbitrary stream of input patterns (Carpenter, Grossberg, Markuzon, Reynolds, and Rosen, 1992; Carpenter, Grossberg, and Reynolds, 1991). The intrinsic stability of ARTMAP allows the system to learn incrementally for an unlimited period of time. System stability properties can be traced to the structure of its learned memories, which encode clusters of attended features into its recognition categories, rather than slow averages of category inputs. The level of detail in the learned attentional focus is determined moment-by-moment, depending on predictive success: an error due to over-generalization automatically focuses attention on additional input details enough of which are learned in a new recognition category so that the predictive error will not be repeated. An ARTMAP system creates an evolving map between a variable number of learned categories that compress one feature space (e.g., visual features) to learned categories of another feature space (e.g., auditory features). Input vectors can be either binary or analog. Computational properties of the networks enable them to perform significantly better in benchmark studies than alternative machine learning, genetic algorithm, or neural network models. Some of the critical problems that challenge and constrain any such autonomous learning system will next be illustrated. Design principles that work together to solve these problems are then outlined. These principles are realized in the ARTMAP architecture, which is specified as an algorithm. Finally, ARTMAP dynamics are illustrated by means of a series of benchmark simulations.

A Neural Network for Synthesizing the Pitch of an Acoustic Source

This article describes a neural network model capable of generating a spatial representation of the pitch of an acoustic source. Pitch is one of several auditory percepts used by humans to separate multiple sound sources in the environment from each other. The model provides a neural instantiation of a type of "harmonic sieve". It is capable of quantitatively simulating a large body of psychoacoustical data, including new data on octave shift perception.