Neural Systems Underlying Perceptual Adjustment to Non-Standard Speech Tokens

What we do (not) know about the mechanisms underlying adaptive speech perception: A computational framework and review

Speech from unfamiliar talkers can be difficult to comprehend initially. These difficulties tend to dissipate with exposure, sometimes within minutes or less. Adaptivity in response to unfamiliar input is now considered a fundamental property of speech perception, and research over the past two decades has made substantial progress in identifying its characteristics. The mechanisms underlying adaptive speech perception, however, remain unknown. Past work has attributed facilitatory effects of exposure to any one of three qualitatively different hypothesized mechanisms: (1) low-level, pre-linguistic, signal normalization, (2) changes in/selection of linguistic representations, or (3) changes in post-perceptual decision-making. Direct comparisons of these hypotheses, or combinations thereof, have been lacking. We describe a general computational framework for adaptive speech perception (ASP) that-for the first time-implements all three mechanisms. We demonstrate how the framework can be used to derive predictions for experiments on perception from the acoustic properties of the stimuli. Using this approach, we find that-at the level of data analysis presently employed by most studies in the field-the signature results of influential experimental paradigms do not distinguish between the three mechanisms. This highlights the need for a change in research practices, so that future experiments provide more informative results. We recommend specific changes to experimental paradigms and data analysis. All data and code for this study are shared via OSF, including the R markdown document that this article is generated from, and an R library that implements the models we present.

Using TMS to evaluate a causal role for right posterior temporal cortex in talker-specific phonetic processing

Theories suggest that speech perception is informed by listeners' beliefs of what phonetic variation is typical of a talker. A previous fMRI study found right middle temporal gyrus (RMTG) sensitivity to whether a phonetic variant was typical of a talker, consistent with literature suggesting that the right hemisphere may play a key role in conditioning phonetic identity on talker information. The current work used transcranial magnetic stimulation (TMS) to test whether the RMTG plays a causal role in processing talker-specific phonetic variation. Listeners were exposed to talkers who differed in how they produced voiceless stop consonants while TMS was applied to RMTG, left MTG, or scalp vertex. Listeners subsequently showed near-ceiling performance in indicating which of two variants was typical of a trained talker, regardless of previous stimulation site. Thus, even though the RMTG is recruited for talker-specific phonetic processing, modulation of its function may have only modest consequences.

Right Posterior Temporal Cortex Supports Integration of Phonetic and Talker Information

Though the right hemisphere has been implicated in talker processing, it is thought to play a minimal role in phonetic processing, at least relative to the left hemisphere. Recent evidence suggests that the right posterior temporal cortex may support learning of phonetic variation associated with a specific talker. In the current study, listeners heard a male talker and a female talker, one of whom produced an ambiguous fricative in /s/-biased lexical contexts (e.g., epi?ode) and one who produced it in /∫/-biased contexts (e.g., friend?ip). Listeners in a behavioral experiment (Experiment 1) showed evidence of lexically guided perceptual learning, categorizing ambiguous fricatives in line with their previous experience. Listeners in an fMRI experiment (Experiment 2) showed differential phonetic categorization as a function of talker, allowing for an investigation of the neural basis of talker-specific phonetic processing, though they did not exhibit perceptual learning (likely due to characteristics of our in-scanner headphones). Searchlight analyses revealed that the patterns of activation in the right superior temporal sulcus (STS) contained information about who was talking and what phoneme they produced. We take this as evidence that talker information and phonetic information are integrated in the right STS. Functional connectivity analyses suggested that the process of conditioning phonetic identity on talker information depends on the coordinated activity of a left-lateralized phonetic processing system and a right-lateralized talker processing system. Overall, these results clarify the mechanisms through which the right hemisphere supports talker-specific phonetic processing.

Perceptual learning of multiple talkers: Determinants, characteristics, and limitations

Research suggests that listeners simultaneously update talker-specific generative models to reflect structured phonetic variation. Because past investigations exposed listeners to talkers of different genders, it is unknown whether adaptation is talker specific or rather linked to a broader sociophonetic class. Here, we test determinants of listeners' ability to update and apply talker-specific models for speech perception. In six experiments (n = 480), listeners were first exposed to the speech of two talkers who produced ambiguous fricative energy. The talkers' speech was interleaved during exposure, and lexical context differentially biased interpretation of the ambiguity as either /s/ or /ʃ/ for each talker. At test, listeners categorized tokens from ashi-asi continua, one for each talker. Across conditions and experiments, we manipulated exposure quantity, talker gender, blocked versus interleaved talker structure at test, and the degree to which fricative acoustics differed between talkers. When test was blocked by talker, learning was observed for different but not same gender talkers. When talkers were interleaved at test, learning was observed for both different and same gender talkers, which was attenuated when fricative acoustics were constant across talkers. There was no strong evidence to suggest that adaptation to multiple talkers required increased quantity of exposure beyond that required to adapt to a single talker. These results suggest that perceptual learning for speech is achieved via a mechanism that represents a context-dependent, cumulative integration of experience with speech input and identity critical constraints on listeners' ability to dynamically apply multiple generative models in mixed talker listening environments.

Listeners are initially flexible in updating phonetic beliefs over time

Perceptual learning serves as a mechanism for listeners to adapt to novel phonetic information. Distributional tracking theories posit that this adaptation occurs as a result of listeners accumulating talker-specific distributional information about the phonetic category in question (Kleinschmidt & Jaeger, Psychological Review, 122, 148-203, 2015). What is not known is how listeners build these talker-specific distributions-that is, if they aggregate all information received over a certain time period, or if they rely more heavily upon the most recent information received and down-weight older, consolidated information. In the present experiment, listeners were exposed to four interleaved blocks of a lexical decision task and a phonetic categorization task in which the lexical decision blocks were designed to bias perception in opposite directions of a "s"-"sh" contrast. Listeners returned several days later and completed the identical task again. In each individual session, listener's perception of a "s"-"sh" contrast was biased by the information in the immediately preceding lexical decision block (though only when participants heard the "sh"-biasing block first, which was likely driven by stimulus characteristics). There was evidence that listeners accrued information about the talker over time since the bias effect diminished in the second session. In general, results suggest that listeners initially maintain some flexibility with their talker-specific phonetic representations, but over the course of several exposures begin to consolidate these representations.

Structural neural correlates of individual differences in categorical perception

Listeners perceive speech sounds categorically. While group-level differences in categorical perception have been observed in children or individuals with reading disorders, recent findings suggest that typical adults vary in how categorically they perceive sounds. The current study investigated neural sources of individual variability in categorical perception of speech. Fifty-seven participants rated phonetic tokens on a visual analogue scale; categoricity and response consistency were measured and related to measures of brain structure from MRI. Increased surface area of the right middle frontal gyrus predicted more categorical perception of a fricative continuum. This finding supports the idea that frontal regions are sensitive to phonetic category-level information and extends it to make behavioral predictions at the individual level. Additionally, more gyrification in bilateral transverse temporal gyri predicted less consistent responses on the task, perhaps reflecting subtle variation in language ability across the population.

Individual Differences in Lexical Contributions to Speech Perception

Purpose The extant literature suggests that individual differences in speech perception can be linked to broad receptive language phenotype. For example, a recent study found that individuals with a smaller receptive vocabulary showed diminished lexically guided perceptual learning compared to individuals with a larger receptive vocabulary. Here, we examined (a) whether such individual differences stem from variation in reliance on lexical information or variation in perceptual learning itself and (b) whether a relationship exists between lexical recruitment and lexically guided perceptual learning more broadly, as predicted by current models of lexically guided perceptual learning. Method In Experiment 1, adult participants (n = 70) completed measures of receptive and expressive language ability, lexical recruitment, and lexically guided perceptual learning. In Experiment 2, adult participants (n = 120) completed the same lexical recruitment and lexically guided perceptual learning tasks to provide a high-powered replication of the primary findings from Experiment 1. Results In Experiment 1, individuals with weaker receptive language ability showed increased lexical recruitment relative to individuals with higher receptive language ability; however, receptive language ability did not predict the magnitude of lexically guided perceptual learning. Moreover, the results of both experiments converged to show no evidence indicating a relationship between lexical recruitment and lexically guided perceptual learning. Conclusion The current findings suggest that (a) individuals with weaker language ability demonstrate increased reliance on lexical information for speech perception compared to those with stronger receptive language ability; (b) individuals with weaker language ability maintain an intact perceptual learning mechanism; and, (c) to the degree that the measures used here accurately capture individual differences in lexical recruitment and lexically guided perceptual learning, there is no graded relationship between these two constructs.

The Role of the Right Hemisphere in Processing Phonetic Variability Between Talkers

Neurobiological models of speech perception posit that both left and right posterior temporal brain regions are involved in the early auditory analysis of speech sounds. However, frank deficits in speech perception are not readily observed in individuals with right hemisphere damage. Instead, damage to the right hemisphere is often associated with impairments in vocal identity processing. Herein lies an apparent paradox: The mapping between acoustics and speech sound categories can vary substantially across talkers, so why might right hemisphere damage selectively impair vocal identity processing without obvious effects on speech perception? In this review, I attempt to clarify the role of the right hemisphere in speech perception through a careful consideration of its role in processing vocal identity. I review evidence showing that right posterior superior temporal, right anterior superior temporal, and right inferior / middle frontal regions all play distinct roles in vocal identity processing. In considering the implications of these findings for neurobiological accounts of speech perception, I argue that the recruitment of right posterior superior temporal cortex during speech perception may specifically reflect the process of conditioning phonetic identity on talker information. I suggest that the relative lack of involvement of other right hemisphere regions in speech perception may be because speech perception does not necessarily place a high burden on talker processing systems, and I argue that the extant literature hints at potential subclinical impairments in the speech perception abilities of individuals with right hemisphere damage.

Lexical Information Guides Retuning of Neural Patterns in Perceptual Learning for Speech

A listener's interpretation of a given speech sound can vary probabilistically from moment to moment. Previous experience (i.e., the contexts in which one has encountered an ambiguous sound) can further influence the interpretation of speech, a phenomenon known as perceptual learning for speech. This study used multivoxel pattern analysis to query how neural patterns reflect perceptual learning, leveraging archival fMRI data from a lexically guided perceptual learning study conducted by Myers and Mesite [Myers, E. B., & Mesite, L. M. Neural systems underlying perceptual adjustment to non-standard speech tokens. Journal of Memory and Language, 76, 80-93, 2014]. In that study, participants first heard ambiguous /s/-/∫/ blends in either /s/-biased lexical contexts (epi_ode) or /∫/-biased contexts (refre_ing); subsequently, they performed a phonetic categorization task on tokens from an /asi/-/a∫i/ continuum. In the current work, a classifier was trained to distinguish between phonetic categorization trials in which participants heard unambiguous productions of /s/ and those in which they heard unambiguous productions of /∫/. The classifier was able to generalize this training to ambiguous tokens from the middle of the continuum on the basis of individual participants' trial-by-trial perception. We take these findings as evidence that perceptual learning for speech involves neural recalibration, such that the pattern of activation approximates the perceived category. Exploratory analyses showed that left parietal regions (supramarginal and angular gyri) and right temporal regions (superior, middle, and transverse temporal gyri) were most informative for categorization. Overall, our results inform an understanding of how moment-to-moment variability in speech perception is encoded in the brain.

Neural Representation of Articulable and Inarticulable Novel Sound Contrasts: The Role of the Dorsal Stream

The extent that articulatory information embedded in incoming speech contributes to the formation of new perceptual categories for speech sounds has been a matter of discourse for decades. It has been theorized that the acquisition of new speech sound categories requires a network of sensory and speech motor cortical areas (the "dorsal stream") to successfully integrate auditory and articulatory information. However, it is possible that these brain regions are not sensitive specifically to articulatory information, but instead are sensitive to the abstract phonological categories being learned. We tested this hypothesis by training participants over the course of several days on an articulable non-native speech contrast and acoustically matched inarticulable nonspeech analogues. After reaching comparable levels of proficiency with the two sets of stimuli, activation was measured in fMRI as participants passively listened to both sound types. Decoding of category membership for the articulable speech contrast alone revealed a series of left and right hemisphere regions outside of the dorsal stream that have previously been implicated in the emergence of non-native speech sound categories, while no regions could successfully decode the inarticulable nonspeech contrast. Although activation patterns in the left inferior frontal gyrus, the middle temporal gyrus, and the supplementary motor area provided better information for decoding articulable (speech) sounds compared to the inarticulable (sine wave) sounds, the finding that dorsal stream regions do not emerge as good decoders of the articulable contrast alone suggests that other factors, including the strength and structure of the emerging speech categories are more likely drivers of dorsal stream activation for novel sound learning.

Neural Correlates of Phonetic Adaptation as Induced by Lexical and Audiovisual Context

When speech perception is difficult, one way listeners adjust is by reconfiguring phoneme category boundaries, drawing on contextual information. Both lexical knowledge and lipreading cues are used in this way, but it remains unknown whether these two differing forms of perceptual learning are similar at a neural level. This study compared phoneme boundary adjustments driven by lexical or audiovisual cues, using ultra-high-field 7-T fMRI. During imaging, participants heard exposure stimuli and test stimuli. Exposure stimuli for lexical retuning were audio recordings of words, and those for audiovisual recalibration were audio-video recordings of lip movements during utterances of pseudowords. Test stimuli were ambiguous phonetic strings presented without context, and listeners reported what phoneme they heard. Reports reflected phoneme biases in preceding exposure blocks (e.g., more reported /p/ after /p/-biased exposure). Analysis of corresponding brain responses indicated that both forms of cue use were associated with a network of activity across the temporal cortex, plus parietal, insula, and motor areas. Audiovisual recalibration also elicited significant occipital cortex activity despite the lack of visual stimuli. Activity levels in several ROIs also covaried with strength of audiovisual recalibration, with greater activity accompanying larger recalibration shifts. Similar activation patterns appeared for lexical retuning, but here, no significant ROIs were identified. Audiovisual and lexical forms of perceptual learning thus induce largely similar brain response patterns. However, audiovisual recalibration involves additional visual cortex contributions, suggesting that previously acquired visual information (on lip movements) is retrieved and deployed to disambiguate auditory perception.

Integrating prediction errors at two time scales permits rapid recalibration of speech sound categories

Speech perception presumably arises from internal models of how specific sensory features are associated with speech sounds. These features change constantly (e.g. different speakers, articulation modes etc.), and listeners need to recalibrate their internal models by appropriately weighing new versus old evidence. Models of speech recalibration classically ignore this volatility. The effect of volatility in tasks where sensory cues were associated with arbitrary experimenter-defined categories were well described by models that continuously adapt the learning rate while keeping a single representation of the category. Using neurocomputational modelling we show that recalibration of natural speech sound categories is better described by representing the latter at different time scales. We illustrate our proposal by modeling fast recalibration of speech sounds after experiencing the McGurk effect. We propose that working representations of speech categories are driven both by their current environment and their long-term memory representations.

People can distinguish words or syllables even though they may sound different with every speaker. This striking ability reflects the fact that our brain is continually modifying the way we recognise and interpret the spoken word based on what we have heard before, by comparing past experience with the most recent one to update expectations. This phenomenon also occurs in the McGurk effect: an auditory illusion in which someone hears one syllable but sees a person saying another syllable and ends up perceiving a third distinct sound.

Abstract models, which provide a functional rather than a mechanistic description of what the brain does, can test how humans use expectations and prior knowledge to interpret the information delivered by the senses at any given moment. Olasagasti and Giraud have now built an abstract model of how brains recalibrate perception of natural speech sounds. By fitting the model with existing experimental data using the McGurk effect, the results suggest that, rather than using a single sound representation that is adjusted with each sensory experience, the brain recalibrates sounds at two different timescales.

Over and above slow “procedural” learning, the findings show that there is also rapid recalibration of how different sounds are interpreted. This working representation of speech enables adaptation to changing or noisy environments and illustrates that the process is far more dynamic and flexible than previously thought.

Lexically guided perceptual learning is robust to task-based changes in listening strategy

Listeners use lexical information to resolve ambiguity in the speech signal, resulting in the restructuring of speech sound categories. Recent findings suggest that lexically guided perceptual learning is attenuated when listeners use a perception-focused listening strategy (that directs attention towards surface variation) compared to when listeners use a comprehension-focused listening strategy (that directs attention towards higher-level linguistic information). However, previous investigations used the word position of the ambiguity to manipulate listening strategy, raising the possibility that attenuated learning reflected decreased strength of lexical recruitment instead of a perception-oriented listening strategy. The current work tests this hypothesis. Listeners completed an exposure phase followed by a test phase. During exposure, listeners heard an ambiguous fricative embedded in word-medial lexical contexts that supported realization of the ambiguity as /∫/. At test, listeners categorized members of an /ɑsi/-/ɑ∫i/ continuum. Listening strategy was manipulated via exposure task (experiment 1) and explicit acknowledgement of the ambiguity (experiment 2). Compared to control participants, listeners who were exposed to the ambiguity showed more /∫/ responses at the test; critically, the magnitude of learning did not differ across listening strategy conditions. These results suggest that given sufficient lexical context, lexically guided perceptual learning is robust to task-based changes in listening strategy.

Listeners are maximally flexible in updating phonetic beliefs over time

Perceptual learning serves as a mechanism for listenexrs to adapt to novel phonetic information. Distributional tracking theories posit that this adaptation occurs as a result of listeners accumulating talker-specific distributional information about the phonetic category in question (Kleinschmidt & Jaeger, 2015, Psychological Review, 122). What is not known is how listeners build these talker-specific distributions; that is, if they aggregate all information received over a certain time period, or if they rely more heavily upon the most recent information received and down-weight older, consolidated information. In the present experiment, listeners were exposed to four interleaved blocks of a lexical decision task and a phonetic categorization task in which the lexical decision blocks were designed to bias perception in opposite directions along a "s"-"sh" continuum. Listeners returned several days later and completed the identical task again. Evidence was consistent with listeners using a relatively short temporal window of integration at the individual session level. Namely, in each individual session, listeners' perception of a "s"-"sh" contrast was biased by the information in the immediately preceding lexical decision block, and there was no evidence that listeners summed their experience with the talker over the entire session. Similarly, the magnitude of the bias effect did not change between sessions, consistent with the idea that talker-specific information remains flexible, even after consolidation. In general, results suggest that listeners are maximally flexible when considering how to categorize speech from a novel talker.

Voice-sensitive brain networks encode talker-specific phonetic detail

The speech stream simultaneously carries information about talker identity and linguistic content, and the same acoustic property (e.g., voice-onset-time, or VOT) may be used for both purposes. Separable neural networks for processing talker identity and phonetic content have been identified, but it is unclear how a singular acoustic property is parsed by the neural system for talker identification versus phonetic processing. In the current study, listeners were exposed to two talkers with characteristically different VOTs. Subsequently, brain activation was measured using fMRI as listeners performed a phonetic categorization task on these stimuli. Right temporoparietal regions previously implicated in talker identification showed sensitivity to the match between VOT variant and talker, whereas left posterior temporal regions showed sensitivity to the typicality of phonetic exemplars, regardless of talker typicality. Taken together, these results suggest that neural systems for voice recognition capture talker-specific phonetic variation.

Phonetic recalibration of speech by text

Listeners adjust their phonetic categories to cope with variations in the speech signal (phonetic recalibration). Previous studies have shown that lipread speech (and word knowledge) can adjust the perception of ambiguous speech and can induce phonetic adjustments (Bertelson, Vroomen, & de Gelder in Psychological Science, 14(6), 592–597, 2003; Norris, McQueen, & Cutler in Cognitive Psychology, 47(2), 204–238, 2003). We examined whether orthographic information (text) also can induce phonetic recalibration. Experiment 1 showed that after exposure to ambiguous speech sounds halfway between /b/ and /d/ that were combined with text (b or d) participants were more likely to categorize auditory-only test sounds in accordance with the exposed letters. Experiment 2 replicated this effect with a very short exposure phase. These results show that listeners adjust their phonetic boundaries in accordance with disambiguating orthographic information and that these adjustments show a rapid build-up.

Lexically guided perceptual tuning of internal phonetic category structure

Listeners use lexical information to retune the mapping between the acoustic signal and speech sound representations, resulting in changes to phonetic category boundaries. Other research shows that phonetic categories have a rich internal structure; within-category variation is represented in a graded fashion. The current work examined whether lexically informed perceptual learning promotes a comprehensive reorganization of internal category structure. The results showed a reorganization of internal structure for one but not both of the examined categories, which may reflect an attenuation of learning for distributions with extensive category overlap. This finding points towards potential input-driven constraints on lexically guided phonetic retuning.

Perceptual learning of degraded speech by minimizing prediction error

Human perception is shaped by past experience on multiple timescales. Sudden and dramatic changes in perception occur when prior knowledge or expectations match stimulus content. These immediate effects contrast with the longer-term, more gradual improvements that are characteristic of perceptual learning. Despite extensive investigation of these two experience-dependent phenomena, there is considerable debate about whether they result from common or dissociable neural mechanisms. Here we test single- and dual-mechanism accounts of experience-dependent changes in perception using concurrent magnetoencephalographic and EEG recordings of neural responses evoked by degraded speech. When speech clarity was enhanced by prior knowledge obtained from matching text, we observed reduced neural activity in a peri-auditory region of the superior temporal gyrus (STG). Critically, longer-term improvements in the accuracy of speech recognition following perceptual learning resulted in reduced activity in a nearly identical STG region. Moreover, short-term neural changes caused by prior knowledge and longer-term neural changes arising from perceptual learning were correlated across subjects with the magnitude of learning-induced changes in recognition accuracy. These experience-dependent effects on neural processing could be dissociated from the neural effect of hearing physically clearer speech, which similarly enhanced perception but increased rather than decreased STG responses. Hence, the observed neural effects of prior knowledge and perceptual learning cannot be attributed to epiphenomenal changes in listening effort that accompany enhanced perception. Instead, our results support a predictive coding account of speech perception; computational simulations show how a single mechanism, minimization of prediction error, can drive immediate perceptual effects of prior knowledge and longer-term perceptual learning of degraded speech.

Talker-specific influences on phonetic category structure

A primary goal for models of speech perception is to describe how listeners achieve reliable comprehension given a lack of invariance between the acoustic signal and individual speech sounds. For example, individual talkers differ in how they implement phonetic properties of speech. Research suggests that listeners attain perceptual constancy by processing acoustic variation categorically while maintaining graded internal category structure. Moreover, listeners will use lexical information to modify category boundaries to learn to interpret a talker's ambiguous productions. The current work examines perceptual learning for talker differences that signal well-defined, unambiguous category members. Speech synthesis techniques were used to differentially manipulate talkers' characteristic productions of the stop voicing contrast for two groups of listeners. Following exposure to the talkers, internal category structure and category boundary were examined. The results showed that listeners dynamically adjusted internal category structure to be centered on experience with the talker's voice, but the category boundary remained fixed. These patterns were observed for words presented during training as well as novel lexical items. These findings point to input-driven constraints on functional plasticity within the language architecture, which may help to explain how listeners maintain stability of linguistic knowledge while simultaneously demonstrating flexibility for phonetic representations.