Does signal reduction imply predictive coding in models of spoken word recognition?

Luthra, Sahil; Li, Monica Y.C.; You, Heejo; Brodbeck, Christian; Magnuson, James S.

doi:10.3758/s13423-021-01924-x

Cited by 12 publications

(10 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, TRACE is proving surprisingly capable of simulating neural level data, including crucial patterns of neural activity hypothesized to be linked to predictive coding. Luthra, Li, You, Brodbeck, and Magnuson (2021) demonstrate that TRACE captures a key pattern observed by Gagnepain, Henson, and Davis (2012) in fMRI: signal reduction when inputs conform to expectations. Brodbeck, Luthra, Gaston, and Magnuson (2021) demonstrate that TRACE captures a complex cross-over interaction observed in representational similarity analyses (Kriegeskorte & Bandettini, 2008) of fMRI (Blank & Davis, 2016) and MEG (Sohoglu & Davis, 2020) data (and demonstrate how this depends crucially on feedback), and also document interesting similarities between MEG studies of speech processing and the sensitivity of different levels of representation in TRACE to information theoretic measures such as surprisal and entropy (and how those relate to specific aspects of TRACE's architecture, such as lateral inhibition and feedback).…”

Section: Discussionmentioning

confidence: 61%

See 1 more Smart Citation

Contra assertions, feedback improves word recognition

Magnuson¹,

Grubb²,

Crinnion³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Norris and Cutler (in press) revisit their arguments that (lexical-to-sublexical) feedback cannot improve word recognition performance, based on the assumption that feedback must boost signal and noise equally. They also argue that demonstrations that feedback improves performance (Magnuson, Mirman, Luthra, Strauss, & Harris, 2018) in the TRACE model of spoken word recognition (McClelland & Elman, 1986) were artifacts of converting activations to response probabilities. We first evaluate their claim that feedback in an interactive activation model must boost noise and signal equally. This is not true in a fully interactive activation model such as TRACE, where the feedback signal does not simply mirror the feedforward signal; it is instead shaped by joint probabilities over lexical patterns, and the dynamics of lateral inhibition. Thus, even under high levels of noise, lexical feedback will selectively boost signal more than noise. We demonstrate that feedback promotes faster word recognition and preserves accuracy under noise whether one uses raw activations or response probabilities. We then document that lexical feedback selectively boosts signal (i.e., lexically-coherent series of phonemes) more than noise by tracking sublexical (phoneme) activations under noise with and without feedback. Thus, feedback in a model like TRACE does improve word recognition, exactly by selective reinforcement of lexically-coherent signal. We conclude that whether lexical feedback is integral to human speech processing is an empirical question, and briefly review a growing body of work at behavioral and neural levels that is consistent with feedback and inconsistent with autonomous (non-feedback) architectures.

show abstract

Section: Discussionmentioning

confidence: 61%

“…4 As such, it is also poised to bridge perceptual, cognitive, and neural studies of language processing. It is obviously useful for researchers interested in making those connections (as we discussed above, TRACE is being used productively to investigate neural data: Brodbeck et al, 2021;Luthra, Li, et al, 2021).…”

Section: Discussionmentioning

confidence: 99%

Contra assertions, feedback improves word recognition

Magnuson¹,

Grubb²,

Crinnion³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…To speculate, predictive coding models may account for activity in the N1 in previously tested paradigms without accurately describing the underlying neural processes. For instance, Luthra et al (2021) showed that, in spoken word recognition, interactive activation models may provide an alternative account of the ERP amplitude reduction observed in response to prediction violations, without invoking key features of predictive coding models. Indeed, effects indicative of predictive processing may emerge in a system that that lacks any representations of, or mechanisms implementing, predictions or prediction errors, instead only implementing "pattern completion" (Falandays et al, 2021).…”

Section: Discussionmentioning

confidence: 99%

Can Prediction Error Explain Predictability Effects on the N1 during Picture-Word Verification?

Taylor,

Rousselet,

Sereno

2023

Preprint

View full text Add to dashboard Cite

Do early effects of predictability in visual word recognition reflect prediction error? Electrophysiological research investigating word processing has demonstrated predictability effects in the N1, or first negative component of the event-related potential (ERP). However, findings regarding the magnitude of effects and potential interactions of predictability with lexical variables have been inconsistent. Moreover, past studies have typically used categorical designs with relatively small samples and relied on by-participant analyses. Nevertheless, reports have generally shown that predicted words elicit less negative-going (i.e., lower amplitude) N1s, a pattern consistent with a simple predictive coding account. In our preregistered study, we tested this account via the interaction between prediction magnitude and certainty. A picture-word verification paradigm was implemented in which pictures were followed by tightly matched picture-congruent or picture-incongruent written nouns. The predictability of target (picture-congruent) nouns was manipulated continuously based on norms of association between a picture and its name. ERPs from 68 participants revealed a pattern of effects opposite to that expected under a simple predictive coding framework.

show abstract

“…This response reduction is compatible with either a prediction error account (through the suppression of responses tuned to input features that are predicted) or a sharpened signal mechanism (through the suppression of responses tuned away from the input features; for a more detailed explanation, see Introduction and refs. [5,6,19]).…”

Section: Prediction Error Computations During Natural Listeningmentioning

confidence: 99%

“…(19 in the MEG experiment, 21 in the fMRI experiment) were Cambridge Psychology Research Ethics Committee. All were right-handed, native speakers of English, aged between 18 and 40 years and had no self-reported history of hearing impairment or neurological disease.…”

mentioning

confidence: 99%

Convergent neural signatures of speech prediction error are a biological marker for spoken word recognition

Sohoglu,

Beckers,

Davis

2023

Preprint

View full text Add to dashboard Cite

We used MEG and fMRI to determine how predictions are combined with speech input in superior temporal cortex. We compared neural responses to bisyllabic spoken words in which the first syllable strongly or weakly predicts the form of the second syllable (e.g. “bingo” versus “tango”). We further compared neural responses to the same second syllables when heard in an unfamiliar pseudoword and therefore in a situation in which predictions mismatch with sensory input (e.g. “snigo” and “meago”). Across multiple imaging modalities and analysis approaches (including information-based pattern analysis methods), we show that neural representations of second syllables are suppressed by strong predictions when predictions match sensory input. However, neural representations of the same second syllables show the opposite effect (i.e. enhanced representations following strongly than weakly-predicting syllables) when predictions mismatch with sensory input. Computational simulations show that this interaction between prediction strength and (mis)match is consistent with prediction error but not alternative (sharpened signal) computations. Neural signatures of prediction error are observed early (beginning 200 ms after the onset of the second syllable), localise to early auditory regions (in fMRI, bilateral Heschl’s gyrus and STG) and are expressed as changes in low-frequency (theta and alpha) power. Finally, we show that even a non-speech noise that replaces a predicted second syllable can evoke informative neural activity patterns, consistent with negative prediction errors. Our study therefore provides convergent neural evidence that speech perception is supported by the computation of prediction errors in auditory brain regions. These prediction error computations play a central role in the identification of familiar spoken words and perception of unfamiliar pseudowords.

show abstract

Does signal reduction imply predictive coding in models of spoken word recognition?

Cited by 12 publications

References 40 publications

Contra assertions, feedback improves word recognition

Contra assertions, feedback improves word recognition

Can Prediction Error Explain Predictability Effects on the N1 during Picture-Word Verification?

Convergent neural signatures of speech prediction error are a biological marker for spoken word recognition

Contact Info

Product

Resources

About