The Neuronal Basis of Predictive Coding Along the Auditory Pathway: From the Subcortical Roots to Cortical Deviance Detection

Abstract

In this review, we attempt to integrate the empirical evidence regarding stimulus-specific adaptation (SSA) and mismatch negativity (MMN) under a predictive coding perspective (also known as Bayesian or hierarchical-inference model). We propose a renewed methodology for SSA study, which enables a further decomposition of deviance detection into repetition suppression and prediction error, thanks to the use of two controls previously introduced in MMN research: the many-standards and the cascade sequences. Focusing on data obtained with cellular recordings, we explain how deviance detection and prediction error are generated throughout hierarchical levels of processing, following two vectors of increasing computational complexity and abstraction along the auditory neuraxis: from subcortical toward cortical stations and from lemniscal toward nonlemniscal divisions. Then, we delve into the particular characteristics and contributions of subcortical and cortical structures to this generative mechanism of hierarchical inference, analyzing what is known about the role of neuromodulation and local microcircuitry in the emergence of mismatch signals. Finally, we describe how SSA and MMN are occurring at similar time frame and cortical locations, and both are affected by the manipulation of N-methyl-D-aspartate receptors. We conclude that there is enough empirical evidence to consider SSA and MMN, respectively, as the microscopic and macroscopic manifestations of the same physiological mechanism of deviance detection in the auditory cortex. Hence, the development of a common theoretical framework for SSA and MMN is all the more recommendable for future studies. In this regard, we suggest a shared nomenclature based on the predictive coding interpretation of deviance detection.

Dopamine modulates prediction error forwarding in the nonlemniscal inferior colliculus

“…Therefore, together with the oddball paradigm, we implemented the cascade sequence [19] in a subset of experiments. The CAS allows to assess which proportions of SSA (measured as iMM) can be attributed to repetition suppression of the STD response (measured as iRS) or to pure prediction error signaling activity (measured as iPE) if any [18]. Thereby we could determine how these two functional components were differentially affected by dopamine.…”

“…Therefore, the difference between the CAS and the STD response must be due to repetition suppression. Once repetition suppression has been accounted for, we considered the remaining response in SSA (whether present) as an extra component that could only correspond to prediction error signaling [18]. The cascade sequence thereby allows to dissociate SSA in 2 components: repetition suppression and prediction error (Fig.…”

“…Thus, we used the cascade sequence to control for the repetition effects induced by the oddball paradigm (for extended rationale, see [18,19]) and decompose SSA into prediction error and repetition suppression components. To allow comparison between responses from different neurons, we normalized the spike count evoked by each tone in DEV, STD and CAS as follows: …”

“…According to this interpretation, the oddball paradigm induces two processes: (1) repetition suppression during STD, and (2) prediction error signaling during DEV [18]. A way to estimate the relative contribution of these two components to the total amount of SSA is to compare DEV and STD responses to those elicited by the cascade sequence (Fig.…”

Fronto-temporal coupling dynamics during spontaneous activity and auditory processing

“…That is, the rapid non-lemniscal pathway could relay faithful information into FAF about the auditory 441 stimuli, which can in turn be compared with information received from primary AC and the 442 thalamus. Alternatively, a further proposition would be that prediction errors, relayed "upwards" 443 along the auditory hierarchy (Carbajal and Malmierca, 2018), and generated in the SGN and the AC, predictive coding could be thoroughly tested in future experimental work. 450…”