Perceptual learning of tone patterns changes the effective connectivity between Heschl's gyrus and planum temporale

Abstract

Learning of complex auditory sequences such as music can be thought of as optimizing an internal model of regularities through unpredicted events (or “prediction errors”). We used dynamic causal modeling (DCM) and parametric empirical Bayes on functional magnetic resonance imaging (fMRI) data to identify modulation of effective brain connectivity that takes place during perceptual learning of complex tone patterns. Our approach differs from previous studies in two aspects. First, we used a complex oddball paradigm based on tone patterns as opposed to simple deviant tones. Second, the use of fMRI allowed us to identify cortical regions with high spatial accuracy. These regions served as empirical regions‐of‐interest for the analysis of effective connectivity. Deviant patterns induced an increased blood oxygenation level‐dependent response, compared to standards, in early auditory (Heschl's gyrus [HG]) and association auditory areas (planum temporale [PT]) bilaterally. Within this network, we found a left‐lateralized increase in feedforward connectivity from HG to PT during deviant responses and an increase in excitation within left HG. In contrast to previous findings, we did not find frontal activity, nor did we find modulations of backward connections in response to oddball sounds. Our results suggest that complex auditory prediction errors are encoded by changes in feedforward and intrinsic connections, confined to superior temporal gyrus.

Musicianship and melodic predictability enhance neural gain in auditory cortex during pitch deviance detection

“…That we found a reduction in top-down inhibition from secondary to primary auditory areas and a lack of modulation of forward connections contrasts with most previous studies in which both forward and backward connections show oddball-related effects (Auksztulewicz & Friston, 2015; Chennu et al, 2016; Garrido et al, 2007, 2008; Garrido, Kilner, Kiebel, et al, 2009; Lumaca et al, 2020; Schmidt et al, 2013). From a predictive coding perspective (Friston, 2005; Garrido, Kilner, Stephan, et al, 2009; Huang & Rao, 2011), forward and backward communication between brain areas reflects the update of predictive models by prediction error.…”

“…Another finding that differs from previous research is the lack of involvement of frontal areas in the generation of the MMN, as indicated by the low probability of the opercular family. This is consistent with the lack of frontal generators previously reported for the same dataset (Quiroga-Martinez et al, 2019a) and in a recent fMRI study using simple musical stimuli (Lumaca et al, 2020). This suggests that the opercular peak found in the present source-level statistical analyses may reflect source leakage.…”

Musicianship and melodic predictability enhance neural gain in auditory cortex during pitch deviance detection

“…That we found a reduction in top-down inhibition from secondary to primary auditory areas and a lack of modulation of forward connections contrasts with most previous studies in which both forward and backward connections show oddball-related effects (Auksztulewicz & Friston, 2015;Chennu et al, 2016;Garrido, Kilner, Kiebel, & Friston, 2007;Garrido et al, 2008;Garrido, Kilner, Kiebel, et al, 2009;Lumaca, Dietz, Hansen, Quiroga-Martinez, & Vuust, 2020;Schmidt et al, 2013). From a predictive coding perspective (Friston, 2005;Garrido, Kilner, Stephan, et al, 2009;Huang & Rao, 2011) and elicit little model updating (i.e., learning) at higher stages where melodic expectations are likely processed.…”

“…Another finding that differs from previous research is the lack of involvement of frontal areas in the generation of the MMN, as indicated by the low probability of the opercular family. This is consistent with the lack of frontal generators previously reported for the same dataset (Quiroga-Martinez et al, 2019b) and in a recent fMRI study using simple musical stimuli (Lumaca et al, 2020). This suggests that the opercular peak found in the present source-level statistical analyses may reflect source leakage.…”

Neurocomputational underpinnings of expected surprise

“…Rare but robust empirical supports for predictive coding at play during oddball processing have been reported this past decade, obtained at the psychological level with trial-by-trial computational modelling (Lieder et al, 2013; Ostwald et al, 2012; Stefanics et al, 2018; Weber et al, 2020) and at the physiological level with DCM (Chennu et al, 2016; Fogelson et al, 2014; Garrido et al, 2009; Lumaca et al, 2021; Moran et al, 2013). Novel evidence at the both levels of analysis is provided here in the controlled work using exactly the same brain data informed by simultaneous EEG and MEG.…”