4.1 N1 Adaptation at the Initial Stage
The pronounced amplitude decrease observed from the 1st tone to the 2nd tone in the N1 adaptation effect aligns with our hypothesis and is consistent with findings from prior studies (e.g., Bourbon et al., 1987; Budd et al., 1998; Lagemann et al., 2012; Recasens et al., 2015; Rosburg, 2004; Rosburg et al., 2006). This finding suggests that N1 adaptation occurs quickly in response to simple tone stimuli presented in rapid succession when expectations are controlled. Nevertheless, these results may not generalize to conditions with longer inter-stimulus intervals (ISI). Previous studies have shown that while the steep amplitude decrease is evident with short ISIs, a further reduction occurs with longer ISIs (e.g., Fruhstorfer et al., 1970; Özesmio et al., 2000; Sambeth et al., 2004; c.f. Bourbon et al., 1987; Budd et al., 1998; Rosburg et al., 2010). Moreover, a previous study has indicated that the primary source of the N1 response is the temporal lobe when employing a one-second ISI condition. However, with longer ISI, activity from brain regions beyond the auditory cortex becomes more prominent (Hari et al., 1982). Given that the present study employed a short ISI (580 ms), the observed steep decrease in amplitude may be ascribed to the ISI.
Additionally, a rebound was observed between the 2nd and 3rd tones in the N1 adaptation curve (Figure 2a). This phenomenon, not anticipated by the adaptation account, may be linked to the findings from a previous study utilizing dynamic causal modelling (DCM) to examine adaptation effects represented by the intrinsic connections in A1 (Garrido et al., 2009a). Based on their interpretation, the initial decrease in intrinsic connectivity from the 1st tone to the 2nd tone indicated reduced estimated precision of predictions triggered by the deviants, while the subsequent increase reflected gradual recovery due to learning. Hence, the rebound observed in the N1 adaptation effect between the 2nd and 3rd tones in our study could align with the notion proposed by Garrido et al. (2009a): the adaptation effect in A1 may incorporate a predictive component, and the brain requires only a few repetitions to generate predictions. Importantly, the present finding extends this interpretation to situations where expectations are controlled, suggesting that the predictive component in the adaptation effect can be automatically generated.