Figure 1.
Control sequence designs used in the current investigation.
(A) Both Study 1 and Study 2 used a flip-flop design for oddball sequences. This design allows for the comparison of the response to a stimulus when it is a rare, unexpected deviant to the same tone when it is a common, expected standard, controlling for differences in responses to the physical characteristics of the stimuli, but not for differential adaptation. (B) The many-standards control sequence was used in Study 2 to test responses to stimuli that would have prompted the same level of adaptation as the deviant stimulus. Comparing the responses to the deviant used in the oddball sequence, where it defies established stimulus regularity predictions, to the control in the many-standards sequence, which does not defy regularity predictions (there is no established regularity) yields a measure of adaptation-independent deviance detection. Comparing the responses to the control (presented rarely) in the many-standards sequence to the standard (presented often) in the oddball sequence yields a measure of adaptation to a rare stimulus vs. a common stimulus. (C) The cascade sequence designs were used for Study 3. The oddball sequences are similar to those used in Study 1 and Study 2, except that a flip-flop design was not used, and the stimuli presented in the ascending and descending sequences are respectively on the upper and lower end of the range of sequences used in the control sequence. (D) The control sequence (like the many-standards sequence) presents the stimuli used as deviants in the oddball sequences at the same probability as they are when deviants in a context where they do not defy established patterns in regularity, thus enabling the control of differential adaptation.
Figure 2.
(A,B) ERPs to the oddball deviant (red) and the standard (blue) for the low (A) and high (B) frequency stimuli. All stimuli show a similar pattern with the same components (N22, P37, N80): responses to deviants are larger in amplitude in comparison to standards. (C,D) Mean amplitudes (± standard error, SE) of N22, P37 and N80 generated by oddball deviants (red) and standards (blue), showing that responses to the deviant compared to the standard were larger for the N80 component (P = 0.001), and the N22 (although not significantly, P = 0.087).
Figure 3.
Rat ERPs in the many-standards control sequence (Study 2).
ERPs recorded from electrodes implanted into the skull above the left frontal cortex (A), right frontal cortex (B), left auditory cortex (C), right auditory cortex (D), midline (E), and averaged over all of the regions (F).
Figure 4.
Rat ERPs in Study 2, averaged over all regions.
(A,B) ERPs (averaged over all five regions) to the oddball deviant (red), the many-standards control (black) and the standard (blue) for the low (A) and high (B) frequency stimuli. All stimuli show a similar pattern with the same components (P13, N18, P30, N55 and N85): responses to deviants are larger in amplitude in comparison to the controls for high frequency, but not low frequency stimuli for P13 (Pb = 0.023, d = 0.73) and N55 (Pb = 0.010, d = 0.87). (C,D) Mean amplitudes (± standard error, SE) of P13, N18, P30, N55 and N85 generated by oddball deviants (red), many-standards controls (grey) and standards (blue), averaged over all five regions. Significance levels for statistical comparisons between stimulus types for each component are shown above the bars for their respective components. Asterisks indicate statistical significance under 0.05, with * 0.050<P>0.010, ** 0.010<P>0.001, *** P<0.001.
Figure 5.
Rat ERPs in Study 3, averaged over all regions.
(A,B) ERPs (averaged over all five regions) to the oddball deviant (red) and the cascade control (black) for the low (A) and high (B) frequency stimuli. As in Study 2, all stimuli show a similar pattern with the same components (P13, N18, P30, N55 and N85), however in this case, responses to deviants were not larger in amplitude in comparison to the controls and the standards. (C,D) Mean amplitudes (± standard error, SE) of P13, N18, P30, N55 and N85 generated by oddball deviants (red) and the cascade controls (grey), averaged over all five regions.
Table 1.
Oddball effects, Deviance detection and adaptation for each frequencies and component combination in Study 2.