Fig. 1: Auditory omission responses consistently localize to the temporal association area.
From: The representation of omitted sounds in the mouse auditory cortex
A Widefield calcium imaging was performed from the auditory cortex in awake, head-fixed mice at a sampling rate of 100 Hz. B The fast and bright calcium indicator jGCaMP8m was expressed by injecting the viral construct in 7 locations across the auditory cortex (triangular pattern, 1 mm separation) to achieve rather uniform expression. C jGCaMP8m enables clear separation of the onset and offset responses (average ΔF/F0 over last 10 trials, top) even on a single-trial basis (bottom), here in response to brief white noise sounds (100 ms, 50 dB SPL, average over primary region, blue isoline in H). D jGCaMP8m’s fast rise time of ~2.5 ms allows a response latency map to be estimated from widefield recordings, which shows a low-latency core, composed of A1, A2, and AAF (division shown here based on tonotopy, see E). Stimulus as in (C); transparency adapted based on overall response size. E The overall division of the auditory cortex into subfields was based on the low-latency center and further subdivided based on classical tonotopic gradient reversals. This revealed both primary (AI, AII, AAF) and secondary areas (DM, DP), including the temporal association area, subdivided spatially into vTeA, mTeA, and pTeA (see “Methods” section36,37). F We presented sequences of identical sounds which were unexpectedly interrupted by omissions, i.e., a stimulus that was randomly left out. In different trials, the sounds (“standards”) were either pure tones or white noise (50 dB SPL, 75 ms, 125 ms pause). Overall, the sequence contained 1440 sounds and 200 omissions (see “Methods” section for details). G The neural activity following an omission (left) differed across the auditory cortex (right), with high activity observed in more posterior regions (referred to as the ‘omission-responsive region’ ORR, red outline) in comparison to the low-latency primary regions (PR, blue outline). The omission response (OR) showed an inflection around omission onset and remained elevated until the response to the post-omission stimulus (PR & ORR defined as adjacent pixels with high stimulus response or OR, respectively, area estimate overlaid in gray). H Across subjects, pTeA showed significantly higher OR than vTeA (p = 0.0001), DM (p < 0.0001), and AAF (p = 0.001, Kruskal–Wallis test after Bonferroni correction, N = 10, OR z-scored within subjects, error bars indicate SEM). Source data are provided with this paper.
