14Primary cortical areas contain maps of sensory features, including sound frequency in primary 15 auditory cortex (A1). Two-photon calcium imaging in mice has confirmed the presence of these large-16 scale maps, while uncovering an unexpected local variability in the stimulus preferences of individual 17 neurons in A1 and other primary regions. Here we show that fractured tonotopy is not unique to 18 rodents. Using two-photon imaging, we found that local variance in frequency preferences is 19 equivalent in ferrets and mice. Much of this heterogeneity was due to neurons with complex 20 frequency tuning, which are less spatially organized than those tuned to a single frequency. Finally, 21we show that microelectrode recordings may describe a smoother tonotopic arrangement due to a bias 22 towards neurons with simple frequency tuning. These results show that local variability in the 23 tonotopic map is not restricted to rodents and help explain inconsistencies in cortical topography 24 across species and recording techniques. 25 26 27 restricted to single neuron recordings, the spacing between neurons is typically ~100 µm or greater, so 56 local variations in stimulus preferences cannot be examined 24 . Second, multielectrode recordings may 57 be biased more towards the most robustly responding neurons in granular layers that receive direct 58 thalamic input, whereas most two-photon imaging studies have been restricted to the superficial layers 59 of the cortex 14,25 . However, Tischbirek et al. report similar tonotopic organization across all layers of 60 mouse auditory cortex 26 . Third, the variations in tonotopy across studies may partly reflect a species 61 difference in the cortical organization of rodents and higher mammals, particularly as local thalamic 62 inputs to A1 in mice are also heterogeneous in their frequency tuning 27 . Two-photon studies of 63 primary visual cortex have revealed poorer spatial organization of orientation tuning in rodents than in 64 cats 23,28 , tree shrews 29 , and ferrets 30 . The same may be true of tonotopy in A1. This view is further 65 supported by a recent study in marmosets, in which A1 was reported to be more tonotopically 66 organized than in rats 31 . 67 68 It is not clear how spatial organization of tuning to a single sound frequency coincides with the known 69 functions and complex frequency receptive fields of auditory cortex. Throughout the ascending 70 auditory pathway, neurons progressively integrate spectral and temporal features of sound 32 , and the 71 receptive fields of many A1 neurons are poorly predicted by a model of linear tuning to a single sound 72 frequency 33,34 . Recent studies have shown that the preferred frequencies of weakly tuned A1 neurons 73 are poorly mapped 20,26 . An outstanding question we address here is how neurons with reliable tuning 74 to multiple frequency peaks impact on tonotopic organization. 75
76The present experiments also examine whether local heterogeneity in tonotopy is a general feature of 77 mammalian A1, or a peculia...