Representation of individual elements of a complex call sequence in primary auditory cortex

Abstract

Conspecific communication calls can be rhythmic or contain extended, discontinuous series of either constant or frequency modulated harmonic tones and noise bursts separated by brief periods of silence. In the guinea pig, rhythmic calls can produce isomorphic responses within the primary auditory cortex (AI) where single units respond to every call element. Other calls such as the chutter comprise a series of short irregular syllables that vary in their spectral content and are more like human speech. These calls can also evoke isomorphic responses, but may only do so in fields in the auditory belt and not in AI. Here we present evidence that cells in AI treat the individual elements within a syllable as separate auditory objects and respond selectively to one or a subset of them. We used a single chutter exemplar to compare single/multi-unit responses in the low-frequency portion of AI—AI(LF) and the low-frequency part of the thalamic medial geniculate body—MGB(LF) in urethane anaesthetized guinea pigs. Both thalamic and cortical cells responded with brief increases in firing rate to one, or more, of the 8 main elements present in the chutter call. Almost none of the units responded to all 8 elements. While there were many different combinations of responses to between one and five of the elements, MBG(LF) and AI(LF) neurons exhibited the same specific types of response combinations. Nearby units in the upper layers of the cortex tended to respond to similar combinations of elements while the deep layers were less responsive. Thus, the responses from a number of AI units would need to be combined in order to represent the entire chutter call. Our results don't rule out the possibility of constructive convergence but there was no evidence that a convergence of inputs within AI led to a complete representation of all eight elements.

BOLD fMRI study of ultrahigh frequency encoding in the inferior colliculus

“…This was in parallel with many electrophysiological studies that recorded responses to similar vocalizations at neuronal levels (Bauer et al, 2002;Mayko et al, 2012;Pincherli Castellanos et al, 2007;Suta et al, 2007;Wallace et al, 2005Wallace et al, , 2013. The UHF vocalizations used in this study was in the category of "50 kHz calls" (the parallel category "20 kHz calls" was not examined in this study) (Brudzynski, 2009;Portfors, 2007).…”

Neuronal coding of multiscale temporal features in communication sequences within the bat auditory cortex

“…Other studies, particularly in the guinea pig, have reported spiking responses in which the temporal structure of communication calls is well-preserved 9 , 46 , 47 . Such responses are called “isomorphic”, since the spiking and the stimulus have similar shape over time.…”

“…birds, bats, and non-human primates) emit communication calls with multiscale temporal structures. These calls, often referred to as sequences, are typically composed of individual acoustic elements like syllables, or groups of syllables, which ultimately make up whole vocalizations 4 , 9 – 11 . For example, in bats, syllabic structures have been observed in the context of distress calling 2 , 12 , 13 .…”

Neural processing of natural sounds

“…These investigations in model systems led to the discovery of cricket-song selective neurons and their contribution to the females' phonotaxis behavior [77], of call selective neurons in frogs [78] and guinea pigs [79], of song selective neurons in songbirds [7,[80][81][82], of neurons selective to the echolocation signal in bats [5], and of brain regions selective for conspecific calls in primates [83]. Selectivity for conspecific communication calls can be reflected not only in the mean rate of single neurons but also (and sometimes only) in time-varying responses [84,85] or ensemble responses [86,87]. Thus, the auditory system is not only selective to natural sounds in a broad sense but appears to also exhibit specialized circuitry for the sole purpose of detecting and processing conspecific communication calls.…”