The inferior colliculus (IC) receives convergent excitatory and inhibitory inputs from several auditory brainstem nuclei, and consequently displays a variety of complex response properties including frequency tuning (Davis, 2005), sensitivity to interaural differences (Yin et al., 2019), duration tuning (Casseday et al., 1994), and tuning to amplitude modulation (AM) (Langner and Schreiner, 1988; Krishna and Semple, 2000). Sensitivity of IC neurons to the velocity (speed and direction) of frequency sweeps has also been studied. Studies of bat IC responses report neurons that prefer, or even respond exclusively to, a particular sweep direction, a possible adaptation for echolocation (Suga, 1965; Fuzessery, 1994; Fuzessery and Hall, 1996; Gordon and O'Neill, 1998; Pollak et al., 2011). Additionally, sounds with statistics resembling vocal spectral-peak (formant) transitions have been used to produce IC spectrotemporal receptive fields (STRFs) in cats and bats (Escabí and Schreiner, 2002; Escabí et al., 2003; Andoni et al., 2007).

Recently, IC sensitivity to the velocity of sweeps has been explored in anesthetized gerbils (Steenken et al., 2022) and awake budgerigars (Henry et al., 2023). These studies used Schroeder-phase (SCHR) harmonic complexes, an advantageous stimulus for studying frequency-sweep sensitivity. The harmonic components of SCHR stimuli have flat magnitude spectra and curved phase spectra that create periodic acoustic waveforms with maximally flat temporal envelopes (Schroeder, 1970). The phase differences across harmonic components result in linear frequency sweeps within each fundamental period, extending from the lowest component, typically the fundamental frequency (F0) or its first harmonic, to the highest harmonic component.

As a tool for interrogating neural velocity sensitivity, the SCHR stimulus is unique in several ways. First, SCHR stimuli are notable for the speed of frequency sweeps, which travel from F0 to the maximum harmonic (here, 16 kHz) over the fundamental period (here, between 2.5 and 20 ms). The resulting velocities are faster than most commonly studied frequency-modulated signals, such as formant transitions (Liberman and Mattingly, 1989). We refer to these fast sweeps as SCHR chirps. Second, as a harmonic complex, the SCHR stimulus resembles voiced speech sounds. SCHR chirps emerge from phase differences between harmonic components. Similar phase differences are present in sounds like vowels, introduced into the glottal pulse by vocal-tract filtering. Therefore, neural sensitivity to chirp velocity would likely influence IC responses to speech sounds.

However, as harmonic stimuli, SCHR stimuli also have strong periodicity. IC neurons are notable for their tuning to the modulation frequency of amplitude-modulated (AM) sounds, often having enhanced or suppressed response rates to AM stimuli relative to an unmodulated stimulus (Krishna and Semple, 2000; Joris et al., 2004; Kim et al., 2020). It is unclear whether sensitivity to chirps requires a periodic context and whether the sensitivity is modulated by stimulus periodicity. Here, we use a new stimulus that isolated the velocity sensitivity from periodicity tuning. We recorded IC responses to both periodic SCHR stimuli and to aperiodic, random-chirp stimuli to assess velocity sensitivity. We explored the differences in prevalence of direction selectivity between the two stimuli and quantified the relative contributions of periodicity and velocity sensitivity in predicting SCHR response rates.