Cineradiography of monkey lip-smacking reveals putative precursors of speech dynamics

Abstract

A key feature of speech is its stereotypical 5 Hz rhythm. One theory posits that this rhythm evolved through the modification of rhythmic facial movements in ancestral primates. If the hypothesis has any validity, then a comparative approach may shed some light. We tested this idea by using cineradiography (X-ray movies) to characterize and quantify the internal dynamics of the macaque monkey vocal tract during lip-smacking (a rhythmic facial expression) versus chewing. Previous human studies showed that speech movements are faster than chewing movements, and the functional coordination between vocal tract structures is different between the two behaviors. If rhythmic speech evolved through a rhythmic ancestral facial movement, then one hypothesis is that monkey lip-smacking versus chewing should also exhibit these differences. We found that the lips, tongue, and hyoid move with a speech-like 5 Hz rhythm during lip-smacking, but not during chewing. Most importantly, the functional coordination between these structures was distinct for each behavior. These data provide empirical support for the idea that the human speech rhythm evolved from the rhythmic facial expressions of ancestral primates.

Figure 1

The Anatomy of the Macaque Monkey Vocal Tract as Imaged with Cineradiography (A) Shown are identifying marks for the key vocal tract structures, including the lips, tongue, and hyoid. (B) Shown is a schematic of the vocal tract elements and the reference points for measurements.

Figure 2

Relative Spatial Distribution of the Lips, Tongue and Hyoid (A) Shown is the scatterplot for the relative spatial distribution of the upper lip (light green), lower lip (dark green), upper and lower tongues (red), upper and lower hyoid (blue), during lip-smacking, for all frames, all sessions, and all monkeys. Observe that the scale is normalized such that the upper hyoid is always at the origin (coordinate [0,0]) and the lower tongue is always at position (1,0). (B) Shown is the scatterplot for the same structures as (A) but during chewing. (C and D) Shown are the centroids and 95% confidence intervals for the spatial distributions of different structures shown in (A) and (B), respectively.

Figure 3

Rhythmic Structure of Orofacial Movements Time-displacement plot of the tongue, interlip distance, and hyoid for two examples of lip-smacking from two different monkeys (A) and chewing from two different monkeys (B). The x axis depicts the time in s and the y axis the displacement in cm. Power spectra of the time-varying displacement of the tongue, interlip distance, and hyoid during lip-smacking (C) and chewing (D) are shown. x axis depicts the frequency in hertz, and the y axis depicts the value of the logarithm of power spectrum in normalized units. The upper plots show the power spectrum of the displacement of tongue (red), interlip distance (green), and hyoid (blue). The lower plots show the power spectrum of the same structures during the chewing.

Figure 4

Temporal Coordination of Vocal Tract Elements (A) PDC plots between tongue, interlip distance, and hyoid for lip-smacking. x axis depicts the frequency in hertz and the y axis the PDC value in normalized units (between 0 and 1). The solid black line in each plot indicates the PDC value for the respective pair of structures. The brown area corresponds to the 95% confidence intervals. The significant PDC values with p values < 0.01/6 are shown in red (tongue to lips) and in green (lips to tongue). The arrows in the center depict the direction of significant influence from each structure onto to the other two as measured by the PDC analysis. (B) PDC plots between tongue, lips, and hyoid for the chewing. The conventions follow (A).