Abstract

Introduction Our previous study demonstrated that the near-infrared topography (NIRT) could be used for the investigation of neonates linguistic activity (Figure 1). [1] NIRT is a new method of neuroimaging that measures the brain activity by the estimation of the relative changes in the concentration of oxy, deoxy, and total hemoglobin (total-Hb). Each Hb concentration change is measured from the changes in light attenuation probed by a combination of two optical fibers: one for irradiation (I) and one for detection (D). The amplitude of the changes in Hb concentration induced by brain activity is dependent on the distances between I and D because measured Hb concentrations are amplified by the scattering path-length in the cortex. This study therefore explores the quality of the brain response at different I-D distances in neonates during the same type of language stimulus as our previous study.[2] Methods and Results To evaluate the relation between amplitude of the signal and I-D distances, we produced a 6-channel headset comprising two probes for both hemispheres. Three different I-D distances, 22±2 , 29±2, and 32±2 mm, were employed to measure the amplitude of the changes in total-Hb concentration associated with speech sound in neonates at the same brain areas in both temporal lobes. Figure 2 shows the fiber arrangement in the headset. The seven analyzed infants were two- to five-days old and belong to monolingual French-speaking families. Parents were informed about the study by one of the authors and gave their written consent. We exposed the neonates to three experimental conditions presented in a counterbalanced order: (1) passive listening to French speech (FW), (2) passive listening to reversed speech (BW), and (3) non-auditory stimulation (SI). The stimuli were presented in periods of 15 sec followed by a silent period of variable duration (from 25 to 35 sec). Infants were exposed to ten periods under the same condition, and then other conditions were presented. Figure 3 shows measured relative concentration changes with total-Hb at each I-D distance in both hemispheres (LH and RH). The values imply mean total-Hb between 5 and 10 seconds after stimuli onset. Figure 4 shows the relationship between simulated regional scattering path-length (rSPL) in the cortex for each I-D distance and measured mean total-Hb. The value of rSPL is estimated from the figure in the previous paper.[2] Discussion FW stimulation induced larger homodynamic changes than the other stimulations. This result is consistent with our previous study. The amplitude increased with increasing I-D distance when neonates were exposed to the FW stimulus. This result indicates that longer I-D distance can amplify homodynamic responses induced by linguistic activity. As shown in Fuigure 4, the relation between the amplitude and rSPL in the cortex is linear (R2=0.990). From this result, it is estimated that the absolute concentration change of total-Hb in the cortex is about 0.002 mM, which equals the gradient of the linear regression in Figure 4. References [1] M. Pena, et al., Proc. Natl. Acad. USA, 100, 11702-5(2003). [2] Y. Fukui, et al., Appl. Opt., 42(16), 2881-7(2003).