Individual Submission Summary
Share...

Direct link:

Poster #41 - Newborn Infant Speech Perception Predicts Math Skills At Four Years

Thu, March 21, 12:30 to 1:45pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Background: Early identification and intervention for cognitive disabilities have long been thought to be crucial for maximizing cognitive developmental outcomes. Several studies have attempted to identify risk for disabilities in preschool and elementary children (Aarnoudse-Moens et al., 2013; Kiechl-Kohlendorfer et al., 2013). While these studies laid groundwork for prediction models, they primarily focused on later periods of development. However, Molfese and colleagues had previously noted success in identifying infants at birth who were at-risk for developing later language impairments. They found that event-related potentials (ERPs) could successfully identify infants shortly after birth who were at-risk for impaired language development (Espy et al., 2004; Maitre et al., 2013; Molfese et al, 1985).

Hypotheses: Based on previous work elucidating the link between brain maturation at birth and subsequent developmental language outcomes (Espy et al., 2004), this study investigated the generalizability of predicting brain maturation for other cognitive processes. More specifically, the present research investigated possible precursors of math disabilities utilizing ERP brain responses recorded from infants 36 hours after birth. We hypothesized that infant brain responses to speech and nonspeech stimuli would predict individuals predisposed to developing math difficulties later in life.

Methods: Auditory stimulus was presented to 84 term infants within 36 hours after birth. Auditory ERP brain responses to the stimuli were recorded from six scalp electrode sites. Stimuli consisted of two computer-synthesized, consonant-vowel speech syllables (/ba/, /ga/), each composed of three formants with bandwidths of 60, 90, and 120 hz (formant 1, 2, and 3 respectively). Two nonspeech analogs were used as control, each composed of three sine waves with width 1 hz bandwidths matched to the center frequency for each of the three formants of the speech syllables. Subsequent math abilities were assessed using the quantitative reasoning test of the Stanford-Binet math subtest administered at four years of age.
Results: The averaged ERP amplitude and latency for the N1, N2, and P2 peaks elicited by each stimulus and each infant were analyzed using a regression procedure. The speech stimuli were used to predict quantitative performance as measured on the Stanford-Binet at four years of age. ERPs to five stimuli produced an r-square of 0.332 in predicting subsequent performance.

Discussion: Results indicate that early auditory discrimination abilities were related to later emerging quantitative skills at four years of age. Children with better quantitative skills exhibited more advanced nervous system development at birth. This was indexed by ERP components that made finer discriminations between speech sounds.

Authors