Brain responses to cross-modal semantic priming in Italian twenty-months-olds

Children begin to establish lexical–semantic representations during their first year of life: this results in vocabulary growth that is initially very slow but speeds up around 18 months (vocabulary spurt). At the neurophysiological level, lexical-semantic processing is reflected in the N400 component of event-related brain potential (ERP), a centro-parietal negativity occurring between 200 and 700 ms (and peaking around 400 ms) after the presentation of a semantically incongruent stimulus [1]. A few studies investigated the development of lexical–semantic processing at the electrophysiological level in the age range of interest for vocabulary spurt (18 to 24 months) [2,3,4]. In the present study, we investigate the brain mechanisms responsible for the N400 effect elicited by incongruous words and non-words in 20-month-old Italian toddlers, with the aims of: 1) testing an electrophysiological procedure for assessing word comprehension and recognition in very young children; 2) investigating whether typically developing children of the same chronological age (20 months), but of different vocabulary sizes, perform differently in such a comprehension task. Fifteen Italian-speaking toddlers ranging between 20.3 and 22.5 months were included in the study (mean age = 20.8). Linguistic skills were assessed using standardized (a and b) and experimental (c) tasks: a) Language Development Survey (LDS) [5], provides parental reports of expressive vocabulary. b) PinG Test [6], a picture naming task with norms on the Italian population (age 19-37 months). c) Electrophysiological measures of semantic comprehension assessed though a cross-modal picture–word priming paradigm (see Figure 1 for a graphical representation of the paradigm). While looking at pictures of known objects, subjects listened to spoken words that were either congruous or incongruous to the picture contents or to legal nonwords. All stimuli were controlled for age of acquisition, frequency and length in milliseconds. ERPs were recorded from 60 scalp sites using the EGI recording system. Analysis were conducted on 16 centro-parietal sites (Figure 1). At the behavioral level, children show extreme variability in both comprehension and production skills. Expressive vocabulary (a) ranged between 0 and 235 words (M=84.76, SD=68.08) and Z-scores in the picture naming task (b) ranged between -1.65 and +1.65 (M=0.75, SD=1.05). However, brain responses to lexical-semantic priming (c) did not look different across participants. Both incongruous words and nonwords elicited an N400, especially at left posterior sites (see Figure 1). In particular, mean amplitude was calculated in two subsequent time windows (TW1: 200-400 and TW2: 400-600 msec), resulting in earlier N400 effect for nonwords than for incongruous words. No differences emerged according to language proficiency. The results suggest different brain responses to incongruous words and non-words in picture context in 20-month-olds. In addition, the preliminary results reveal that very early neural mechanisms underlying N400 generation are not related to the infants’ state of behavioral language development in this age range of high interest for vocabulary development. The sample will be enlarged in the next months, and these data will be included in a wide research project aiming at identifying early neuropsychological markers for language impairment.