A preliminary account of the effect of otitis media on 15-month-olds' categorization and some implications for early...

In the present study, 24 infants 14 to 16 months old (M = 15.3 months) with a history of otitis media (OM) and tube placement were tested for categorical responding within a visual familiarization-discrimination paradigm (cf. Roberts & Jacob, 1991; Roberts, 1995a). Sixteen infants were, on average, 4.6 months post-tubes. At the time of testing, hearing was reported to be normal. During both the familiarization and testing phases, reiterant speech ulterances (e.g., "ti ti the sasasa," "bo bo the sasasa," "gu gu the sasasa," "sasasa") were randomly presented contingent on infants' looking at line drawings of selected animals. The inclusion of the article the in these utterances provides an important grammatical cue signaling an upcoming noun label. Using these same procedures, utterances, and animal stimuli, normal-hearing 15-month-olds without a reported history of OM had categorized successfully in two previous experiments. In contrast, the OM infants did not respond categorically under these conditions. This suggests that even mild hearing loss may adversely affect categorical responding under specific input conditions and that this effect may persist after normal hearing is restored. A plausible account is that fluctuating hearing loss may cause the low-phonetic-substance article to vacillate above and below some attentional threshold. The resulting inconsistency in the article's availability would disrupt the co-occurrence relation between the article and cues already connected to joint attentional/referential interactions and categorical treatment of objects (cf. Roberts, 1995a). Such a disruption has implications for the acquisition of words and other aspects of language.