Infants' Selective Attention to Reliable Visual Cues in the Presence of Salient Distractors

Abstract

With many features competing for attention in their visual environment, infants must learn to deploy attention toward informative cues while ignoring distractions. Three eye tracking experiments were conducted to investigate whether 6- and 8-month-olds (total N = 102) would shift attention away from a distractor stimulus to learn a cue-reward relation. While 8-month-olds showed evidence of increasingly selective attention toward the predictive cues, even when the distractors were highly salient, 6-month-olds shifted attention toward the predictive cues only when the distractors were equally (not more) engaging. These experiments suggest that attention in infancy is highly dependent on the relative weightings of predictiveness and visual salience, which may differ across development and context.

Infants’ Looking to Surprising Events: When Eye-Tracking Reveals More than Looking Time

“…Our results suggest a new interpretation of the divergent findings in the infant reasoning literature. Specifically, looking times in single sampling events (from large populations) may be sensitive to relative likelihood (i.e., converging with previous reports using search measures with probabilistic containers [16]), but only when considering gaze to specific locations. We are currently investigating this hypothesis in an eye tracking procedure that uses probabilistic, rather than deterministic sampling.…”

“…For example, do crawling paradigms (e.g., [16,17]) capitalize on infants’ motivation to search for desired objects in ways that are distinct from LT paradigms? How might various measures of LT—compared to search paradigms, for example—tap inferences that are either predictive or post-dictive (e.g., [35])?…”

“…Increasingly, both infant and adult research studies measure other types of gaze information, asking how looking to specific locations reveal more subtle measures of cognitive ability. Such techniques have been used to explore implicit theory-of-mind [7–10], perceptual categorization abilities [11–15], as well as many other aspects of cognitive and perceptual processing (e.g., [16–20]). In the present report, we focus on a research domain in which these advanced eye-tracking techniques have not yet been applied: Infants’ understanding of surprising sampling events in procedures that measure inferential reasoning, and which have garnered significant attention in the cognitive development literature in recent years [21–23].…”

“…One difference is that these latter experiments used an explicit choice task (i.e., crawling and searching) when testing samples drawn from large collections of objects ([16,17]; see [19] for similar findings with apes). In sum, LT experiments on infants’ reasoning abilities have revealed some failures to make inferences about single samples from distributions containing more items than can be individually tracked, but other tasks have revealed success.…”

Infant Statistical Learning

“…This suggests that sensitivity to input statistics is context dependent. A series of studies by Tummeltshammer and colleagues (2014a,b; 2017) have expanded upon the issue of context, showing that visual statistical learning is mediated by stimulus salience, source reliability, and mode of presentation.…”