Statistical Learning Is Not Age‐Invariant During Childhood: Performance Improves With Age Across Modality

Abstract

Humans are capable of extracting recurring patterns from their environment via statistical learning (SL), an ability thought to play an important role in language learning and learning more generally. While much work has examined statistical learning in infants and adults, less work has looked at the developmental trajectory of SL during childhood to see whether it is fully developed in infancy or improves with age, like many other cognitive abilities. A recent study showed modality‐based differences in the effect of age during childhood: While visual SL improved with age, auditory SL did not. This finding was taken as evidence for modality‐based differences in SL. However, since that study used auditory linguistic stimuli (syllables), the differential effect of age may have been driven by stimulus type (linguistic vs. non‐linguistic) rather than modality. Here, we ask whether age will affect performance similarly in the two modalities when non‐linguistic auditory stimuli are used (familiar sounds instead of syllables). We conduct a large‐scale study of children's performance on visual and non‐linguistic auditory SL during childhood (ages 5–12 years). The results show a similar effect of age in both modalities: Unlike previous findings, both visual and non‐linguistic auditory SL improved with age. These findings highlight the stimuli‐sensitive nature of SL and suggest that modality‐based differences may be stimuli‐dependent, and that age‐invariance may be limited to linguistic stimuli.

Do current statistical learning tasks capture stable individual differences in children? An investigation of task reliability across modality

“…The pattern of correlations was stable across sessions, another indication of task stability: The two auditory tasks were not correlated, whereas visual task and the nonlinguistic auditory task were. This pattern of correlation may seem counterintuitive because the tasks do not group together on the basis of modality; however, it is compatible with a set of recent findings suggesting that tasks using linguistic stimuli (like syllables) behave differently from tasks using nonlinguistic stimuli, regardless of modality (Siegelman et al, 2018a;Shufaniya & Arnon, 2018). An additional stable pattern was the positive correlation between the SL tasks and working memory, a pattern that has been reported in some adult studies (Misyak & Christiansen, 2012) but not others .…”

“…SL tasks using linguistic stimuli (syllables) are more affected than visual tasks by the exact stimuli used and show evidence of L1 influences (Siegelman et al, 2018b). Developmentally, the effect of age on performance seems different for tasks using linguistic stimuli: Whereas performance on visual and nonlinguistic auditory SL improves during childhood (ages 5-12 years), performance on linguistic auditory SL does not (Shufaniya & Arnon, 2018). Since studies finding modalitybased differences often use linguistic auditory tasks (Emberson et al, 2011;, previously reported differences (and similarities) may be driven not only by modality by also by the specific stimuli use.…”

Statistical learning abilities and their relation to language

“…For example, while it is well documented that humans are sensitive both to the transitional statistics and to the aggregated frequency and variance of exemplars in the input, Thiessen, Kronstein, and Hufnagle () claimed that from a computational perspective, the assimilation of these two types of regularities is likely subserved by different processes. Concurring behavioral evidence also showed patterns of informational‐specificity (specifically, in learning adjacent vs. non‐adjacent contingencies: e.g., Newport & Aslin, , and verbal vs. nonverbal material: Shufaniya & Arnon, ; Siegelman, Bogaerts, Elazar, Arciuli, & Frost, ). Lastly, joining these group‐level observations are preliminary individual‐differences studies revealing patterns of modality and informational‐specificity in SL.…”

Assessing Visual Statistical Learning in Early-School-Aged Children: The Usefulness of an Online Reaction Time Measure

“…The development of online measures is especially important for studies investigating SL in early-school-aged children due to the fact that the traditional 2-AFC questions require explicit decision-making, a skill that young children have difficulties with (Bialystok, 1986). Children’s performance on 2-AFC questions in VSL tasks is known to increase between the ages of 5 and 12 years (Arciuli and Simpson, 2011; Raviv and Arnon, 2017; Shufaniya and Arnon, 2018). For this reason, solely using 2-AFC questions to assess early-school-aged children’s performance may not provide a complete picture of their SL abilities.…”