Figure 1.

Word comprehension data. Top panel: Six datasets showing nonlinear improvements in online spoken word comprehension (y-axis) as a function of age (x-axis), as measured in the lab (left to right: Data replotted from four articles by Bergelson and Swingley (B&S + publication year, n = 73, 70, 49, 33), Bergelson et al. (2017; n = 44 (longitudinal), and Garrison et al., 2020; n = 30). 13–38% of infants’ mothers have less than a college degree across datasets. Bottom panel: Cross-sectional data (left; from Wordbank, Frank et al., 2020; n = 1,804; of the 59% reporting maternal education, 55% have less than a college degree) and longitudinal data (right, from Bergelson, 2016; n = 44; 25% of infants’ mothers have less than a college degree), showing gradual increase in overall vocabulary with age, in contrast to the top panel’s real-time comprehension data. Bottom right panel and fifth-top panel show longitudinal data from the same children, highlighting the nonlinear versus linear shift in in-lab word comprehension versus overall reported comprehension vocabulary. Vertical lines show the onset of the comprehension boost, ~12–14 months.