Sensitive Period Diversity: Insights From Evolutionary Models

• Published in: The Quarterly review of biology Vol. 97; no. 4; pp. 243 - 295
• Main Authors: Stamps, Judy Ann, Luttbeg, Barney
• Format: Journal Article
• Language: English
• Published: Baltimore The University of Chicago Press 01.12.2022; University of Chicago, acting through its Press
• Subjects: Adaptiveness; Bayesian analysis; Biology; Critical period; Deprivation; Design of experiments; Evolution; Experimental design; Mathematical models; Reliability aspects; Stimuli; Variation; Bayesian updating; adolescence; sensitive windows; developmental plasticity; critical periods
• ISSN: 0033-5770; 1539-7718
• DOI: 10.1086/722637

Recent models of the evolution of sensitive periods in response to informative stimuli (i.e., cues) provide insights into reasons why empiricists within and across disciplines might observe variation in the patterns they observe when they study sensitive periods. We consider what an evolutionary perspective can tell us about the appropriate age to begin and end a study of sensitive periods, and how the experimental design and the predispositions of the subjects in an empirical study might affect its results. Using models based on Bayesian updating, we show how several factors, including the protocol used to study sensitive periods, the relative reliability of the information provided to the subjects, and the predispositions of the subjects, are expected to affect the presence and shapes of sensitive periods that occur in response to information-only cues. Our results suggest that investigators will observe considerable variation in the patterns reported in empirical studies of sensitive periods simply based on the protocol they use to study them; this is relevant because investigators working in different disciplines tend to rely on different protocols. We show how theory can help shed light on the adaptive significance of patterns reported by empiricists, e.g., why we might expect to observe heightened sensitivity to particular cues during adolescence. We describe existing empirical support for some of the models’ predictions, e.g., that the duration of sensitive periods will be extended if subjects are first maintained in noninformative conditions before they are first exposed to informative stimuli. We highlight novel predictions of these models that can be readily tested by empiricists, e.g., that the effects of deprivation treatments on sensitive periods will vary, depending on their timing. More generally, we show how an evolutionary approach to sensitive periods reveals that at least some of the considerable variation observed in the results reported in empirical studies of sensitive periods may be attributable to variation in the methods that are used to study them.