All (tilt) models are wrong, but some are useful: A reply to Sorjonen et al.'s (2023) critique of tilt

Tilt refers to a pattern of specific abilities and is based on within subject differences in two abilities (e.g., math and verbal), yielding an ability profile with strengths and weaknesses. A common type of tilt is ability tilt based on math and verbal scores, yielding math tilt (math>verbal) and verbal tilt (verbal>math). This article responds to Sorjonen, Ingre, Nilsonne, and Melin's (2023, this issue) critique of tilt and investment theories. In their critique, Sorjonen et al. claim that tilt results are spurious and that investment theories cannot explain tilt. Contra Sorjonen et al.'s claims, the current article argues that the nomological network of tilt relations is not spurious but is parsimonious, falsifiable, and supports the predictions of investment theories. The nomological network indicates that (a) tilt levels increase with age in adolescence and are mediated by processing speed; (b) males show math tilt, whereas females show verbal tilt; and (c) math tilt correlates positively with analogous criteria (e. g., science and math) and negatively with competing criteria (e.g., verbal), whereas verbal tilt shows the opposite pattern. It is argued that these (and other) findings are not spurious but can be parsimoniously explained by investment theories, which assume that investment in a particular domain (e.g., STEM; science, technology, engineering, math) boosts the development of analogous abilities (e.g., math) and inhibits the development of competing abilities (e.g., verbal). The article concludes with a discussion of future research on tilt, focusing on factors that may affect the development of tilt and its predictive power (e.g., trait complexes and developmental dedifferentiation).

Automated summary

Tilt refers to a pattern of specific abilities based on within subject differences in math and verbal scores. This article responds to a critique of tilt and investment theories, arguing that the nomological network of tilt relations is not spurious but supports the predictions of investment theories. The article concludes with a discussion of future research on tilt, focusing on factors that may affect its development and predictive power.