The robustness and generalizability of findings on spontaneous false belief sensitivity: a replication attempt

Abstract

Influential studies showed that 25-month-olds and neurotypical adults take an agent's false belief into account in their anticipatory looking patterns (Southgate et al. 2007 Psychol. Sci. 18, 587–592 (doi:10.1111/j.1467-9280.2007.01944.x10.1111/j.1467-9280.2007.01944.x); Senju et al. 2009 Science 325, 883–885 (doi:10.1126/science.117617010.1126/science.1176170)). These findings constitute central pillars of current accounts distinguishing between implicit and explicit Theory of Mind. In our first experiment, which initially included a replication as well as two manipulations, we failed to replicate the original finding in 2- to 3-year-olds (N = 48). Therefore, we ran a second experiment with the sole purpose of seeing whether the effect can be found in an independent, tightly controlled, sufficiently powered and preregistered replication study. This replication attempt failed again in a sample of 25-month-olds (N = 78), but was successful in a sample of adults (N = 115). In all samples, a surprisingly high number of participants did not correctly anticipate the agent's action during the familiarization phase. This led to massive exclusion rates when adhering to the criteria of the original studies and strongly limits the interpretability of findings from the test phase. We discuss both the reliability of our replication attempts as well as the replicability of non-verbal anticipatory looking paradigms of implicit false belief sensitivity, in general.

On the domain specificity of the mechanisms underpinning spontaneous anticipatory looks in false‐belief tasks

“…While most published studies on spontaneous mentalizing have reported positive evidence, from adults as well as from infants as young as 7 months (Kovács et al, 2010), recent works have also reported failures to replicate earlier findings (e.g., Burnside, Ruel, Azar, & Poulin-Dubois, 2018;Schuwerk, Priewasser, Sodian, & Perner, 2018). Some of these works are best described as conceptual replications (e.g., ,…”

Infants attribute false beliefs to a toy crane

“…Other paradigms have been developed to test false belief understanding in infancy (e.g. anticipatory looking, helping) but they have also proven difficult to replicate (Burnside, Ruel, Azar, & Poulin-Dubois, 2018;Crivello & Poulin-Dubois, 2018;Priewasser, Rafetseder, Gargitter, & Perner, 2018;Schuwerk, Priewasser, Sodian, & Perner, 2018). Additionally, there appears to be a lack of convergence in performance when the same infants are administered pairs of non-traditional false belief tasks (e.g.…”

Do infants understand false beliefs? We don’t know yet – A commentary on Baillargeon, Buttelmann and Southgate’s commentary

“…With the possible exception ofWang & Leslie, 2016. This study replicated only one of the original conditions (FB1), however. And see alsoSchuwerk et al (2018) for critical concerns about the analyses that do deviate considerably from the original ones. 3 It should be noted, though, that only the FB2 condition ofSouthgate et al (2007) was tested here.…”