Hysteresis affects approximate number discrimination in young children.

Abstract

Perceptual decisions are often affected not only by the evidence gathered during a trial but also by the history of preceding trials. This effect—termed perceptual hysteresis—provides evidence for how perceptual information is represented and how it is used. The present research focuses on how the difficulty of preceding trials affects subsequent ones—we find that how well 5-year-old children perform in a 2-alternative forced-choice numerical discrimination task depends on whether they have had a prior history of easier discriminations or a prior history of harder discriminations. Furthermore, this effect is modulated by the feedback children receive. In 3 experiments, we demonstrate that these effects are not related to practice or loss of interest due to negative feedback, or simply to trial difficulty or discriminability. Instead, children appear to have state-dependent confidence states such that prolonged experience making low-confidence decisions degrades performance, whereas prolonged experience making high-confidence decisions improves it. These results are discussed in the context of dynamical psychophysics, representations of confidence, and work on children’s and adults’ number perception abilities.

Dot comparison stimuli are not all alike: The effect of different visual controls on ANS measurement

“…This finding can provide clarification on the conflicting findings regarding congruency effects that have been reported in the literature to date; differences are likely due to some researchers failing to consider convex hull (e.g. Odic et al, 2013Odic et al, , 2014. The presence of convex hull congruency effects in trials created with both protocols provides support for the developing hypothesis that participants may recruit different mechanisms, such as inhibitory control skills, to complete incongruent trials (Fuhs & McNeil, 2013;Gilmore et al 2013;Nys and Content 2012;Szűcs et al 2013).…”

“…However, other studies have failed to find this effect (Odic et al, 2013(Odic et al, , 2014, or find the opposite congruency effect (Gebuis & Van der Smagt, 2011). It is possible that this is partly due to the diverse methodologies for controlling visual cues employed in the tasks.…”

“…There have been some suggestions that w scores can be used to compare performance on dot comparison tasks that use different methodological formats (Piazza et al, 2013), but this is unlikely to be the case given Smets et al's (2014) results. Additionally, Odic et al (2014) found that the order of trial presentation in a dot comparison task significantly affected w scores in a within subjects design. Participants' w scores were superior on tasks that became increasingly more difficult, in comparison to tasks that became increasingly easier, despite both manipulations of the study containing exactly the same trials overall.…”

“…This is the default setting on the Panamath software 5 , and this type of control has been used in multiple studies of the ANS (e.g. Halberda et al, 2012;Hellgren, Halberda, Forsman, Ådén, & Libertus, 2013;Libertus et al, 2011;Libertus et al, 2012;Libertus, Feigenson, & Halberda, 2013a, 2013bMazzocco et al 2011a;Mazzocco, Feigenson, & Halberda, 2011b;Odic, Libertus, Feigenson, & Halberda, 2013;Odic, Hock, & Halberda, 2014).…”

“…For example, unlike the protocol designed by Gebuis and Reynvoet (2011), some methods do not allow researchers to systematically vary the convex hull of the dot arrays created. Notably, studies which have not found standard congruency effects did not manipulate convex hull size (Gebuis & Van der Smagt, 2011;Odic et al, 2013Odic et al, , 2014. It is therefore important to understand more about when congruency effects occur, and how both 9 convex hull size and cumulative surface area influence task performance in one group of participants, across trials created with different visual cue controls.…”