Poster #39 - Do Toddlers Understand The Temporal Priority Priniciple in Causal Reasoning?

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Cognitive Processes

Integrative Statement

The temporal priority principle states that causes must precede their effects. Adults have a robust understanding of this principle (Burns & McCormack, 2009), and infant looking-time results suggest temporal priority may be an innate or early developing causality cue (Mascalzoni et al., 2013). However, findings based on children’s verbal judgments are mixed. While some studies suggest successful use of the principle in children as young as 3-years-old (Bullock & Gelman, 1979), others suggest that consistent use may not develop until 4-5 years, or even later (Shultz & Mendelson, 1975; Sophian & Huber, 1984; Rankin & McCormack, 2013; Lohse et al., 2015). No study has investigated temporal priority in children younger than 3 years. Since previous studies used verbal judgments about causality, linguistic ability may have played a role in younger children’s mixed performance (Rankin & McCormack, 2013). In the current study, we investigate whether 12-35-month-old toddlers (N=83) can use temporal priority to guide their causal reasoning, using a non-linguistic behavioral measure.

Consistent with previous work in older children (Bullock & Gelman, 1979; Rankin & McCormack, 2013), we used an A-Effect-B paradigm, where action A precedes an effect that is then followed by action B. The effect was the ejection of a sticker from a machine, while A and B were two actions (e.g., a dial and a lever) on either side of the machine’s dispenser. Participants were randomly assigned to one of two between-subject conditions, connected or disconnected, manipulating the spatial relation between the temporally prior action and the effect (Figure 1). The disconnected condition pitted spatial cues against temporal cues and was consequently a more stringent test of temporal priority, which 3-5-year-olds have previously performed well in (Bullock et al., 1979, 1982). In the connected condition, both actions were spatially contiguous with the effect. In both conditions, participants watched an adult perform the sequence A-Effect-B twice. The demonstrator then left the testing area and the toddler was encouraged to interact with the machine and obtain up to five stickers of their own.

Toddlers’ first action was on the temporally prior action A significantly more often than action B on their first sticker (binomial test, connected: 29/41, p=0.01; disconnected: 29/42, p=0.02; Figure 2a), and across all five stickers (single-sample-t-test, connected: M=0.81, SD=0.39, t(40)=7.20, p<0.0001; disconnected: M=0.78, SD=0.41, t(41)=7.55, p<0.0001; Figure 2b). Moreover, toddlers touched A first equally often in the connected and disconnected conditions, on both their first sticker (Fisher’s exact test, p=1.0) and across all five stickers (independent-t-test; connected: M=0.81, SD=0.39; disconnected: M=0.78, SD=0.41; t(81)=0.51, p=0.31), preferring the temporally prior action even when it was spatially disconnected from the effect.

By inferring causal reasoning through participant behavior rather than verbal judgments, we were able to demonstrate that children as young as 12 months seem to grasp the temporal priority principle, and that this cue to causality dominates spatial contiguity in this age group. Previously reported failures of older children to understand temporal priority were likely due to the additional cognitive demands of making explicit verbal judgments.

Authors

• Pingki Mazumder, University of Toronto
  Presenting Author

• Emma Tecwyn, Cardiff University
  Non-Presenting Author

• Daphna Buchsbaum, University of Toronto
  Non-Presenting Author