Looking time and eye-tracking dwell times from infant numerical perception experiments

These data files largely contain preferential looking data for pairs of experiments that were run in tandem. One file (relating to Experiment 19 and 20) contains Tobii eye tracker dwell times to the two objects in the outcome display (the same are currently being coded for the other experiments). The document containing figures outlines the manipulation concerned in each case. All studies were based closely on Wynn’s addition and subtraction work, and in addition to replicating her conditions, these experiments systematically manipulate the identity and position of the objects that appear in the outcome display. Thus, in some cases of addition the correct numerical outcome is presented but either the originally placed object or the added object has a changed identity or is in the incorrect place. The file containing figures illustrates on the left-hand side the initial state and addition or subtraction operation, and the right-hand side indicates the outcome. Most studies consist of a familiarisation phase, and six test trials – three non-violation outcomes and three violation outcomes. Where indicated, there were four test trials. See experiment numbers for the nature of the incorrect outcomes, as illustrated in the figures illustrating the experiments.<p>Following the controversial claim that young infants understand the basic principles of addition and subtraction (Wynn, 1992), several alternative, more parsimonious, accounts have been offered for Wynn’s data. However, this work has proceeded more or less in a piecemeal fashion, and there is a need to carry out a systematic investigation of factors that may be influential in tasks measuring infants' ability to keep track of objects and their identity.

This project provides that investigation, in a programme of experiments evaluating the numerical interpretation against lower level perceptual accounts based on single and multiple object tracking. The work will mainly use infant looking time as indication of violation of expectation. Specifically, we will measure infant looking time in response to object addition and subtraction manipulations involving:</p>