Neuroscience Letters 369 (2004) 197–202 Age differences in estimating arrival-time Nicolas Benguigui a,∗ , Michael Broderick b , Hubert Ripoll c a Centre de Recherches en Sciences du Sport, Universit´ e Paris-Sud(11), UFR STAPS, Bˆ atiment 335, 91405 ORSAY Cedex, France b Arizona State University, USA c University of the Mediterranean, Marseille, France Received 9 June 2004; received in revised form 15 July 2004; accepted 19 July 2004 Abstract The present study examined the accuracy in extrapolating an occluded trajectory in relation to observer age. Adults and children aged 7, 10, and 13 were tested in a prediction-motion task which consisted of judging, after the occlusion of the final part of its path, the moment of arrival of a moving stimulus towards a specified position. Results showed that children as young as 7 years old are able to use the same strategy as adults in the extrapolation of an occluded moving object. However, accuracy in responses improves most significantly for occlusion times equal to or more than 400 ms and this improvement occurs mainly between 7 and 10 years of age. This confirms that children are less efficient in performing the computations necessary to extrapolate in time an occluded trajectory. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Arrival-time; Motion extrapolation; Prediction-motion task; Development A large number of studies have shown that children have more difficulty than adults when trying to accurately coordi- nate their actions with the displacement of moving objects in real-world tasks such as intercepting or catching a ball (e.g., [1,24]) and crossing a road (e.g., [11]). It is reasonable to as- sume that some of the age-related differences in these tasks come from an inability to correctly estimate the arrival-time (AT) of the moving object at the position where contact will occur (e.g., [22]). One method currently used to study the accuracy of AT estimates is the occlusion method, which involves the use of prediction-motion (PM) tasks (e.g., [3,23]). These tasks consist of presenting a moving object that is occluded just before it reaches the observer or a specified position. The observer is required to make a simple response (e.g., press a button) that will coincide temporally with the moving ob- ject’s immediate arrival at the observer’s position or another specified position in space. The numerous studies carried out in this field have shown that a linear relationship exists be- ∗ Corresponding author. Tel.: +33 169 15 43 11; fax: +33 169 15 62 37. E-mail address: nicolas.benguigui@staps.u-psud.fr (N. Benguigui). tween estimates of the arrival-time after the occlusion and actual arrival-time (e.g., [5,22]). According to Yakimoff et al. [26], this relationship can be expressed by the equation: T r = α(T a )+ θ, where T a is the actual arrival-time of the moving object, T r is the response time between the occlu- sion and the response (which corresponds to the participant’s estimate of T a ), and α and θ are the two parameters charac- terizing the accuracy of extrapolation. It has generally been observed that the slope α are much lower than 1 and the in- tercept θ are greater than zero. This means that participants underestimate the AT for the longer occlusions and overes- timate the AT for the shorter occlusions. Generally, the tran- sition point between under- and overestimations is 1 s (e.g., [15,21]). Yakimoff et al. [26] pointed out that the linear model would only be applicable for occlusions greater than or equal to 200 ms. In making this assertion, they took into account the duration of a visuo-motor delay during which information about the time remaining before the arrival of the moving ob- ject could not be used to coordinate the response. Following this assumption, one might infer that for occlusions shorter than 200 ms, the accuracy of responses should not be differ- 0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.07.051