The agent-based approach emphasizes the importance of learning through organism-environment interaction. This approach is part of a recent trend in computational models of learning and development toward studying autonomous organisms that are embedded in virtual or real environments. In this paper we introduce the concepts of online and offline sampling and highlight the role of online sampling in agent-based models. After comparing the strengths of each approach for modeling particular developmental phenomena and research questions, we describe a recent agent-based model of infant causal perception. We conclude by discussing some of the present limitations of agent-based models and suggesting how these challenges may be addressed. © 2001 Academic Press Computational models of learning and development are playing an increasingly critical role in child development research (Cassidy, 1990;

While computational models are playing an increasingly important role in developmental psychology, at least one lesson from robotics is still being learned: modeling epigenetic processes often requires simulating an embodied, autonomous organism. This paper first contrasts prevailing models of infant cognition with an agent-based approach. A series of infant studies by Baillargeon (1986; Baillargeon & DeVos, 1991) is described, and an eye-movement model is then used to simulate infants' visual activity in this study. I conclude by describing three behavioral predictions of the eyemovement model, and discussing the implications of this work for infant cognition research.

Much of what psychologists know about infant perception and cognition is based on habituation, but the process itself is still poorly understood. Here the authors offer a dynamic field model of infant visual habituation, which simulates the known features of habituation, including familiarity and novelty effects, stimulus intensity effects, and age and individual differences. The model is based on a general class of dynamic (time-based) models that integrate environmental input in varying metric dimensions to reach a single decision. Here the authors provide simulated visual input of varying strengths, distances, and durations to 2 coupled and interacting fields. The 1st represents the activation that drives “looking, ” and the 2nd, the inhibition that leads to “looking away, ” or habituation. By varying the parameters of the field, the authors simulate the time course of habituation trials and show how these dynamics can lead to different depths of habituation, which then determine how the system dishabituates. The authors use the model to simulate a set of influential experiments by R. Baillargeon (1986, 1987a, 1987b) using the well-known “drawbridge ” paradigm. The dynamic field model provides a coherent explanation without invoking infant object knowledge. The authors show that small changes in model parameters can lead to qualitatively different outcomes. Because in typical infant cognition experiments, critical parameters are unknown, effects attributed to conceptual knowledge may be explained by the dynamics of habituation.

We present new evidence to support criticisms of infant looking time experiments. One such experiment, in which Baillargeon concluded from looking times that infants understand object permanence, is examined in detail. An alternative model of infant cognition, using the idea that infants look longer at particular scenes based on visual processing at the pre-attentive level, rather than based on any understanding of the objects they are seeing, is suggested. The model is implemented on Nico, a humanoid robot currently being developed in the Yale Social Robotics Lab. The validity of the model is established by running Nico through a reenactment of Baillargeon’s initial experiment and achieving comparable results. The argument is made that while these results do not prove the suggested model is correct, they do prove that the suggested model is sufficient for explaining Baillargeon’s results. We conclude that the demonstrated validity of the model prevents Baillargeon from claiming that the initial experiment provides proof of an understanding of object permanence in infants. We suggest that the model could be further validated by running Nico through other looking time experiments. 1.

The capacity for infants to form mental representations of hidden or occluded objects can be decomposed into two tasks: one process that identifies salient objects and a second complementary process that identifies salient locations. This functional decomposition is supported by the distinction between dorsal and ventral extrastriate visual processing in the primate visual system. This approach is illustrated by presenting an eye-movement model that incorporates both dorsal and ventral processing streams and by using the model to simulate infants ’ reactions to possible and impossible events from an infant looking-time study (R. Baillargeon, “Representing the existence and the location of hidden objects: object permanence in 6- and 8-month-old infants”, Cognition, 23, pp. 21–41, 1986.). As expected, the model highlights how the dorsal system is sensitive to the location of a key feature in these events (i.e. the location of an obstacle), whereas the ventral system responds equivalently to the possible and impossible events. These results are used to help explain infants’reactions in looking-time studies. Keywords: Object representations; Dorsal–ventral model; Infant perception 1.

Abstract. Schlesinger (2003, Adaptive Behavior, 11: 97–107) recently proposed a model of eye movements as a tool for investigating infants ’ visual expectations. In the present study, this gazedirection model was evaluated by: (a) generating a set of predictions concerning how infants distribute their attention during possible and impossible events; and (b) testing these predictions in a replication of Baillargeon’s ‘car study ’ (Baillargeon, 1986, Cognition, 23: 21–41, Baillargeon and DeVos, Child Development, 62: 1227–1246). We found that the model successfully predicts general features of infants ’ gaze direction, but not specific differences obtained during the possible and impossible events. The implications of these results for infant cognition research and theory are discussed.

Manual search for totally occluded objects was investigated in 10-, 12- and 14-monthold infants. Infants responded to two types of total hiding in different ways, supporting the inference that object permanence is not a once-and-for-all attainment. Occlusion of an object by movement of a screen over it was solved at an earlier age than occlusion in which an object was carried under the screen. This dissociation was not explained by motivation, motor skill or means–ends coordination, because for both tasks the same object was hidden in the same place under the same screen and required the same uncovering response. This dissociation generalized across an experimentally manipulated change in recovery means—infants removed cloths while seated at a table in Expt 1 and were required to crawl through 3-D space to displace semi-rigid pillows in Expt 2. Further analysis revealed that emotional response varied as a function of hiding, suggesting an affective correlate of infant cognition. There are four empirical findings to account for: developmental change, task dissociation, generalization of the effects across recovery means, and emotional reactions. An identity-development theory is proposed explaining these findings in terms of infants’

Smith, 1999) criticized several reports by investigators (e.g., Baillargeon, 1999; Spelke, 1998; Wynn, 1998) who offered a so-called “rich interpretation ” of findings gathered from young infants. The fundamental danger associated with such rich interpretations is that some simpler lower level mechanism may be discovered that can account for these same findings. This commentary is directed to three articles that appear in this issue of Infancy (Bogartz, Shinskey, & Schilling; Cashon & Cohen; Schilling). All three articles present empirical evidence from 4-, 5-, and 8-month-old infants that the authors contend is more consistent with a perceptually based interpretation than with the rich interpretation offered by two classic studies of object permanence in young infants (Baillargeon, 1987a; Baillargeon, Spelke, & Wasserman, 1985). I argue that the alternative explanation offered by these three sets of authors is inconsistent with both the original studies and with their own data. THE ORIGINAL “DRAWBRIDGE ” STUDIES In Baillargeon et al. (1985), 5-month-old infants were habituated to a screen that rotated 180 ° (fore and aft), much like a drawbridge. No object (other than the screen) was present during habituation, and the now-standard infant-control habituation procedure was used. Both types of test trials involved the placement of an Requests for reprints should be sent to Richard N. Aslin, Department of Brain and Cognitive Sciences,

When an object moves behind an occluder and re-emerges, 4-month-old infants perceive trajectory continuity only when the occluder is narrow, raising the question of whether time or distance out of sight is the important constraining variable. One hundred and forty 4-month-olds were tested in five experiments aimed to disambiguate time and distance out of sight. Manipulating the object’s visible speed had no effect on infants ’ responses, but reducing occlusion time by increasing object speed while occluded induced perception of trajectory continuity. In contrast, slowing the ball while it was behind a narrow or intermediate screen did not modify performance. It is concluded that 4-month-olds perceive trajectory continuity when time or distance out of sight is short. In everyday environments, the objects that we encounter frequently pass into and out of sight as our view of them is obstructed by nearer objects, and yet we perceive these objects as enduring entities. Additionally, when objects move, we perceive their trajectories as continuous even though they may be invisible for part of their path of movement. This ability to fill in the gaps in perception is a fundamental aspect of object perception, and important questions arise regarding its developmental origins. Early work investigating young infants ’ perception of events in which an object passes behind a screen was framed in terms of infants ’ knowledge of object permanence. For instance, Bower, Broughton, and Moore (1971) reported that 2-month-olds anticipated the re-emergence of an object from behind a screen, evidence for object permanence. Moreover, they showed tracking disruption (i.e., oculomotor search) when an object emerged from behind a screen too soon, but not when it changed its form behind the screen, and Bower et al. (1971) interpreted this as indicating a difficulty in integrating object

Abstract not found