Where to look first for an explanation of induction with uncertain categories

Recent research has examined how people predict unobserved features of an object when its category membership is ambiguous. The debate has focused on whether predictions are based solely on information from the most likely category, or whether information from other possible categories is also used. In the present experiment, we compared these category-based approaches with feature conjunction reasoning, where predictions are based on a comparison among exemplars (rather than categories) that share features with a target object. Reasoning strategies were assessed by examining patterns of feature prediction and by using an eye gaze measure of attention during induction. The main findings were (1) the majority of participants used feature conjunction rather than categorical strategies, (2) people predominantly gazed at the exemplars that were most similar to the target object, and (3) although people gazed most at the most probable category to which an object could belong, they also attended to other plausible category alternatives during induction. These findings question the extent to which category-based reasoning is used for induction when category membership is uncertain.     

Attention, Joint Attention, and Social Cognition

ABSTRACT— Before social cognition there is joint processing of information about the attention of self and others. This joint attention requires the integrated activation of a distributed cortical network involving the anterior and posterior attention systems. In infancy, practice with the integrated activation of this distributed attention network is a major contributor to the development of social cognition. Thus, the functional neuroanatomies of social cognition and the anterior–posterior attention systems have much in common. These propositions have implications for understanding joint attention, social cognition, and autism.     

Support vector machines categorize the scaling of human grip configurations

In previous work (Cesari & Newell, 2002), we used a graphical dimensional analysis to show that grip transitions obey the body-scaled relation K = lnL(o) + InM(o)h/(a + bM(h) + cL(h)), where L. and Mo are the object's length and mass, and Lh and Mh the length and mass of the grasper's hand. However, the generality of the equation was limited by the ad hoc graphical method that defined the lines for grip separation and by the assumption that these lines be negatively sloped and parallel to one another. This article reports an independent test of this relation by the geometrical and statistical categorization of body-scaled invariants for the transition of human grip configurations through support vector machines (SVMs). The SVM analysis confirmed the fit of linear, negatively sloped, and approximately parallel transition boundaries in the scaling of human grip configuration within a single hand. The SVM analysis has provided a theoretical refinement to the scaling model of human grip configurations.     

Virtual Time-to-Collision and Human Postural Control

This article is a report on 3 experiments designed so that the role of virtual time-to-collision (VTC), which specifies the spatiotemporal proximity of the center of pressure to the postural stability boundary in the regulation of posture in upright stances, could be examined. Virtual time-to-collision was estimated for normal upright stance with different bases of support, and for postural oscillations in which the speed of movement and instructional constraints on the coordination mode used were manipulated. The results showed that virtual time-to-collision was predictably reduced as (a) the base of support was reduced, (b) the speed of the postural oscillation was increased, and (c) the number of biomechanical degrees of freedom regulated in the coordination mode increased. Over a range of task conditions, the coefficients of variation of the VTC time-series were significantly lower than the coefficients of variation for the velocity and acceleration time series of the center of pressure. The absolute values of VTC increased with the increment of the ground reaction forces a performer generated to avoid falling while approaching the stability boundary. These findings are consistent with the proposition that VTC may serve as an organizing informational control parameter for posture.     

A note on the speed-amplitude function in movement control

An experiment is reported that documents the maximum average speed-amplitude relationship across the full range of motion for elbow flexion. Minimum movement time increased as a negative exponential within the movement range up to 94-97% of the maximum range of motion. At this point a discontinuity occurred with movement time increasing at an increasing rate probably due to anatomical and morphological constraints. These results suggest that the maximum average velocity-amplitude boundary to the movement speed-accuracy relationship is curvilinear. Kinematic analysis of the movements as a function of range of motion suggests that a simple pulse-step model of movement control cannot account for the present findings.     

Visual control of discrete aiming movements

An experiment is reported which investigated the visual control of discrete rapid arm movements. Subjects were required to move as rapidly as possible to several target width-movement distance combinations under both visual and non-visual conditions. The movement time (MT) data were supportive of Fitts' Law in that MT was linearly related and highly correlated to the Index of Difficulty (ID). MT was also similar for different target width-distance combinations sharing the same ID value. The error rate analysis, which compared visual to non-visual perfromance, indicated that vision was only used, and to varying degrees, when MT exceeded 200 ms (3.58 ID level). There was some evidence that vision was differentially used within target width-distance combinations sharing the same ID. Estimates of endpoint variability generally reflected the results of the error rate analysis. These results do not support the discrete correction model of Fitts' Law proposed by Keele (1968).