Walter Gerbino

The effect of amodal completion on visual matching

Abstract In a series of five experiments, we investigated how amodal completion affects pattern recognition, and tested possible models of processes underlying completion of simple shapes. Inferences about processing models were based mainly upon the comparison of ‘same’ latencies in a simultaneous matching task. The major result of experiments 1–4 regards two conditions where a complete target had to be matched with a given stimulus region, belonging to a composite comparison pattern. Matching is faster when this stimulus region is amodally completed than when it looks like an incomplete shape. In experiment 5 we compared complete vs incomplete targets, that were either phenomenally or topographically identical to a given region of the comparison pattern. The failure to show any effect of target completeness suggests that phenomenal identity may be as effective as topographical identity.

Contour interpolation by vector-field combination.

We model the visual interpolation of missing contours by extending contour fragments under a smoothness constraint. Interpolated trajectories result from an algorithm that computes the vector sum of two fields corresponding to different unification factors: the good continuation (GC) field and the minimal path (MP) field. As the distance from terminators increases, the GC field decreases and the MP field increases. Viewer-independent and viewer-dependent variables modulate GC-MP contrast (i.e., the relative strength of GC and MP maximum vector magnitudes). Viewer-independent variables include the local geometry as well as more global properties such as contour support ratio and shape regularity. Viewer-dependent variables include the retinal gap between contour endpoints and the retinal orientation of their stems. GC-MP contrast is the only free parameter of our field model. In the case of partially occluded angles, interpolated trajectories become flatter as GC-MP contrast decreases. Once GC-MP contrast is set to a specific value, derived from empirical measures on a given configuration, the model predicts all interpolation trajectories corresponding to different types of occlusion of the same angle. Model predictions fit psychophysical data on the effects of viewer-independent and viewer-dependent variables.

Mental image reversal and verbal recoding: When ducks become rabbits

A recent debate has concerned whether classical bistable configurations (e.g., duck/rabbit) can be reinterpreted using mental imagery. Research in this field indicates that image reversal is possible only when subjects change their specification of orientation. In a series of four experiments, we demonstrate that mental reversal of classical bistable configurations( CBCs) is impeded by verbally recoding the visual pattern at the time of input. When subjects were prevented from verbally recoding visual stimuli in short-term memory, they fared systematically better in mentally reversing the CBC, even when they received no instructions to change their reference frame or specification of orientation. On this basis, we suggest a model of image reversal that takes into account the interaction between memory codes and provides a new perspective on verbal recoding, verbal overshadowing, and mental discoveries.

Classical and inverted White's effects.

In classical Whites effect, intermediate-luminance targets appear lighter when they interrupt the dark stripes of a grating and darker when they interrupt the light stripes. The effect is reversed when targets are of double-increment or double-decrement luminance, relative to the luminances of grating stripes. To find a common explanation for classical and inverted effects, we ran two experiments. In experiment 1, we utilised intermediate-target displays to show that perceived transparency dominates over occlusion only when the target luminance is close to the luminances of top regions. This result weakens transparency-based accounts of Whites effect. In experiment 2, we varied grating contrast and target luminance to measure the classical effect in seven intermediate-target cases, as well as the inverted effect in four double-increment and four double-decrement cases. Both types of effect are explained by a common model, based on assimilation to the top region and contrast with the interrupted region, weighted by adjacency along the luminance continuum.

Blindness to inconsistent local signals in motion transparency from oscillating dots

Visual processing involves hierarchical stages in which local features are initially analyzed and subsequently grouped into objects and surfaces. In the domain of motion perception, transparent motion has been used as a powerful tool to investigate the mechanisms underlying the grouping of local features. Here, we report a novel way of creating motion transparency from oscillating dots (MTOD). In this stimulus, individual dots move back and forth over a small distance. When the dots are oscillating in synchrony, global surfaces are also perceived as moving back and forth. However, when the oscillation desynchronizes, the percept turns into two moving surfaces that are sliding over each other continuously (streaming motion). The percept of MTOD is similar to conventional transparent motion, where individual dots move only in one direction. Also, when streaming motion is perceived, the detection of oscillation is impaired. This blindness to the oscillation becomes stronger, as the signal strength for the streaming motion is increased. These findings suggest that when global visual representations are constructed, weak and inconsistent local signals are discarded.

Low-Level and High-Level Processes in the Perceptual Organization of Three-Dimensional Apparent Motion

An experiment is described in which a bistable motion display is shown in four combinations of monoptic and dichoptic viewing. When two triangles are replaced by one of them, one of two competing phenomena can be seen: (i) local disappearance of the triangle that has not been replaced; (ii) a 3-D rotation of a rigid triangle from one location to another. The dependence of the first sensation on the duration of the interstimulus interval and on monocular availability of the local-identity information confirms previous evidence about the existence of two processes which mediate apparent motion. The localization of the high-level process, however, is questioned on the basis of a comparison of data obtained when motion information was monoptic or only dichoptic. As monoptic motion information seems more effective, the ‘figural’ higher process cannot be wholly located at the binocular level.

Visual interpolation is not scale invariant.

According to the scale-dependence hypothesis, the visual interpolation of contour fragments depends on the retinal separation of endpoints: as the retinal size of a partially occluded angle increases, the interpolated contour gradually deviates from the shortest connecting path and approaches the shape of the unoccluded angle. In the field model, as the retinal size increases the strength of good continuation increases while the strength of the minimal-path tendency decreases. To test the scale-dependence hypothesis--as well as other hypotheses connected to inclusion, support-ratio dependence, and extended relatability--we ran two experiments using the probe localization technique. Stimuli were regular polygons with rectilinear contours bounding symmetrically occluded angles. Retinal size was manipulated by changing viewing distance. Observers were asked to judge if a probe, briefly superposed on the occlusion region, was inside or outside the amodally completed angle. Retinal size strongly influenced the penetration of interpolated trajectories in the predicted direction. However, support ratio and interpolated angle size interacted with retinal size, consistently with the idea that unification factors are effective within a spatial window. We modified the field model to include the size of such a window as a new parameter and generated model-based trajectories that fitted empirical data closely.

Amodal Completion and Illusory Figures: An Information-Processing Analysis

An illusory figure is an occluding figure. In all compelling instances, perception of the illusion occurs together with amodal completion of the inducing elements. However, it is presently uncertain whether or not this relation should be interpreted causally. The purpose of this chapter is to look at completion from an information-processing point of view. A better understanding of the completion process might help clarify its role in the perception of illusory figures and give us some insight into how these illusions are produced within the visual system.

Body actions change the appearance of facial expressions.

Perception, cognition, and emotion do not operate along segregated pathways; rather, their adaptive interaction is supported by various sources of evidence. For instance, the aesthetic appraisal of powerful mood inducers like music can bias the facial expression of emotions towards mood congruency. In four experiments we showed similar mood-congruency effects elicited by the comfort/discomfort of body actions. Using a novel Motor Action Mood Induction Procedure, we let participants perform comfortable/uncomfortable visually-guided reaches and tested them in a facial emotion identification task. Through the alleged mediation of motor action induced mood, action comfort enhanced the quality of the participant’s global experience (a neutral face appeared happy and a slightly angry face neutral), while action discomfort made a neutral face appear angry and a slightly happy face neutral. Furthermore, uncomfortable (but not comfortable) reaching improved the sensitivity for the identification of emotional faces and reduced the identification time of facial expressions, as a possible effect of hyper-arousal from an unpleasant bodily experience.

Referring in multimodal systems: the importance of user expertise and system features

This paper empirically investigates how humans use reference in space when interacting with a multimodal system able to understand written natural language and pointing with the mouse. We verified that user expertise plays an important role in the use of multimodal systems: experienced users performed 84% multimodal inputs while inexpert only 30%. Moreover experienced are able to efficiently use multimodality shortening the written input and transferring part of the reference meaning on the pointing. Results showed also the importance of the system layout: when very short labels (one character) are available users strongly adopt a redundant reference strategy, i.e. they referred to the object in a linguistic way and use pointing too. Starting from these facts some guidelines for future multimodal systems are suggested.