Showing papers on "Perceptual learning published in 1992"

How to build a baby: II. Conceptual primitives.

Abstract: A mechanism of perceptual analysis by which infants derive meaning from perceptual activity is described. Infants use this mechanism to redescribe perceptual information into image-schematic format. Image-schemas create conceptual structure from the spatial structure of objects and their movements, resulting in notions such as animacy, inanimacy, agency, and containment. These earliest meanings are nonpropositional, analogical representations grounded in the perceptual world of the infant. In contrast with most perceptual processing, which is not analyzed in this fashion, redescription into image-schematic format simplifies perceptual information and makes it potentially accessible for purposes of concept formation and thought. In addition to enabling preverbal thought, image-schemas provide a foundation for language acquisition by creating an interface between the continuous processes of perception and the discrete nature of language.

Self-organizing neural network that discovers surfaces in random-dot stereograms

Abstract: The standard form of back-propagation learning is implausible as a model of perceptual learning because it requires an external teacher to specify the desired output of the network. We show how the external teacher can be replaced by internally derived teaching signals. These signals are generated by using the assumption that different parts of the perceptual input have common causes in the external world. Small modules that look at separate but related parts of the perceptual input discover these common causes by striving to produce outputs that agree with each other. The modules may look at different modalities (such as vision and touch), or the same modality at different times (for example, the consecutive two-dimensional views of a rotating three-dimensional object), or even spatially adjacent parts of the same image. Our simulations show that when our learning procedure is applied to adjacent patches of two-dimensional images, it allows a neural network that has no prior knowledge of the third dimension to discovery depth in random dot stereograms of curved surfaces.

Motion sickness and perception: a reappraisal of the sensory conflict approach.

Abstract: This review examines the role of activity and perceptual learning in motion sickness by means of a survey of the two kinds of recent research relevant to this topic The first is a body of literature concerned not with motion sickness as such, but with perception of orientation and self-motion under the conditions of ‘sensory conflict’ which are thought to provoke motion sickness The second consists of investigations into the prediction and prevention of motion sickness itself A major weakness is identified in the methodologies employed in both types of research: namely, a neglect of the way in which responses to unusual and disorienting environments, whether nauseogenic or not, may be affected by the activities, skills and strategies of the perceiver New directions are outlined for future research into immediate reactions and longer-term adaptation to such environments

Active and exploratory perception

Abstract: The main goal of this paper is to show that there is a natural flow from active perception through exploration to perceptual learning. W have attempted to conceptualize the perceptual process of an organism that has the top-level task of surviving in an unknown environment. During this conceptualization process. Four necessary ingredients have emerged for either artificial or biological organisms. First, the sensory apparatus and processing of the organism must be active and flexible. Second, the organism must have exploratory capabilities. Third, the organism must be selective in its data acquisition process. Fourth, the organism must be able to learn. In the section on learning, we have clearly delineated the difference between what must be innate and what must be learned. In order to test our theory, we present the system's architecture that follows from the perceptual task decomposition. The predictions of this theory are that an artificial system can explore and learn about its environment modulo its sensors, manipulators, end effectors, and exploratory procedures/attribute extractors. It can describe its world with respect to the built-in alphabet, that is, the set of perceptual primitives.

Learning to See Random-Dot Stereograms:

Abstract: In the present study some specific properties of the learning effects reported for random-dot stereograms are examined. In experiment 1 the retinal position-specific learning effect was reproduced and in a follow-up experiment it was shown that the position specificity of learning can be accounted for by selective visual attention. In experiments 2 and 3 evidence was obtained that suggests that observers can learn, to a certain degree, monocular random-dot patterns and that this learning facilitates the depth percept. This result indicates that the traditional belief that random-dot stereograms are devoid of monocularly recognizable or useful forms should be reconsidered. In the second set of experiments the learning of two binocular surface properties of random-dot stereograms, depth edges and internal depth regions, was investigated. It was shown in experiment 4 that the depth edges of random-dot stereograms are not learned, whereas the results of experiment 5 indicate that the internal depth regions are learned. Finally, in experiment 6 it was shown that depth edges are learned when the internal depth regions of the stereogram are ambiguous. The results are discussed in terms of the importance of the particular type of stimulus used in the learning process and in terms of perceptual learning and attention.

Perceptual Invariances in the Comparative Psychology of Music

Abstract: If one stimulus pattern is transposed to another, and if the two are recognizably the same, then they are said to be perceptually invariant. Usually, transpositions that lead to perceptual invariance involve changes on a ratio scale between the stimuli comprising the two patterns. In this paper, we survey the literature with a view to the conditions of pitch structure (melody and harmony), spectral structure (timbre), intensity structure (loudness), and temporal structure (rhythm, meter, and tempo) that produce perceptual invariance. The review compares perceptual invariance for human infants, young children, and adults and nonhuman animals. For the most part, perceptual invariance holds at all levels of development and for all species throughout the acoustic dimensions surveyed. However, for melody perception, there is evidence that humans go through a stage in early childhood in which absolute (as distinguished from relative) pitch perception plays a role. Without doubt, absolute pitch is important in perception of serial acoustic (melodic) structures by animals. For both humans and nonhumans, melody perception appears to be governed by a hierarchy of perceptual strategies that includes both absolute and relative pitch. The survey suggests the value of a comparative perspective in understanding the perceptual principles underlying music perception in humans and the principles by which human infants and nonhuman animals process acoustic information.

Learning to categorize objects using temporal coherence

Abstract: The invariance of an objects' identity as it transformed over time provides a powerful cue for perceptual learning. We present an unsupervised learning procedure which maximizes the mutual information between the representations adopted by a feed-forward network at consecutive time steps. We demonstrate that the network can learn, entirely unsupervised, to classify an ensemble of several patterns by observing pattern trajectories, even though there are abrupt transitions from one object to another between trajectories. The same learning procedure should be widely applicable to a variety of perceptual learning tasks.

Eleanor J. Gibson: Learning to perceive and perceiving to learn.

Abstract: \i Eleanor Gibson essentially defined the domain of perceptual learning, which includes both improvement in perception as a function of experience and learning and acquisition of knowledge as a function of changes in perception. In her view,differentiation, as opposed to association, is the process underlying perceptual development as well as perceptual learning. She considered perceptual development to be an important aspect of cognitive development. To a considerable degree, children's acquisition of knowledge and their increasingly complex conceptual sophistication can be attributed to their ability to detect more and more meaningful aspects of the rich stimulation impinging on them. This theoretical analysis was instantiated in empirical research on a wide range of topics. Imagine yourself a participant in an experiment. You are such confusion errors decreased with trials for participants of shown a complex graphic "scribble" consisting of a four-coil all ages. spiral. You are then shown successively a series of similar and Why is this experiment considered a classic? Consider the identic~1 drawings. Your ta~k is to pick out the identical ones in time when it was carried out. It was the early mid-fifties, in the ( the senes. The first few scnbbles seem about the same as each heyday of behaviorism with its emphasis on the association of other and the same as what you remember of the original stan- stimuli and responses (e.g., Hull and Spence) or in some cases dard. However, you gradually notice that there is variation in the association of stimuli and stimuli (e.g., Guthrie). Psycholonumber of coils in the spiral, perhaps degree of tightness of the gists interested in learning generally all agreed that learning spiral, and even direction of rotation of the spiral (clockwise vs. was the forming of associations or, at the very least, emphasized counterclockwise). You do not know if you have noticed all the the response side of learning. The scribble experiment was a ways the scribbles differed or all the particular differences simple and vivid demonstration that a change in performance among them, because the experimenter-never tells you specifi- as a function of experience did not have to involve the formacally whether you are right or wrong. However, at the end of the tion of associations between stimuli and responses but could series the experimente~ asks you to repeat the procedure, giving consist of improvement in perception. Moreover, this improveyou another op~rtum~y to l~ok.at the standard, and the~ to go ment in perception occurred without reinforcement in the through the senes again. ThIS tIme you are more certaIn that sense of drive reduction or even reinforcement in the sense of you have noti~ed the types of difference among the scri?bles, extrinsic information about correctness (i.e., knowledge of reand you are fal~ly su:e you have detected most of the scnbbles suIts). What was required was merely the opportunity to examthat were not. IdentIcal to the standard. Once ~ore 7"°u ar.e ine and to study the stimuli. What was learned was not a new ~sked to ~xamme the standar? a~d to select t~ose IdentIcal to It association, but rather how the stimuli differed from one anm the senes. At the end of this tnal the expenmenter tells you, other. Besides the quantitative data about errors from the experfinally, that you have now gotten them all correct and thanks . t th t t f th b. t . II

Chapter 5 Adaptive Eye-Hand Coordination: Implications of Prism Adaptation for Perceptual-Motor Organization

Abstract: Publisher Summary Perceptual-motor adaptability is certainly one of the more salient and important human capacities The most obvious example of such adaptability is skilled performance, ranging from primitive athletic and linguistic abilities to more modern skills, such as machine control and performance in alien work environments Less obvious, but in many ways more fundamental, is the ability to recover from systematic encoding dysfunctions arising from growth, pathology, or drift that misrepresent stimulus information and therefore, specify inappropriate responses The study of such perceptual adaptation is commonly called “prism adaptation” Skill learning and perceptual adaptation seem to be fundamentally different processes The chapter illustrates a system analysis comparison of the two processes The stimulus for skill learning is an objective change or temporal sequence of changes in the distal stimulus while the stimulus for perceptual adaptation is a systematic distortion such that the relationship between distal and proximal stimuli is changed in some important respect Skill learning consists in determining the parameters of the changing stimulus (perceptual learning) and modifying motor programs (motor learning) such that the response becomes spatially congruent with the stimulus

Reading skills and auditory processing ability in children with chronic otitis media in early childhood.

Abstract: The effect of recurrent middle ear disease during the first 2 years of life on auditory perceptual skills and reading ability was examined in two groups of 6- and 7-year-old children who were pair-matched by age, gender, socioeconomic status, and receptive vocabulary. Group 1 consisted of children with documented chronic otitis media at an early age, and group 2 had no history of middle ear problems. Tests of auditory perceptual skills and reading ability were administered. Significant differences in performance on all tests of auditory processing ability and reading ability were noted.

The Effectiveness of Vocabulary Visual Aids for Auditory and Visual Foreign Language Students

Abstract: A classroom experiment involving seventy-nine first semester Spanish students at the University of Central Arkansas measured the relationship between foreign language vocabulary visual aids, vocabulary achievement scores, and student modality strengths. The objective was to determine whether students of different perceptual learning styles (e.g., auditory and visual) would achieve differential scores on a multiple-choice test over a three-part foreign language (Swahili) vocabulary presentation using pictorial, verbal, and combination pictorial-verbal aids. The results showed significant positive correlation between the visual parameter as measured by the Swassing-Barbe Modality Index and scores on vocabulary items presented and tested with combination pictorial-verbal aids. Conversely, there was a negative correlation between the auditory parameter and the combination pictorial-verbal vocabulary section. Median scores on the pictorial and verbal sections were too close for conclusions to be drawn regarding the efficiency of those visual aids for different learning styles. However, the overall results suggest that further investigation will prove useful in devising the best combinations of visual aids, particularly for the predominantly visual or auditory student. A serendipitous finding was that the auditory students in this study had generally lower grades and ACT scores, and may be most benefited by individualized study aids.

Role of context in perceptual learning in maze discriminations.

Abstract: Three experiments with rats in a maze examined the effects of pre-exposure to the relevant discriminative stimuli (rubber and sandpaper-covered maze arms) or the extra-maze context (the maze was surrounded either by black curtains or by variety of extra-maze landmarks) on the learning of a discrimination between rubber and sandpaper arms. In Experiment 1, pre-exposure to the extra-maze context facilitated subsequent discrimination learning. Experiments 2 and 3 showed that pre-exposure to rubber and sandpaper arms facilitated subsequent discrimination learning only when these cues were presented in the same context during pre-exposure and discriminative training. Taken together, the results are consistent with the hypothesis that a major cause of perceptual learning is the latent inhibition of stimuli or features common to the two discriminative stimuli, and that such latent inhibition may be disrupted by a radical change of context.

Learning of visual modules from examples: a framework for understanding adaptive visual performance

Abstract: Networks that solve specific visual tasks, such as the evaluation of spatial relations with hyperacuity precision, can be easily synthesized from a small set of examples. The present paper describes a series of simulated psychophysical experiments that replicate human performance in hyperacuity tasks. The experiments were conducted with a detailed computational model of perceptual learning, based on HyperBF interpolation. The success of the simulations provides a new angle on the purposive aspect of human vision, in which the capability for solving any given task emerges only if the need for it is dictated by the environment. We conjecture that almost any tractable psychophysical task can be performed better after suitable training, provided the necessary information is available in the stimulus.

Learning sequences in an associative network: a step towards cognitive structure

Abstract: A crucial step towards the representation of structured, symbolic knowledge in a connectionist system is an associative, neurophysiologically plausible theory of sequence learning. Learning internal representations for sequences is important on adaptive as well as theoretical grounds, since it provides the capability of anticipating probable next states of the world. Through computer simulation of a connectionist, neurocognitive model, this work begins the development of a theory of sequence learning. Sequence learning is characterized as (1) the formation of an internal representation from a series of perceptual events experienced in order over time, and (2) the use of this internal representation to predict events in the absence of supporting environmental input. Development of the model proceeds in a failure-driven fashion, wherein modifications and new mechanisms are introduced to correct specific shortcomings revealed in the simulations. The starting point is a network architecture designed for perceptual learning of compact representations for features and objects. Units of the network do not model single neurons but cell assemblies: multi-neuron groups which exhibit a persistent and complex time course of activity. Unit dynamics are modelled with a system of difference equations which compute state trajectories for three unit state variables, with activation being the primary variable. At any time the content represented by a unit is either latent (no activation), conscious (peak activation), or held in immediate memory (intermediate activation). A modified Hebbian learning rule is introduced which strengthens connection weights in response to consecutive, rather than concurrent, activity in connected units. The importance of controlling the activity generated by positive feedback within and between units is emphasized and specific dynamic control functions are elucidated. Using simulation and conceptual analysis, a set of complimentary control mechanisms, each satisfying a particular functional requirement, is identified. Together these mechanisms provide effective control of activity which permits training and testing the system on sequence learning tasks. A series of simulations then demonstrate elementary learning of single and intersecting sequences. It is concluded that an associative explanation of sequence learning is possible within a neurophysiologically plausible connectionist architecture.

Selective attention to perceptual groups: the phase tracking mechanism

Abstract: We introduce a new mechanism of selective attention among perceptual groups as part of a computational model of early vision. In this model, selection of objects is a two-stage process: perceptual grouping is first performed in connectionist networks which dynamically bind together the neural activities triggered in response to related features in the image; secondly, by locking its output on the quasi-periodic bursts of activity associated with a single perceptual group, a system called the phase-tracker produces a temporal filter which retains the selected group for further processing while rejecting the unattended ones. The properties of our model are investigated in a number of simulations and interpreted in the context of current theories of visual attention developed in psychology.

Structure and Process in Perceptual Organization

Abstract: Publisher Summary This chapter presents the theoretical concepts of structure and process for a better understanding of the perceptual organization/cognitive system. A particularly stubborn and enduring issue in the psychology of perception is the way in which perception might be organized—the primacy of wholes versus parts. Two basic positions on this topic can be traced back to the controversy between two schools of perceptual thought: structuralism and gestalt. The chapter elaborates on the growing usage of the term “wholistic” rather than “analytic” to describe perception. The relationship between the conceptual structure/process distinction and process models is that the stimulus structure and processing aspects, as well as their mutual interaction, need to be represented in any complete information-processing model of the perceptual/cognitive system. That is, analysis guided by the conceptual distinction between structure and process provides useful constraints on what aspects of human information processing must be embodied in the process model. However, it need not predetermine how those aspects are to be represented in the model. A clear notion of the perceptual structure of the stimulus is an important prerequisite for asking meaningful questions about processing. As a general rule, it is important to consider the perceptual structure of the stimuli used to test any processing hypothesis. Stimulus structure and organismic mode of processing must be considered together for understanding the performance on a psychological task.

Part 2: Perceptual and cognitive training: its nature and importance

Abstract: Infants who see are abundantly stimulated by the luminous energy that is, by far, the best source of the spatial information they need for proper perceptual and cognitive development. Learning to attend to this stimulus energy is not a problem for them; it is ubiquitous and not easily ignored. On the other hand the stimulation experienced by infants who cannot see is not as informative and lacks the

Are artificial neural nets as smart as a rat

Abstract: The origins of artificial neural networks are related to animal conditioning theory: both are forms of connectionist theory which in turn derives from the empiricist philosophers' principle of association. The parallel between animal learning and neural nets suggests that interaction between them should benefit both sides. The paper examines relationships between neural networks and a range of key ideas and findings in modern learning theory. It draws on studies of both conditioning and perceptual learning. The need to avoid simplistic comparisons is stressed. Not all issues which have aroused interest in learning theory are relevant to neural net research, because old and new connectionism diverge in some important ways. It is also necessary to recognize that many learning phenomena do not lend themselves to simulation by a single net. However, once these points are recognized, the findings of learning theory provide a range of well-defined challenges which are potentially important for those who are con...

Self-Perception and Self-Conception

Abstract: Butterworth's article addresses a topic that is both interesting and important: the origins of self-knowledge. Using some of J. J. Gibson's ideas, he argues that awareness of the self begins very early in life and is based in direct perception of the self (proprioception) through all modalities. He goes on to suggest that perceptual specification of the self may form the basis for a later self-concept. This is an important idea, because it acknowledges the continuity between direct perception and other human modes of knowing, including those that involve symbolizing. Certainly, one of the persistent problems of J. J. Gibson's ecological approach to perception is that of accounting for phenomena that are usually called conceptual (cf. Pick & Heinrichs, 1989). Thus, I welcome Butterworth's attempt to bridge the developmental gap between early awareness of the self (as object and actor) and more sophisticated understanding of the self. I also welcome his emphasis on the infant's ability to perceive the self directly in meaningful ways. Nevertheless, although I am quite in sympathy with this general approach, I am puzzled by some aspects of Butterworth's argument linking selfperception with self-concept. Through his review of research on infant development, Butterworth demonstrates very nicely that even the youngest infants give evidence of self-perception in several modalities. Unfortunately, thereafter his argument is in some respects inconsistent with the premises he seems to adopt in his discussion of J. J. Gibson's theory and its application to self-awareness. There is a shift in viewpoint that is difficult to reconcile: Whereas some primitive, bodily aspects of the self are supposed to be directly perceived, the "meaningful" aspects of the self are dealt with by means of representations. For example, having agreed with Gibson about the earliest basis for self-awareness, he goes on to accept, uncritically, that mirror behaviors such as those exhibited in the markedface task with young children, must necessarily be explained by some "higher level" mechanism (i.e., self-concept, selfrepresentations, etc.). Similarly, in his discussion of movingroom studies with infants, he seems to say that the infant falls down because misleading, perceptual information has not yet been overcome by "objective conceptual knowledge of self." In general, his argument is that the infant's developmental task is to replace reliance on bad, misleading perceptual information with good, conceptual knowledge. For several reasons, I do not agree. I do not see why perceptual information is good and veridical for specifying the self as actor in the environment, but bad and misleading for specifying the self in other ways. This seems to me to be a version of the ancient and by now somewhat worn-out distinction between appearance and reality, which assumes that perception is not veridical, because it is based on shifting and subjective sensations; it thus also assumes that perception is indirect. J. J. Gibson (1966, 1979/1986) argued that perception is always veridical, but that perceivers sometimes do not pick up all the information they need and so are unable to act effectively. This seems to be the case for the infants in the moving room, who perceive a change in their own orientation based on visual information, but do not make use of other available information (e.g., kinesthetic information). Under ordinary conditions, perceptual learning takes place, and the perceiver becomes more skilled at picking up the necessary information to guide action (E. J. Gibson, 1969). This process seems to me to be essential to many of the achievements of infancy, including such things as maintaining an upright posture while moving. However, Butterworth deals quite differently with the development of increasing skill at such activities. He connects Butterworth and Cicchetti's (1978) finding that infants became more skilled with increasing age in the moving-room situation with the idea that they are beginning to "attribute an independent cause to the discrepant visual feedback." Why not simply say that perceptual learning has taken place? At the same time, Butterworth makes a statement with which I do agree: ". . . perceptual specification of self is embedded within congruent self-knowledge, in those old enough and with the cognitive capacities to have acquired it. It is not a form of information for self that we outgrow." This statement leads in a more fruitful direction, because it suggests something of the way in which perceptual understanding and conceptual understanding might be developmentally related. However, it leaves open the question of whether the relation between the two is continuous or discontinuous. Butterworth, like most others, seems to endorse the latter idea. Pick and Heinrichs (1989), by contrast, argued for the continuity of perceptual and conceptual understanding:

Concept of visual sensation in a theory of visual attention: a theoretical note.

Abstract: In "Concept of Visual Sensation" (I), I proposed a sharp dl

A Quasi-Experimental Study of 5th-Graders' Use of Selected Self-Directing Perceptions and Learning Strategies

Abstract: A major change being advocated in education is that of making students more self-directing; that is, helping them become more responsible for their own learning. The focus of this investigation was on fifth grade students' use of self-directed learning strategies and self-directed perceptual skills. An experimental study was conducted using the nested design for analyzing data obtained from the Guglielmino Self-Directed Learning Readiness Scale, the Zimmerman and Martinez-Pons1 Self-Regulated Learning Interview Schedule, and the Bradley-Lane Self-Directing Perceptual Scale. One hundred fifty-two fifth graders were involved in the eight week study along with their six teachers. Both students and teachers were immersed in a module of training that included emphasis upon self-directing behaviors and learning strategies. Two striking findings emerged; namely, (a) in comparing the average number of learning strategies acquired after treatment, the experimental group (low, middle, and high IQ levels) scored 40%, 50% and 29% higher respectively, than did the control group; (b) in comparing the fifth grade students use of learning strategies it was found that most students nearly doubled the number of learning strategies they had previously acquired. Thus, it was concluded that children who do not have actual teaching of information or data concerning learning strategies will likely never acquire the same repertoire of skills that students acquire when exposed to this critical information in some specific, systematic fashion. A primary product developed for the purposes of this investigation was the Bradley-Lane Self-Directing Perceptual Scale—a 132-item Likert Scale designed to identify the self-perceptions of elementary and middle school students. After field-testing, a chi-square treatment was applied to each item of the Perceptual Scale resulting in a reliability of p<.01 for the majority (79%) of the test items, while an additional twelve items (9%) were found to be reliable at the .02 level of significance.

Reliability of perceptual learning style preferences of chronically mentally ill adults.

Abstract: 36 chronically mentally ill adults were administered the perceptual learning style items (auditory, visual, kinesthetic) from the Productivity Environmental Preference Survey. A retest was given 6 months later. Analysis showed that, although the instrument can be reliably scored, content and time sampling error more than account for the variance in these subtests. Self-reported perceptual learning style appears to be a transient phenomenon of limited practical utility in this population.

Early perception and structural identity: neural implementation

Abstract: It is suggested that there exists a minimal set of rules for the perceptual composition of the unending variety of spatio-temporal patterns in our perceptual world. Driven by perceptual discernment of "sudden change" and "unexpectedness", these rules specify conditions (such as co-linearity and virtual continuation) for perceptual grouping and for recursive compositions of perceptual "modalities" and "signatures". Beginning with a smallset of primitive perceptual elements, selected contextually at some relevant level of abstraction, perceptual compositions can graduate to an unlimited variety of spatiotemporal signatures, scenes and activities. Local discernible elements, often perceptually ambiguous by themselves, may be integrated into spatiotemporal compositions, which generate unambiguous perceptual separations between "figure" and "ground". The definition of computational algorithms for the effective instantiation of the rules of perceptual grouping remains a principal problem. In this paper we present our approach for solving the problem of perceptual recognition within the confines of one-D variational profiles. More specifically, concerning "early" (pre-attentive) recognition, we define the "structural identity of a k-norm, k ∈ K,"--SkID--as a tool for discerning and locating the instantiation of spatiotemporal objects or events. The SkID profile also serves a s a reference coordinate framework for the "perceptual focusing of attention" and the eventual assessment of resemblance. Neural network implementations of pre-attentive and attentive recognition are also discussed briefly. Our principles are exemplified by application to one-D perceptual profiles, which allows simplicity of definitions and of the rules of perceptual composition.