Encoding Shape and Spatial Relations: The Role of Receptive Field Size in Coordinating Complementary Representations

An effective functional architecture facilitates interactions among subsystems that are often used together. Computer simulations showed that differences in receptive field sizes can promote such organization. When input was filtered through relatively small nonoverlapping receptive fields, artificial neural net-works learned to categorize shapes relatively quickly; in contrast, when input was filtered through relatively large overlapping receptive fields, networks learned to encode specific shape exemplars or metric spatial relations relatively quickly. Moreover, when the receptive field sizes were allowed to adapt during learning, networks developed smaller receptive fields when they were trained to categorize shapes or spatial relations, and developed larger receptive fields when they were trained to encode specific exemplars or metric distances. In addition, when pairs of networks were constrained to use input from the same type of receptive fields, networks learned a task faster when they were paired with networks that were trained to perform a compatible type of task. Finally, using a novel modular architecture, networks were not preassigned a task, but rather competed to perform the different tasks. Networks with small nonover-lapping receptive fields tended to win the competition for categorical tasks whereas networks with large overlapping receptive fields tended to win the competition for exemplar/metric tasks.     

The cutaneous saltatory area and its presumed neural basis

The saltatory area, defined as that region of the skin within which two successive and spatinily separated signals will interact to produce apparent mislocalization of the prior signal, is measured by a campimetric technique applied to the palm, finger pad, and volar forearm. As contrasted with earlier measures on larger anatomical units, saltatory areas of the hand region are considerably compressed. Comparisons between cortical receptive fields and saltatory areas lead to speculation concerning the operation of the neural principle of “magnification” in the establishment of saltatory limits. The correspondence of configurational properties of both receptive fields and saltatory areas suggests a commonality of principles underlying both, but demands more information on the temporal dimensions of cortical interaction to confirm the parallelism.     

Processing spatial relations with different apertures of attention

Neuropsychological studies suggest the existence of lateralized networks that represent categorical and coordinate types of spatial information. In addition, studies with neural networks have shown that they encode more effectively categorical spatial judgments or coordinate spatial judgments, if their input is based, respectively, on units with relatively small, nonoverlapping receptive fields, as opposed to units with relatively large, overlapping receptive fields. These findings leave open the question of whether interactive processes between spatial detectors and types of spatial relations can be modulated by spatial attention. We hypothesized that spreading the attention window to encompass an area that includes two objects promotes coordinate spatial relations, based on coarse coding by large, overlapping, receptive fields. In contrast, narrowing attention to encompass an area that includes only one of the objects benefits categorical spatial relations, by effectively parsing space. By use of a cueing procedure, the spatial attention window was manipulated to select regions of differing areas. As predicted, when the attention window was large, coordinate spatial transformations were noticed faster than categorical transformations; in contrast, when the attention window was relatively smaller, categorical spatial transformations were noticed faster than coordinate transformations. Another novel finding was that coordinate changes were noticed faster when cueing an area that included both objects as well as the empty space between them than when simultaneously cueing both areas including the objects while leaving the gap between them uncued.     

The shape of inhibitory fields in the human visual system

Five point sources of light were displayed sequentially in a horizontal line, with .45 in. of separation between each point, employing a computer-based cathode-ray tube display system. If a particular display order and appropriate display rate is employed, then the first two points being displayed will not be perceived. By systematically varying the spatial location of these two blanked points relative to the three unblanked points, a start is made towards mapping the shape of visual inhibitory fields in man, and this general technique is related to the work of Hartline and of Ratliff on lateral inhibition and to that of Hubel and Wiesel with receptive fields in the cat and monkey striate cortex.     

Relating developments in children's counterfactual thinking and executive functions

The performance of 93 children aged 3 and 4 years on a battery of different counterfactual tasks was assessed. Three measures: short causal chains, location change counterfactual conditionals, and false syllogisms—but not a fourth, long causal chains—were correlated, even after controlling for age and receptive vocabulary. Children's performance on our counterfactual thinking measure was predicted by receptive vocabulary ability and inhibitory control. The role that domain general executive functions may play in 3- to 4-year olds' counterfactual thinking development is discussed.     

Gestalt Issues in Modern Neuroscience

We present select examples of how visual phenomena can serve as tools to uncoverbrain mechanisms. Specifically, receptive field organization is proposed as a Gestalt-like neural mechanism of perceptual organization. Appropriate phenomena, such as brightness and orientation contrast, subjective contours, filling-in, and aperture-viewed motion, allow for a quantitative comparison between receptive fields and their psychophysical counterparts, perceptive fields. Phenomenology might thus be extended from the study of perceptual qualities to their transphenomenal substrates, including memory functions. In conclusion, classic issues of Gestalt psychology can now be related to modern “Gestalt psychophysics” and neuroscience.     

Neither strong nor weak space constancy is coded in striate cortex

Space constancy (the perception that the world remains stable despite eye movements) might be mediated by receptive fields with constant locations in the world rather than constant locations on the retina. In a strong form the hypothesis, receptive fields of single neurons would compensate for all eye movements, whereas in a weak form they would compensate only for the small movements of fixation that occur without retargeting. Early receptive field mapping confounded world-centered and retina-centered organizations by fixing the retina relative to the world. This paper links the strong and weak forms of constancy by relating previously isolated data, and illustrates the issues with a previously unpublished figure showing that striate receptive field have constant retinal locations. The first tests of space constancy in striate cortex contradicted the strong hypothesis; recent tests contradict the weak hypothesis, also. The results are consistent with modern theories of space constancy, differentiating between motor-oriented and perceptual functions of vision, and differentiating between constancy during pursuit and constancy across saccades. Received: 5 January 1998 / Accepted: 16 July 1998     

Directional harmonic theory: a computational Gestalt model to account for illusory contour and vertex formation

Visual illusions and perceptual grouping phenomena offer an invaluable tool for probing the computational mechanism of low-level visual processing. Some illusions, like the Kanizsa figure, reveal illusory contours that form edges collinear with the inducing stimulus. This kind of illusory contour has been modeled by neural network models by way of cells equipped with elongated spatial receptive fields designed to detect and complete the collinear alignment. There are, however, other illusory groupings which are not so easy to account for in neural network terms. The Ehrenstein illusion exhibits an illusory contour that forms a contour orthogonal to the stimulus instead of collinear with it. Other perceptual grouping effects reveal illusory contours that exhibit a sharp corner or vertex, and still others take the form of vertices defined by the intersection of three, four, or more illusory contours that meet at a point. A direct extension of the collinear completion models to account for these phenomena tends towards a combinatorial explosion, because it would suggest cells with specialized receptive fields configured to perform each of those completion types, each of which would have to be replicated at every location and every orientation across the visual field. These phenomena therefore challenge the adequacy of the neural network approach to account for these diverse perceptual phenomena. I have proposed elsewhere an alternative paradigm of neurocomputation in the harmonic resonance theory (Lehar 1999, see website), whereby pattern recognition and completion are performed by spatial standing waves across the neural substrate. The standing waves perform a computational function analogous to that of the spatial receptive fields of the neural network approach, except that, unlike that paradigm, a single resonance mechanism performs a function equivalent to a whole array of spatial receptive fields of different spatial configurations and of different orientations, and thereby avoids the combinatorial explosion inherent in the older paradigm. The present paper presents the directional harmonic model, a more specific development of the harmonic resonance theory, designed to account for specific perceptual grouping phenomena. Computer simulations of the directional harmonic model show that it can account for collinear contours as observed in the Kanizsa figure, orthogonal contours as seen in the Ehrenstein illusion, and a number of illusory vertex percepts composed of two, three, or more illusory contours that meet in a variety of configurations.     

基于坐标的脑成像元分析方法:激活似然性评估

随着脑成像研究数量的增长,研究者迫切需要一种能够将众多研究进行整合的元分析方法。本文在简单回顾了以往脑成像元分析方法的基础上,介绍了激活似然性评估法(ALE)的操作实施、优势特征、更新完善,以及其发展与应用。该元分析方法将每一个激活的坐标看作一种分布的可能性,通过特定的算法与拟合获得激活似然值,最后,将不同文献汇聚计算而来的ALE值与虚无分布获得的ALE值进行对比,以确定激活的可能性分布。近十年来,激活似然性评估法几经修正,已被广泛运用于各领域元分析的研究中。     

Semantic and formational clustering in deaf and hearing subjects’ free recall of signs

To investigate whether formational properties of sign language are used spontaneously to organize long-term memory, 16 deaf college students were given a free recall task with items that could be categorized either by shared semantic category or by shared sign language hand shape. Both presentation and response modes (signed or written) were varied between subjects. Analyses revealed no effects of mode on trials to criterion or number of items recalled at 1 week. The clustering that occurred was exclusively semantic, with significantly higher clustering scores during acquisition trials in subjects required to sign their responses. In Experiment 2, formational clustering was encouraged by including formational similarity as the only experimenter-defined basis of categorization, by increasing formational similarity within categories, and by testing only subjects with high signing skills. Input and output modes were again varied between subjects. Subjects were deaf college students with deaf parents (n = 10) or hearing parents (n = 16), and hearing adults with deaf parents (n = 8). Again, spontaneous clustering by formational similarity was extremely low. In only one case— deaf subjects with hearing parents given signed input—did formational clustering increase significantly across the eight acquisition trials. After the categorical nature of the list was explained to subjects at a 1-week retention session, all groups clustered output by formational categories. Apparently, fluent signers do have knowledge of the formational structure of signs, but do not spontaneously use this knowledge as a basis of mnemonic organization in long-term memory.     

Independent component analysis: an introduction

Independent component analysis (ICA) is a method for automatically identifying the underlying factors in a given data set. This rapidly evolving technique is currently finding applications in analysis of biomedical signals (e.g. ERP, EEG, fMRI, optical imaging), and in models of visual receptive fields and separation of speech signals. This article illustrates these applications, and provides an informal introduction to ICA.     

Sharing-rule and detection of free-riders in cooperative groups: Evolutionarily important deontic reasoning in the Wason Selection task

Taking a Darwinian approach, we propose that people reason to detect free-riders on the Wason Selection task with the sharing-rule; If one receives the resource, one is an in-group member (standard), or If one is an in-group member, one receives the resource (switched). As predicted, taking the resource-provider's perspective, both undergraduates and children (11 to 12 years old) checked for the existence of out-group members taking undeserved resource. Changing the perspective to that of the resource-recipient did not alter the selection pattern in undergraduates, although the prediction was that another type of free-riding—failure to share by resource-provider—would be checked as well. However, by removing confounding factors in the materials, both undergraduates and children checked for both types of free-riding, which fully supports the prediction. These results indicate that the sharing-rule elicits a thematic content effect that cannot be explained by preceding deontic reasoning theories.     

An investigation of trained neural networks from a neurophysiological perspective

The application of theoretical neural networks to preprocessed images was investigated with the aim of developing a computational recognition system. The neural networks were trained by means of a back-propagation algorithm, to respond selectively to computer-generated bars and edges. The receptive fields of the trained networks were then mapped, in terms of both their synaptic weights and their responses to spot stimuli. There was a direct relationship between the pattern of weights on the inputs to the hidden units (the units in the intermediate layer between the input and the output units), and their receptive field as mapped by spot stimuli. This relationship was not sustained at the level of the output units in that their spot-mapped responses failed to correspond either with the weights of the connections from the hidden units to the output units, or with a qualitative analysis of the networks. Part of this discrepancy may be ascribed to the output function used in the back-propagation algorithm.     

Model of visual adaptation and contrast detection

A three-component model of spatial vision is proposed, consisting of (1) a feedback stage, (2) a feedforward stage, (3) a threshold detector. The components correspond to physiological processes; in particular, the feedforward control signal corresponds to the “surround’s” signal in the receptive fields of retinal ganglion cells. The model makes appropriate qualitative predictions of: (l)a square-root law (Δl ∞ l^1/2) for detection at low luminances, (2) a Weber law (Δl ∞ l) at high luminances, (3) additivity of threshold masking effects at high background luminances, (4) receptive fields that, in the dark, consist only of an excitatory center and that, in the light, also contain inhibitory surrounds, (5) the variation of spatial characteristics of receptive fields depending on the temporal characteristic of the test stimulus used to measure them, (6) the subjective appearance of Mach bands, (7) sine-wave contrast-threshold transfer functions, (8) the frequent failure of disk-detection experiments to demonstrate inhibitory surrounds, and (9) various second-order threshold effects, such as reduced spatial integration for long-duration stimuli, reduced temporal integration for large-area stimuli, and the increased effect of background luminance on the detection of large-area stimuli. Predictions are improved by assuming there exist various sizes of receptive fields that determine thresholds jointly.     

Categorical versus coordinate spatial relations: computational analyses and computer simulations.

Results of 4 sets of neural network simulations support the distinction between categorical and coordinate spatial relations representations: (a) Networks that were split so that different hidden units contributed to each type of judgment performed better than unsplit networks; the reverse was observed when they made 2 coordinate judgments. (b) Both computations were more difficult when finer discriminations were required; this result mirrored findings with human Ss. (c) Networks with large, overlapping "receptive fields" performed the coordinate task better than did networks with small, less overlapping receptive fields, but vice versa for the categorical task; this suggests a possible basis for observed cerebral lateralization of the 2 kinds of processing. (d) The previously observed effect of stimulus contrast on this hemispheric asymmetry could reflect contributions of more neuronal input in high-contrast conditions.     

Neural Network Models as Evidence for Different Types of Visual Representations

Cook (1995) criticizes the work of Jacobs and Kosslyn (1994) on spatial relations, shape representations, and receptive fields in neural network models on the grounds that first-order correlations between input and output unit activities can explain the results. We reply briefly to Cook's arguments here (and in Kosslyn, Chabris, Marsolek, Jacobs & Koenig, 1995) and discuss how new simulations can confirm the importance of receptive field size as a crucial variable in the encoding of categorical and coordinate spatial relations and the corresponding shape representations; such simulations would testify to the computational distinction between the different types of representations.     

Unconscious Communication in the Intersubjective Analytic Field at Times of Separation,Loss, and Termination

ABSTRACT

Unconscious communication, like transference-countertransference, is ubiquitous in life and in the psychoanalytic process. Regardless of a clinician’s theoretical perspective, and despite differences in clinical technique, Freud’s advice to turn our unconscious to the patient’s unconscious “like a receptive organ” has guided generations of analysts toward deeper exploration of the countertransference in the intersubjective analytic field (Freud, 1912a, p. 115). In this clinical article, the recognition and use of unconscious communication, from the ordinary to the more extraordinary or uncanny, is described at moments of separation as harbinger of loss and, ultimately, termination. Such moments hold potential for a depth of emotional resonance with and accessibility to our patient’s psychic realities that may otherwise be unavailable due to our systemic defense against a shared existential anxiety that all things come to an end. The emergence of unconscious communication via the analyst’s reverie and dreams are considered an opening of potential space where ending can be conceived as a bearable thought—a transitional organizing experience for the dyad.     

THE CHRISTIAN'S DILEMMA: ORGANICISM OR MECHANISM?

Is organicism inherently Christian‐friendly, and for that matter, is mechanism inherently religion nonfriendly? They have tended to be, but the story is much more complicated. The long history of the intertwined metaphors of nature taken as an organism, versus that of nature as a machine, reveals that both metaphors have flourished in the endeavors of philosophers, scientists, and persons of faith alike. Different kinds of Christians have been receptive to both organicist and mechanistic models, just as various kinds of nonreligious scientists have been receptive to both holistic and machine metaphors. Although, it is true, organicism has been generally more attractive to persons of faith than mechanism (and vice versa), an overview of the rich and varied history of allegiances to these metaphors—religious and nonreligious alike—shows that debate is much more interesting and complex. A brief inspection of conversation surrounding recent scientific discoveries shows that this debate between metaphors is still very much alive today.     

Representation of visuotactile space in the split brain

Recent neurophysiological research in the monkey has revealed bimodal neuronal cells with both tactile receptive fields on the hand and visual receptive fields that follow the hands as they move, suggesting the existence of a bimodal map of visuotactile space. Using a cross-modal congruency task, we examined the representation of visuotactile space in normal people and in a split-brain patient (J.W.) as the right arm assumed different postures. The results showed that the congruency effects from distracting lights followed the hand around in space in normal people, but failed to do so in the split-brain patient when the hand crossed the midline. This suggests that cross-cortical connections are required to remap visual space to the current hand position when the hand crosses the midline.