DOCSLIB.ORG

Predictive Perceptual Systems

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Predictive Perceptual Systems Synthese (2018) 195:2367–2386 https://doi.org/10.1007/s11229-017-1373-4 S.I. : PREDICTIVE BRAINS Predictive perceptual systems Nico Orlandi1 Received: 5 July 2016 / Accepted: 11 March 2017 / Published online: 20 March 2017 © Springer Science+Business Media Dordrecht 2017 Abstract This article attempts to clarify the commitments of a predictive cod- ing approach to perception. After summarizing predictive coding theory, the article addresses two questions. Is a predictive coding perceptual system also a Bayesian system? Is it a Kantian system? The article shows that the answer to these questions is negative. Keywords Predictive coding · Bayes · Perception · Kant · Expectations · Error · Free-energy What does a predictive coding approach to perception (or PCP) commit us to? This article suggests that predictive coding is primarily a theory of information transmis- sion. It is a theory that, in cognitive science, is geared at explaining how perceptual and sensory parts of the brain transmit information in a way that is efficient for the guidance of action. The article questions some additional commitments that the theory is assumed to have. The focus is on two such commitments. First, although PCP and Bayesian accounts of the mind and brain are usually presented hand in hand, I suggest that predictive coding and Bayesianism do not necessarily go together. A predictive coding perceptual system need not also be a Bayesian system. Second, some propo- nents of predictive coding interpret the theory as putting a Kantian and intellectualist gloss on mental processes (Gładziejewski 2016; Hohwy 2013). I explain, instead, This article develops themes from Orlandi 2014. I thank Jona Vance and Martin Thomson-Jones for assistance in understanding Bayesian decision theory. Needless to say, any mistake present in the paper is my own. I also thank two anonymous referees selected by Synthese, for constructive criticism on an earlier version of this article. B Nico Orlandi [email protected] 1 UC Santa Cruz, Santa Cruz, CA, USA 123 2368 Synthese (2018) 195:2367–2386 that predictive coding is fully compatible with embedded and ecological accounts of cognition. Why focus on these two themes? Doing so hopes to deliver a better understanding of PCP. But the themes are also interesting from a historical point of view. PCP was initially developed in cognitive science in the field of active vision, and it was thought to be good news for proponents of ecological and embodied understandings of perception (Rao and Ballard 1999). It is a curious development that it would be taken up by proponents of more intellectualist accounts. This article questions this type of uptake. In Sect. 1, I present what a predictive coding framework is, and describe how it may apply to perception. In Sect. 2, I argue that predictive perceptual systems are not necessarily Bayesian systems. In Sect. 3, I discuss the relationship between predictive coding and intellectualism, showing that the posits of predictive coding are in line with an ecological way of understanding the perceptual process. 1 Predictive coding and predictive perception Suppose that two people make a plan to meet in a certain location at a certain time. Once they make the plan, they decide that they will communicate prior to the meeting only if a problem arises. In doing this, the protagonists employ an efficient technique for transmitting information. They assume the meeting will take place, and communicate only if a change is needed. In other words, they communicate only if there is error in the original plan. This style of communication is the idea behind predictive or efficient coding. Predictive coding is originally a label for an information-transmission strategy employed, for example, in television (Harrison 1952; Oliver 1952). Television requires the continuous transmission of images. One of the problems faced by engineers in the image transmission field is the overload of the transmission medium. Transmission channels have limited capacity. To avoid overload, a strategy is to have a code that transmits information in a way that gets rid of redundancy. Consider transmitting a “still” image. We can either have a signal that transmits the still twice, risking overload, or we can have a signal that transmits the still once, and then communicates that there is no change in the image from the previous transmission (Harrison 1952, p. 766). Similarly, in transmitting a changing picture, we could have a signal that informs concerning every aspect of the changing picture, or we could have a signal that only informs concerning what changes from the initial image. This type of signal does not encode regions of the changed picture that are highly predictable given the past. This would be a predictive coding signal. A predictive code informs about error or what is unexpected, rather than what is present (Clark 2013; Shi and Sun 1999). Error, expectation and prediction are thought to be three key concepts of this approach.1 1 In this first sketch of the theory, I treat expectations and predictions as separate elements. Depending on the specific example, however, there may be no difference between the two (see, predictive coding in lateral and temporal inibition below). In Sect. 3, following the insight of an anonymous referee, I put this distinction into question. 123 Synthese (2018) 195:2367–2386 2369 When an image is transmitted, an expectation is created. The expectation is that the subsequent image will be roughly similar to the previous image. This expectation may in turn create a prediction of what the following pixel values in the incoming signal should be. The prediction is then either confirmed or disconfirmed by the incoming signal, which serves as an error sign. Starting in the 1980s, the predictive coding idea was applied to understanding the response behavior of neurons in preliminary stages of sensory processing—for example, in retinas and LGNs of flies, cats and monkeys (Barlow 1981; Dan et al. 1996; Srinivasan et al. 1982). Neurons in early sensory areas display both spatial lateral inhibition, and temporal inhibition. Inhibition means that neurons can exhibit behavior that is hampering, rather than excitatory. They can inhibit neighboring neurons, and they can self-inhibit. An example of spatial lateral inhibition is “center-surround” antagonism, where neurons in a given retinal layer are organized in such a way that the center is excitatory while the surround is inhibitory. Because of this arrangement, the behavior of excitatory neurons is not simply a function of the amount of light that hits them, but also a function of the inhibitory activity of nearby neurons. Similarly, in temporal inhibition (or self-inhibition) neurons have a “phasic” response profile. They may have an initial excitatory response, followed by a neutral or inhibitory response. A constant intensity of light may evoke virtually no sustained response other than the response due to the initial exposure. A flash of light, by con- trast, may evoke an initial excitatory response, followed by an inhibitory response (Srinivasan et al. 1982, p. 438). Several theories have been put forward to explain lateral and temporal inhibition. One hypothesis is that inhibition removes redundancy by reducing the response range of neurons, thus also enhancing their efficiency in transmitting information (Barlow 1961). Predictive coding is a framework used to spell out this idea. Lateral inhibition can be seen as the realization of a predictive code. In center-surround antagonism, surrounding neurons that have been exposed to nat- ural images generate a statistical estimate of the intensity expected at a particular point in the middle. By subtracting this best estimate (through inhibition) from the signal actually entering at the center, the center neurons communicate if there is something unexpected. In this way, redundant information is suppressed and neurons instantiate an efficient code. In temporal inhibition, a constant intensity light generates an initial response, but then virtually no response. Neurons “predict” the intensity coming in based on how they have already been excited. They subsequently stop responding unless there is a change in intensity—an error. Predictive coding is then introduced as a theory of information transmission that makes sense of the response behavior of some sensory neurons in vertebrates. Why make it into a more general idea about brains and cognitive activity? This is not always made explicit, but a number of factors likely contributed. Initially, the theory was used to understand the feedback connections between lay- ered neurons (Rao and Ballard 1999). As described so far, lateral inhibition involves the horizontal interplay of nearby neuronal elements. Inhibitory interactions, how- ever, can also occur “top-down”. The visual cortex is organized in layers of neurons 123 2370 Synthese (2018) 195:2367–2386 that often have reciprocal connections. Neurons at higher levels (for instance in V2) may send predictions down to neurons in lower levels (V1). The feedback connections serve as (inhibitory) predictions. The feedforward connections from V1 to V2 serve as error signals. Since not just the visual cortex, but the cerebral cortex in general is hierarchically structured, we can see how this theory can serve as a story of how the whole brain processes information (Rao and Ballard 1999, p. 85). Later, and perhaps more importantly, predictive coding insights were used to under- stand the effective control of action (Clark 2016, ch. 4). One example is given by skilled activity. A major puzzle in understanding how skilled activity takes place is given by information transmission problems. Information transmission problems arise because of the limitations of nerves and synapses. Such limitations produce signaling delays that would seem to impede fluid motion. In skilled (fast) reaching, for example, the brain has to receive, and respond to, a stream of proprioceptive information concerning the position and trajectory of one’s arm and hand, as well as information about the location of objects (Clark and Toribio 1994 p.
Load more

Recommended publications
  • Emulating the Perceptual System of the Brain for the Purpose of Sensor Fusion
    HSI 2008 Krakow, Poland, May 25-27, 2008 Emulating the Perceptual System of the Brain for the Purpose of Sensor Fusion Rosemarie Velik, Member, IEEE, Roland Lang, Member, IEEE, Dietmar Bruckner, Member, IEEE, and Tobias Deutsch, Member, IEEE Abstract — This work presents a bionic model derived II. STATE OF THE ART from research findings about the perceptual system of the There have already been made various attempts to merge human brain to build next generation intelligent sensor fusion systems. Therefore, a new information processing information coming from various sensory sources. The principle called neuro-symbolic information processing is research area generally first mentioned when discussing introduced. According to this method, sensory data are such tasks is the area of sensor fusion. In literature, there processed by so-called neuro-symbolic networks. The basic cannot be found only one but various definitions of the processing units of neuro-symbolic networks are neuro- term sensor fusion. Principally, sensor fusion is concerned symbols. Correlations between neuro-symbols of a neuro- with the combination of sensor data or data derived from symbolic network can be learned from examples. Perception sensory data in order to produce enhanced data in form of is based on sensor data as well as on interaction with an internal representation of the process environment. The cognitive processes like focus of attention, memory, and achievements of sensor fusion are robustness, extended knowledge. Additionally, a mechanism for evaluating perception by emotions is suggested. spatial and temporal coverage, increased confidence, reduced ambiguity and uncertainty, and improved Keywords — Humanlike Perception, Sensor Fusion, resolution [8].
    [Show full text]
  • 1 What Do Models of Visual Perception Tell Us About Visual Phenomenology?
    What do models of visual perception tell us about visual phenomenology? Rachel N. Denison1, Ned Block2, Jason Samaha3 1 Department of Psychology and Center for Neural Science, New York University 2 Department of Philosophy, New York University 3 Department of Psychology, University of California, Santa Cruz 1 Abstract Computational models of visual processing aim to provide a compact, explanatory account of the complex neural processes that underlie visual perception and behavior. But what, if anything, do current modeling approaches say about how conscious visual experience arises from neural processing? Here, we introduce the reader to four commonly used models for understanding visual computations, neural activity, and behavior: signal detection theory, drift diffusion, probabilistic population codes, and sampling. In an attempt to bridge these modeling approaches with experimental and philosophical work on the neural basis of conscious visual perception, we lay out possible relationships between the components of the models and the contents of phenomenal visual experience. We find no unique relation between model components and phenomenal experience in any model; rather, there are multiple logically possible mappings from models to experience. Going forward, we suggest that there are scientific opportunities to develop models that predict and explain a variety of subjective reports and philosophical opportunities to consider what aspects of phenomenal experience are promising scientific targets. Keywords: vision, perception, consciousness, subjective report, computational model 2 1 Introduction The science of visual perception has a long history of developing quantitative, computational models to explain and predict visual performance on a variety of tasks. These models have typically been developed to account for objective visual performance—like observer’s accuracy, reaction time, or perceptual sensitivity in discriminating different visual stimuli.
    [Show full text]
  • Chapter 7 Information Theory and Sensory Perception
    Chapter 7 Information theory and sensory perception M.D. Plumbley & S.A. Abdallah Department of Electronic Engineering, Queen Mary University of London, London, UK. Abstract In this chapter, we explore Shannon’s information theory and what it may tell us about the biological processes of sensory perception. We begin by discussing some historical theories of sensory perception, including concepts of objects and invariances as well as principles from Gestalt psychology. We consider the ecological principle that perception should be useful for an organism, and introduce information theory as a possible way to measure this usefulness. After introducing some of the key concepts from information theory, such as entropy, information, redundancy and factorial coding, we draw parallels between communication in engineering systems and sensory perception. We discuss Attneave’s early proposal that perception should create an economical description of sensory inputs, as measured by information theory, and Barlow’s suggestion of lateral inhibition as a possible mechanism to achieve this. We discuss the important role played by noise in an information processing system, and its importance when determining how to optimise the communication of information. This leads to the concept of information-optimal filters which change their characteristics at different signal-to-noise levels, a feature exhibited by fly and human visual systems. For information processed across topographic maps, optimal information processing leads to a principle of uniform information density, whereby a larger area in the cortex allows higher information density, and hence higher accuracy or sensitivity, at the corresponding sensors. We also discuss some recent investigations of information theory applied to spiking neural systems, and the concept of economy of impulses.
    [Show full text]
  • Defining the Role of Attention in Hierarchical Auditory Processing
    Review Defining the Role of Attention in Hierarchical Auditory Processing Caitlin N. Price 1,2,* and Deborah Moncrieff 1,2 1 Institute for Intelligent Systems, University of Memphis, Memphis, TN 38152, USA; [email protected] 2 School of Communication Sciences & Disorders, University of Memphis, Memphis, TN 38152, USA * Correspondence: [email protected] Abstract: Communication in noise is a complex process requiring efficient neural encoding through- out the entire auditory pathway as well as contributions from higher-order cognitive processes (i.e., attention) to extract speech cues for perception. Thus, identifying effective clinical interventions for individuals with speech-in-noise deficits relies on the disentanglement of bottom-up (sensory) and top-down (cognitive) factors to appropriately determine the area of deficit; yet, how attention may interact with early encoding of sensory inputs remains unclear. For decades, attentional the- orists have attempted to address this question with cleverly designed behavioral studies, but the neural processes and interactions underlying attention’s role in speech perception remain unresolved. While anatomical and electrophysiological studies have investigated the neurological structures contributing to attentional processes and revealed relevant brain–behavior relationships, recent elec- trophysiological techniques (i.e., simultaneous recording of brainstem and cortical responses) may provide novel insight regarding the relationship between early sensory processing and top-down at- tentional influences. In this article, we review relevant theories that guide our present understanding of attentional processes, discuss current electrophysiological evidence of attentional involvement in auditory processing across subcortical and cortical levels, and propose areas for future study that will inform the development of more targeted and effective clinical interventions for individuals with speech-in-noise deficits.
    [Show full text]
  • The Function of Perception
    The Function of Perception Peter J. Graham 1 Fitness, Functions, and Functional Analysis I believe human perceptual systems—especially visual systems—have producing reliably accurate perceptual representations as a biological function (Graham 2010 , 2012 , 2014 ). I defend this against an argument from Tyler Burge. Burge argues that perceptual states cannot have representing accurately as a biological function, for there is a “root mismatch” between representational success and failure, on the one hand, and biological success and failure, on the other. Truth and accuracy are seman- tical, not practical, matters, and biology only cares about practical matters. Representational success and failure thus cannot be biological functions of any psy- chological state or system. Burge is not alone, as many have argued that truth and accuracy cannot be biological functions. 1 I shall argue this isn’t necessarily so. In the fi rst Sect. 1 say a few words about biological functions before saying why, in the second, I think human perception has accurately representing the environment as a biological function. In the third and fourth I state Burge’s case for thinking this isn’t so. In the fi fth I explain why Burge’s grounds do not make his case and then in the sixth I critically examine an example Burge offers to buttress his case. In the seventh I say why the issue matters to Burge and why, even though I reject his argument, we are not at cross-purposes. 2 There is an everyday sense of ‘fi tness’ and a technical sense. I trust we would all like to stay fi t—to stay in shape.
    [Show full text]
  • The Color of Smell
    The Color of Smell A cross-modal interactive installation for individual expression Hannah Weiser Interaction Design Master’s Program One-year master 15 Credits Spring semester 2016 Supervisor: Simon Niedenthal [Main field of study – See the programme syllabus for the main field of study] [Level, specify Bachelor, One-year master or Two-year master] [Specify the number of credits for this thesis] [Semester/Year when this thesis is submitted] Supervisor: [First name Surname] “The Color of Smell” A cross-modal interactive installation for individual expression Table of Content 1. Abstract 5 2. Introduction 5-6 3. Research Questions 6 4. Theory 7-16 4.1 The potential of smell: Psychological and physiological properties 4.2 Olfactory communication in HCI 4.3 The olfactory medium 4.4 Smell in a cross-modal context 4.5 Synesthetic Design - A cross-sensory approach 4.5.1 Synesthetic perception 4.5.2 Embodied Cognition in the context of synesthetic perception 4.6 Cross-sensory connections for the creation of Synesthetic Design 4.6.1 Cross-modal analogy 4.6.2 Concrete association - iconic conncetion 4.6.3 Symbolic connection 4.7 The benefits of Synesthetic Design 5. Related work - relevant examples 17-20 5.1 Lexus - A Journey Through the Senses 5.2 Odor Emoticons, Olfacticons: An olfactory application that conveys emotion 5.3 The Sound of Touch 5.4 The I/O Brush: Drawing with Everyday Objects 6. Methods 21-23 6.1 Research Through Design 6.2 Co-Design 6.3 Prototyping 6.3.1 The Wizard of Oz 6.3.2 Rapid video prototyping 7.
    [Show full text]
  • Introductory Psychology Chapter 4: Sensation & Perception Winter 2014
    Van Selst Introductory Psychology Chapter 4: Sensation & Perception Winter 2014 Sensation and Perception Chapter 4 of Feist & Rosenberg Psychology: Perspectives & Connections Chapter 4: Sensation & Perception Sensation: receiving physical stimulation, encoding the input into the nervous system; The processes by which our sensory organs receive information from the environment. Perception: the process by which people select, organize, and interpret (recognize) the sensory information, the act of understanding what the sensation represents Transduction: Physical energy neural impulses Chapter 4: Sensation & Perception Absolute threshold: the minimum amount of stimulus required for a percept (note: “sub-liminal” perception is perception below the threshold of consciousness); the amount of stimulation required for a stimulus to be detected 50% of the time. Jnd: just noticeable difference: smallest difference between stimuli that people can detect 50% of the time. Weber’s law: physical intensity vs perceptual (psychological) experience; the idea that the jnd of a stimulus is a constant proportion despite variations in intensity. (2% change for weight; 10% change for loudness; 20% for taste of salt) Sensory adaption: a decline in senstitivity to a stimulus that occurs as a result of constant exposure. e.g., the perceived loudness of a nightclub or a plane Chapter 4: Sensation & Perception Signal-Detection Theory: statistical model of decision making (Sensitivity & Bias) Chapter 4: Sensation & Perception The five classic senses: • Vision (day,
    [Show full text]
  • Object Apprehension Using Vision and Touch
    Object Apprehension Using Vision and Touch R. Bajcsy and S.A. Stansfield ._ University of Pennsylvania [ i) " ;'_ Philadelphia, PA 19104 [ ....... 1. Abstract ._._ ..... We define object apprehension as the determination of the properties of an object and the rela;t|bnships among these properties. W_t contrast tla!s_with recognition, which goes a step further to attach a label to/the object as a whole. Apprehension_ls_rfi_d_ental ,to manipulation. This is true whether the manipulation is t:161ngcarried out by an autonomous ro,.Ipo_s"thi3 result of teleoperatlon involving sensory feedback. W6 present an apprehension paradigm using both vision and touch. In this model, we define a representation for object apprehension in terms of aset of primitives and features, along with their relationships. This representation is the mechanism by which the data from the two modallties is combined. It Is also the mechanism which drives the apprehension process. 2. Introduction It h_s been suggested by both psychologists and perceptual roboticists that objects are defined In terms of their Parts and features. It has also been suggested that these features determine not only our recognition of objects, 10ut also our interactions with them. It seems reasonable to say that these features (along with their relationships] are the outputs of the perceptual system. Ore '!rst task, then, in building a robotic perceptual system is to determine what this fixed set of features will be. This is also vital to our representation pardigm, since these features r_ust form the building blocks of objects wh!ch are to be manipulated by the system. We have chosen a hierarchical representation for objects which consists, at the lowest level, of a set of modality dependent primitives.
    [Show full text]
  • Psychology of Auditory Perception Andrew Lotto1∗ and Lori Holt2
    Overview Psychology of auditory perception Andrew Lotto1∗ and Lori Holt2 Audition is often treated as a ‘secondary’ sensory system behind vision in the study of cognitive science. In this review, we focus on three seemingly simple perceptual tasks to demonstrate the complexity of perceptual–cognitive processing involved in everyday audition. After providing a short overview of the characteristics of sound and their neural encoding, we present a description of the perceptual task of segregating multiple sound events that are mixed together in the signal reaching the ears. Then, we discuss the ability to localize the sound source in the environment. Finally, we provide some data and theory on how listeners categorize complex sounds, such as speech. In particular, we present research on how listeners weigh multiple acoustic cues in making a categorization decision. One conclusion of this review is that it is time for auditory cognitive science to be developed to match what has been done in vision in order for us to better understand how humans communicate with speech and music. 2010 John Wiley & Sons, Ltd. WIREs Cogn Sci 2010 DOI: 10.1002/wcs.123 INTRODUCTION a person spoke and a dog barked and asked you to name the number, location, and identity of these magine sitting on a pier with a friend over a sound sources—you would find the challenge much Iswimming beach where the swimmers are out of less intimidating (though you may still question your your direct view. Your friend offers you this challenge: quality of friend). The ability of the auditory system from the movement of the water waves that you see to segregate, locate, and categorize events in the below, can you tell how many swimmers there are? environment is a remarkable accomplishment given Where are these swimmers relative to each other? the complexity and transient nature of sound waves.
    [Show full text]
  • The Use of Illusory Visual Information in Perception and Action
    I Medische E;t:JliDtheek !2oo.)" E.U.R. ~~ ~ The Use of Illusory Visual Information in Perception and Action Het gebruik van illusoire visuele informatie in perceptie en actie Proefschrift ter verkrijging van de graad van doctor aan de Erasmus Universiteit Rotterdam op gezag van de rector magnificus Prof.dr. S.W.J. Lamberts en volgens besluit van het College voor Promoties. De openbare verdediging zal plaatsvinden op donderdag 19 mei 2005 om 13.30 uur door Denise Desiree J osefina de Grave geboren te Winterswijk Promotiecommissie Promotor: Prof.dr. J.G.G. Borst Overige leden: Dr. M.A. Frens Prof.dr. C.J. Erkelens Prof.dr. C.I. de Zeeuw Copromotoren: Dr. E. Brenner Dr. J.B.J. Smeets Dit proefschrift werd mede mogelijk gemaakt door financiele steun van de Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), gebied Maatschappij- en Gedragswetenschappen (MaGW), dossiemummer 490-22-lSOb. Contents Contents Chapter 1 5 General Introduction Chapter 2 17 Are the original and the induced Roelofs effect caused by the same shift in straight ahead? Chapter3 31 Effects ofthe Ebbinghaus figure on grasping are not only due to misjudged size Chapter4 45 The coding of pointing movements ChapterS 61 The coding of saccades Chapter 6 75 Summary and conclusions Hoofdstuk6 87 Samenvatting en conclusies References 95 Dankwoord 107 Curriculum Vitae 109 Publications Ill Chapter 1 General introduction Chapter 1 Humans constantly interact with objects in the environment (e.g. grasp a pencil for writing or pick up a cup of tea) without making many mistakes in these performed actions. To guide these actions, visual information is used.
    [Show full text]
  • Perceptual Co-Reference
    Perceptual Co-Reference Michael Rescorla Abstract: The perceptual system estimates distal conditions based upon proximal sensory input. It typically exploits information from multiple cues across and within modalities: it estimates shape based upon visual and haptic cues; it estimates depth based upon convergence, binocular disparity, motion parallax, and other visual cues; and so on. Bayesian models illuminate the computations through which the perceptual system combines sensory cues. I review key aspects of these models. Based on my review, I argue that we should posit co-referring perceptual representations corresponding to distinct sensory cues. For example, the perceptual system represents a distal size using a representation canonically linked with vision and a distinct representation canonically linked with touch. Distinct co-referring perceptual representations represent the same denotation, but they do so under different modes of presentation. Bayesian cue combination models demonstrate that psychological explanation of perception should attend to mode of presentation and not simply to denotation. §1. Sensory cue combination A familiar picture holds that mental activity involves computation over mental representations (Fodor, 1975, 2008; Gallistel and King, 2009; Pylyshyn, 1984). This paper will discuss representations that figure in computations executed by the perceptual system. My focus is sensory cue combination. The perceptual system estimates distal conditions (e.g. shapes, sizes, colors, and locations of distal objects) based upon proximal sensory input 2 (e.g. retinal stimulations, or vibrations in the inner ear, or muscle spindle activity). When estimating distal conditions, the perceptual system typically exploits relevant information from multiple sensory modalities. If you see a barking dog across the park, then you receive both visual and auditory information relevant to the dog’s location.
    [Show full text]
  • Bayesian Natural Selection and the Evolution of Perceptual Systems
    Received 13 September 2001 Accepted 13 November 2001 Published online 22 April 2002 Bayesian natural selection and the evolution of perceptual systems Wilson S. Geisler* and Randy L. Diehl Department of Psychology and Center for Perceptual Systems, University of Texas at Austin TX 78712, Austin, USA In recent years, there has been much interest in characterizing statistical properties of natural stimuli in order to better understand the design of perceptual systems. A fruitful approach has been to compare the processing of natural stimuli in real perceptual systems with that of ideal observers derived within the framework of Bayesian statistical decision theory. While this form of optimization theory has provided a deeper understanding of the information contained in natural stimuli as well as of the computational principles employed in perceptual systems, it does not directly consider the process of natural selection, which is ultimately responsible for design. Here we propose a formal framework for analysing how the statistics of natural stimuli and the process of natural selection interact to determine the design of percep- tual systems. The framework consists of two complementary components. The rst is a maximum tness ideal observer, a standard Bayesian ideal observer with a utility function appropriate for natural selection. The second component is a formal version of natural selection based upon Bayesian statistical decision theory. Maximum tness ideal observers and Bayesian natural selection are demonstrated in several examples. We suggest that the Bayesian approach is appropriate not only for the study of perceptual systems but also for the study of many other systems in biology. 1.
    [Show full text]