Concerning Phenomenal Expectations

Abstract

Dishabituation experiments that involve an increase in the duration of longer looking times ofinfants have been argued to entail core knowledge. Despite how one chooses to understand theresults of these experiments and whether or not core knowledge is needed to describe and explainthem, what is needed is a theory that explains why infants find a state of affairs more novel andwhy infants behave as they do. Tis theory would account for infants' expectations. In hisupcoming publication, Stephen Buterfill suggest a theory based on what he calls phenomenalexpectations. Buterfill maintains that phenomenal expectations not only describe and explaininfants' behavior in anticipatory looking experiments, but also they are able to overcomeDavidson's challenge of providing a middle-ground between propositional thought and mindlessbehavior. I begin by introducing the discussion concerning dishabituation experiments, coreknowledge and phenomenal expectation, then I investigate the significance of phenomenalexpectations integrated into theories of perception and test whether phenomenal expectationsovercome Davidson's challenge.

1. Introduction

To understand the human mind, many psychologist and philosophers have studied its infancy.Grasping the operation of the infant mind affords an understanding of the initial conditions bywhich the adult mind operates. If claims of core knowledge have any validity, then this knowledgemust be exhibited by the mind of an infant. Since infants are unable to report their experiences,psychologists have developed methods of probing into their mind. Tese experiments rely onfour different methods. Tese methods are habituation, violation-of-expectations, anticipatorylooking and search behaviors. At the heart of these methods is the idea that what appears novel toan infant violates its expectations, and what violates its expectations is indicative of theoperations of the human mind in its infancy.

For the sake of example, imagine an experiment devised to test whether an infant is able toobserve object permanence; that is, whether an infant is aware that objects continue to persistunobserved. At first, the infant is habituated to a scenario: for example, a toy being placed andsubsequently revealed to persist behind an opaque screen. Once the infant has been habituated,measured by the duration of the infant's looking times, the experiment is altered to violate theinfant's supposed expectations. Te toy is placed behind the screen as before, but, unbeknownst tothe infant, the toy is secretly removed. Upon revealing the toy's absence, the hypothesis is thelooking time of the infant will be significantly longer. Te longer looking time indicates that thelater state of affairs violated the infant's expectation of how the scenario would unfold, thusgiving researchers the empirical basis from which theses can be established about the functioningof the human mind in its infancy.

Te finer details of the last few stages of the dishabituation experiments are of profoundinterest for the philosophy of mind. What inferences can we draw from those results? Does thehuman mind possess core knowledge concerning physical objects? What about the behavior ofthe infant? What are these violated expectations exactly? Why does an infant find one scenariomore novel than the other? I shall address all of these questions while focusing on the expectationaspect of the experiment. But before I discuss the details of a theory that seeks to describe andexplain those expectations, I shall discuss the history of core knowledge as it relates topsychology. Te dialectic of the historical progress in developing more refined theories of coreknowledge will set the stage for a discussion of theory of expectations called phenomenalexpectations.

2. Core Knowledge

Te notion that the operation of the human mind is dependent on some form of core knowledge isentailed by many psychological theories and philosophies of mind. Piaget claimed a strongervariant of what might be called enactivism today. Enactivism is the theory that perception isreciprocally modified by embodied action. Te major difference between enactivism and Piaget'stheory is that the later greatly underestimated the perceptual capabilities of infants. His theoryrequired a more robust form of action that infants, due to both physical and mental constraints,are unable to perform. Another theory that held sway was Gestalt theory. It claimed thatperception was based on a series of principles of simplicity and grouping similarities. But in 1989,the cognitive psychologist Spelke was able to disprove these theories with experiments testing thecognitive abilities of infants and young children.¹ To investigate the operation of the human mindin its infancy, Spelke and others developed a series of experiments to test object perception ininfants. I shall describe one of these experiments, particularly the one that disproved both Gestalttheory and Piaget's theory about how the human mind operates, which will be relevant for ourdiscussion later.

In this experiment, an infant is set to observe states of affair. Te experimenters' goal is to discover whether an infant will become dishabituated. Dishabituation will indicate that the infants expectations are violated. Violated expectations indicate conceptual boundaries of human mind's operation; these boundaries are also referred to as signature limits. Similar to how crossing a territorial boundary may set off an alarm and indicate trespassing into a different field, the exposure of signature limits reveals inherent systematic structures.

To describe one of Spelke's experiments, imagine a long stick held diagonally behind ascreen. Te top and botom parts of the stick are seen sticking out from behind the screen. Tequestion is whether removing the screen and revealing there to be either a single stick or twoseparate sticks would cause the duration of the infants' looking times to significantly increase.Gestalt theory, based on it principles of simplicity, claims that the infant will not be dishabituatedto the state of affairs revealing a single stick, while the state of affairs revealing two separatesticks will. Te results of the experiment cast doubt on the principles of simplicity, since infantsfound both scenes to be equally novel.

In the next phase, the experiment was set up the same as before, but the single stick or twosticks respectively were moved in unison from side to side. In this phase, infants becamedishabituated when two separate sticks were revealed. Tis shows that infants only begin toassociate two parts as one when parts are observed moving together. Moreover, the same results were produced regardless of the shape and texture of the observable parts. Tis means, the grouping axioms of Gestalt theory do not hold either. Te only axiom of Gestalt theory able tostand the test of Spelke's experiments was the axiom called common fate; i.e. stimulus elementsare perceived as a unit when they move together. I shall draw upon the axiom of common fatelater on.

I hope that a certain dialectic has started to emerge regarding core knowledge claims. Gestalt theory's claims at core knowledge were based on a priori reasoning and limited experiments with adults. Piaget's claims at core knowledge were based primarily on experiments with older children and their testimony. It wasn't until the discovery of the relationship between dishabituation and looking-time duration that the operation of the infant's intellect could be probed, and only once the mind in its infancy could be experimentally tested, could core knowledge claims be tested and invalidated.

Afer Spelke's experiments had disproved Gestalt theory and Piaget's claims, the next stepwas to collect data from Spelke's and other related experiments and develop a new theory of coreknowledge. Te result was the Simple View. Te Simple View claims core knowledge amounts to asmall number of principles that underlie the perception of objects. Tese principles are commonsense notions of objects such as rigidity, cohesion, no action at a distance, etc. Tough the SimpleView is an atractive option, because it possesses much explanatory power and remains at itsfoundation quite simple, it does make several false predictions concerning children's purposiveactions. On top of that, the Simple View claims the relation between principles of objectperception and infants' minds is based on belief or knowledge, and this claim is in directcontradiction with Davidson's challenge—I shall discuss Davidson's challenge shortly. All of thishas lead many psychologists and philosophers of mind to rejected the Simple View.

3. Object Indexes and Davidson's Challenge

In place of the Standard View, Buterfill suggests looking to recent research thatconjectures infants' abilities to be underpinned by a system of object indexes. To understandobject indexes, consider for a moment your visual field. Look around the many objects placedalong your desk. If one of these objects were to fall, you'd be able to reliably guess the newlocation of the object. According to the suggested conjecture, you're aware of object's newlocation because your mind has assigned an index to object. Object indexes track the location andtrajectory of objects similar to how using your finger to point out the object and track itsmovement would.

Te theory of object indexes is a candidate for replacing the Simple View. Object indexes do not make the false predictions concerning the purposive actions of children and they are not dependent on the child possessing any knowledge or beliefs concerning either the indexes or objects. Tat said, object indexes fail to explain or describe why infants find a state of affairs in dishabituation experiments more novel. Afer an infant has been habituated to a phenomenon, afer an object index has been assigned to each object, and when suddenly two object indexes are needed where only one was assigned, the infant looks longer at this state of affairs than it would were the number of objects and the number of indexes constant. In order to explain an infant's behavior in these experiments, object indexes alone are not enough.

Before we can begin to understand an approach at explaining the behavior of infants indishabituation experiments, we need to discuss Davidson's challenge. Davidson has arguedagainst the possibility of understanding the mind of an infant. He says we have language fordescribing the mind of an adult, as it is based in propositional thought, and we have language fordescribing the operation of mindless behavior. However, we do not have a language for describinga mind that is in between propositional thought and mindless behavior. In other words, we will donot have the language for describing or explaining the mind of an infant. Developing a theorythat is home to the Goldilocks Zone between propositional thought and mindless behavior isDavidson's Challenge. Since infants do not have propositional thought, this means a theoryexplaining infants' behavior cannot depend on an ability to hold beliefs, possess knowledge ormake judgments.

4. Phenomenal Expectations

In order to explain the longer looking times, Buterfill proposes phenomenal expectations. But what are phenomenal expectations exactly? Are phenomenal expectations able to circumvent the need for an infant to have beliefs about objects; i.e. do phenomenal expectations overcome Davidson's challenge? Do phenomenal Expectations explain the novelty behind the longer looking times of infants? In what remains, I shall address these questions.

To illustrate a phenomenal expectation, Buterfill provides one example in his upcomingpublication Te Developing Mind, as well as two more examples on philosophyofrains.com. Wewill consider the first example from his upcoming publication.² Tis example concerns two staticimages. In the first image, we can see a finger partially occluding the outline of what appears to bea triangle. In the second image, the triangle is no longer obscured by the finger and the lines ofwhat before were blocked are revealed. Te triangle isn't a triangle afer all. It is a small diamondatached to an irregular pentagon, but we couldn't see this before because the finger had occluded

[...]

¹ Spelke, E. S, C. von Hofsten and R. Kestenbaum, “Object perception and object-directed reaching in infancy: Interaction of spatial and kinetic information for object boundaries”, Developmental Psychology 25:185-196, 1989.

² Butterfill, Stephen, “The Developing Mind”, upcoming publication, pg. 62.