Block on perceptual variation, attribution, discrimination, and adaptation

The Border between Seeing and Thinking is an instant classic. Exploiting his unparalleled knowledge of cognitive psychology, neuroscience, and vision science in rigorous and subtle philosophical argument, Block provides a sweeping analysis of key aspects of perception. Scientifically astute and philosophically decisive, it constitutes the gold standard of how to do empirically informed philosophy of mind. It would be impossible to do justice to it. We will restrict ourselves to six key issues: borders, perceptual variation, attribution, discrimination, adaption, and singular content. Since his influential distinction between access and phenomenal consciousness, a distinctive feature of Block's methodological approach has been to find borders and joints and then focus the discussion of the relevant issues around these borders and joints. The Border between Seeing and Thinking is the culmination of this approach. It is rich with discussion of such borders: most centrally the border between cognition and perception, but also the border between conscious and unconscious perception, and conceptual and non-conceptual content, to name just a few. The approach of organizing debates around borders has become common in philosophy of mind, largely due to Block's towering influence. Alternative approaches are focused not on borders between, say, perception and other mental capacities, but rather, for example, on determining what is constitutive of perception, on specifying its fundamental nature, or on understanding its mechanism. No doubt, those focused on these projects make lots of distinctions along the way, some of which correlate with joints and borders, and of course, Block has lots to say about the nature of perception. So ultimately, the two approaches may converge. That said, there is a question to be asked about what can be gained from attempting to delineate borders. Here is why. We can all agree that there are mental events that are primarily perceptual and others that are primarily cognitive. But if perception and cognition are intimately interrelated—not just functionally but also regarding their physical implementation (see Pessoa, 2022, Pessoa et al, 2021)—then the question arises whether there is a joint in nature that correlates with the distinction between cognition and perception postulated not just by Block but by researchers from many different scientific domains. More generally if the borders of the mind are fuzzy and permeable in all directions with information processing circuits distributed, then the question arises of how much we can learn about the nature of perception from attempting to specify and delineate borders and joints in the mind and brain. One key feature of perception that Block barely addresses is its perspectival character: a subject who gains information about her environment via perception does so under specific perceptual conditions. Such perceptual conditions include her location relative to the perceived objects and features in her environment, the lighting conditions, the color context, and the acoustic conditions (reverberation, refraction, and diffusion features in our environment), to name just a few. Due to the perspectival nature of perception, the stimuli on which our visual system operates are perpetually in flux. Nonetheless, we see relatively stable properties in our environment, such as the uniform whiteness of the walls of our office or the rectangular shape of our desk. The constancy mechanism of our perceptual system downplays perceptual variation due to changes in lighting conditions, changes in the spatial relation between the perceiver and the perceived features, and other such perceptual conditions. So it downplays changes in our sensory input that are due to the perspectival nature of perception. This allows us to focus on features in our environment that are relatively stable. Our first question for Block is how these facts about perspectival variance and perceptual constancy square with his thesis that the primary goal of vision is to deliver news. One possible response is to say that delivering news is just one of many core functions of perception. Two further core functions include guiding action and tracking objects. Regarding these two latter core functions, the constancy mechanism minimizes most news, namely news that is due to changes in perceptual conditions. A second possible response is to say that there is important news and fleeting news. News due to perceptual variation is fleeting news. The idea would then be that perception aims only at providing important news. Adopting this strategy would require a principled distinction between important and fleeting news. A complicating factor with this second approach is that occasionally, news that is due to perceptual variation is important. A freight train that is barreling towards one as one is taking a stroll on what one thought was a defunct train track is important news. Yet, the rapidly increasing perspectival size of the train is news that is due to perceptual variance. So in this case, we gain important news due to fleeting perspectival news. We are not questioning Block's idea that perception functions to deliver news. We note only that a specification of this idea is required to account for the fact that the perceptual system systematically downplays vast quantities of fleeting information about perceptual variation in favor of perceptual constancy. So it systematically downplays fleeting perspectival news. Block argues that “all the markers of perception [rivalry, pop-out, illusory contours, processing speed, and adaption]… involve perceptual attribution” (p. 144). A further question for Block is: what is attribution? The term “attribution” is a philosophical term of art, championed primarily by Burge (2010). It is not used in vision science, and it is not obvious what, if anything, it would correlate to in any of the many branches of vision science. The same holds true for neuroscience more generally. In discussing the work of psychologists and neuroscientists, Block reports them as making claims about attribution.1 But among the authors he cites, we could not find one who makes claims about attribution, though there are many claims made about generalization, identification, recognition, classification, and categorization. For example, he cites Thoen et al. (2014) in support of his claim that “different perceptual systems can prioritize attribution and discrimination in different ways” (p. 148). But while Thoen et al. discuss recognition, they do not discuss attribution, and Block does not consider recognition necessary for attribution. As we will argue, vision science does not support the idea that “all markers of perception … involve perceptual attribution”, on any notion of what attribution might be. We will argue that perception is fundamentally a matter of discriminating. At its core perception is difference detection over some dimension. That dimension could be time or space, it could be the frequency spectrum of light or sound, positions along the body, or molecules in the air, to name just a few. To detect difference, one needs a discrimination mechanism. In short, no difference, no discrimination, no perception. If there is no discrimination, the neural activity is unmodulated. Discrimination is essentially comparative: discriminative sensitivity is the ability to distinguish between stimuli of different kinds, where one of those stimuli might be a null-stimulus. Detection is discrimination between a stimulus and a null-stimulus, that is, the absence of the relevant stimulus. If a perceiver discriminates a stimulus from a null-stimulus, she detects the stimulus. Attribution is an activity in which a feature is assigned to an x, where that x could be an an a in space, to name just a Discrimination is an activity in which two or a and a in the environment are where those could be features, or So to is to x as So there be an x to which the is on the discrimination there does not to be an x to which the is could be any x as One might the discrimination by how the could conditions. In while conditions require a such as is conditions do For the to be there needs to be an in the A more important difference between the attribution and the discrimination is how to the relation between attribution, and the discrimination sensory discrimination of features, and events in the environment their is a of discrimination on which we between features we but that of discrimination is not sensory Block does not that sensory discrimination is an important of the perceptual mechanism. One question for Block is where sensory discrimination the idea of x as Block his discussion of attribution and discrimination of that there is no to favor attribution or discrimination as more the he makes many claims that are in with this For example, he argues not only that perception is but argues further that most of discrimination require from one in the case, (p. 148). Block allows for to the he in this and that most of discrimination require first properties to the environment so as to be to then those In we do not see this would be the As Block we do not to first is and is to from As we argue, the stimulus from other stimuli is required to features in the The of features in the environment constitutes the for input objects and et In these features in the environment them to objects or and to a with those of in of in the visual processing the in the In short, features are in the environment, and are from the Since the features are in the environment, there is no to Block does not that features are in the So the that features are in our environment, what is there that we do not the features in our is there that in most of perception, we features that first could be a is a If a the it we can that the as an of One could this mechanism as discriminating. It could be that the but does not and so makes a to perceptual attribution. of a we could consider a are of which a A of discrimination as could claim that one of the discriminates between that and no the of discrimination could say that the discriminates between and But then the claim that discrimination is would more of a a claim about what is fundamental to perception. A could be an that its two are different and does not its are the A would be an if we one of the the as (p. Block that there is a difference between detection and discrimination in that the mechanism of detection has one while the mechanism of discrimination has a mechanism is in the of only one stimulus a mechanism is in the of stimulus A and stimulus A from the fact that a is a this does not with the of perception. It that are in a they are The same detection and are the only which a stimulus can the perceptual A is to a of be it a specific of light a of or of or some other In response to the a is that is, there is a in the That in its is a of the stimulus In short, the of detection and discrimination is the The same holds further in an in the visual to stimuli and or not at all to other stimuli only are the same sensory in detection and discrimination, so are the same So there is no in which there is a mechanism for detection that is from a mechanism for if attribution as the then attribution would be discrimination on standard of perception. as Block detection is a of discrimination, the in which a stimulus is from a (p. Block's is the of light The does not if no light is and light is the mechanism is the of light from the null-stimulus. we can detect a feature that feature to an how Block attribution, it be So much for attribution as Block that of perception attribution and discrimination as (p. But “attribution” is not a term used in of perception, and it is what it would provides a to to on which we can then in light of and are used to in which a subject is asked whether a stimulus is or If a perceiver has two is and is not and to the is then this can as from not and as One could argue that a to is is to x, but that is not a term used in the as is a in which there are between which the subject needs to where those could be is and is detection is a of discrimination in of Attribution is is of the mechanism of attribution” (p. Attribution (p. attribution is while discrimination is attribution while discrimination does attribution identification, while discrimination while Block holds that perceptual attribution is he allows for many does not that perceptual is common he to to a cognitive a perceptual So while and make in Block's discussion of attribution, are constitutive of attribution as Block the Block that he of a of perception in which there is discrimination attribution. We are by this If attribution is, as Block an activity in which a feature is to an x it an or in then any of feature perception that feature to an x is a in which we two features those We do not see any as to there could not be a that only More there are many of perception in which there is no in which a feature is to any x, be it an or in Here are a In the two and whether a stimulus in the first or second The stimuli could be color or of The for whether or is no that the feature is to is perception, in which we an of color that our visual In a we are not a feature to et if we the color of the there is no that the color is to an or in this case, one could argue that the color is to a in If we of feature perception that feature to A can be that to anything, be it a or In and it is possible to or features them to course, all these can be so that an x is postulated to which the feature is if that is the then the claim that such involve attribution it impossible to that could the We do not that many of perception involve attribution. features to and in space, in of attribution are those in which at one of the features perceived is not in the are many such all include at one and so the thesis that perception is fundamentally a matter of Here are a few. the a there is no in the the of the sensory So the stimulus on which the visual system operates is not of a But on the stimulus is A different is a in which we see and the shape we may that what we are is a the stimulus is not to a from its A further is a in which we see in a We are not to a but we to our these in common is that they are a of a feature to in the environment that could be in any of one could argue that all features perceived are to some in the perceptual In other one could argue that all of perception are to some the does not such an and we that in most of perception, the features perceived are in the It be that one of the attribution over the discrimination is that the attribution the from the if the fundamental of perception is to features to some x, then there will not be a question as to which features in the environment are with which and a in which to the the of color and In it be that on the there can be such but the to the of in which a subject features and there is an question as to which features are with which objects. It philosophical not be in the of the of such The is a in vision science, and Block it in in his If the to it is the fundamental of perception from the then it would that vision science is with a that does not As the we can detect or features them to an or The arises there is a question as to which or in relevant features in the If such features could not be first been to an or the would It that a to the is in an of the attribution rather, the with the it. one could argue that the of whether perception is fundamentally a matter of or is a But Block of his to this so we that is not his is a difference between the two in so as attribution a feature to an x discrimination does We with a if perceived features are in the environment and if there are features that are not features of or of space, what is the that we can only features they been to or in Block has it that there can be of perceptual attribution would attribution discrimination a of perception in which there is just one stimulus Block say that a of attribution discrimination is a in which the subject a feature to the that stimulus A and any other If a subject is then there is sensory input to perceptual If there is sensory input to perceptual then sensory of perceptual system a to If sensory of perceptual system a to then perceptual system If a subject is perceptual system We see no to that the perceiver features at any of this We will support to in In support of No perception is possible if the sensory system has at the been by a stimulus. If the sensory are then no information the relevant perceptual and perception does not there can be sensory input if the sensory input is the and so to a there is no perception. if the subject is there be sensory So sensory input is necessary but not for perception. One might whether the information could not the system by some other In no doubt, a perceiver can a and the of her might how she perceptual her might how she perceptual But if no perceptual information the perceptual then she is not a in which the sensory can be a of perception. In support of A sensory is it is by a stimulus. If the of the stimulus or the of the relevant sensory the a to For in the and to to to visual in the which in to in the visual It is possible to a response in sensory In the for a response is one et In on the of of is to from is as to which the sensory information from the sensory In sensory to For visual to the are to how they are in the visual et in stimuli activity the et facts do not in all these sensory input was to and all we for perception is input from some to some perceptual processing In support of To we sensory processing of the information they a response that The sensory the so that can it in the the system information that via the sensory that In If there is then the perceptual system will This is the case, if the stimulus is the perceptual system will the uniform stimulus from how this stimulus That is a of Block cites the as if it a for the discrimination But in the is for the discrimination and the attribution A is a of In discussing Block to be about as he is just Discrimination can over any dimension. One important such dimension is that at time we see a At time we see a So at we detect the by it from the at A is a relative If at a perceptual system detects stimulus then the at time is the absence of The not be the absence of any stimulus. It is the absence of the stimulus that at course, the perceptual system does not to the as The is that if a system detects it be to the of A from A was not does the support the discrimination one a there is a namely and the subject is to the That allows the perceiver to see the Due to of any further the sensory system and the No stimulus the sensory and so no is to discrimination of perception So the the thesis that if there is no difference, there can be no discrimination, and so no perception. the attribution the relevant color is and it is the To the thesis that there can be no perception discrimination does not that all aspects of perceptual are due to It only that there can be no of perception that does not include at one stimulus from where one of the two stimuli could be the null-stimulus. Block holds that one of the markers of perception is One Block argues that attribution is for perception is that he holds that it be in As he of the of all on attribution” he argues that in we (p. as Block is a of perception and on attribution but does not involve discrimination, would that attribution, but not discrimination is a key feature of perception. In we agree with Block that is a key of perception. can at the of the perceptual namely the of the at the of the visual system and the of sensory in perceptual we that many are It is not obvious that they involve attribution on any of the one might attribution. To a of Block's key to an of attribution of or of the of the of attribution of the for attribution at (p. 144). We can all agree that the for features, but there is no to the philosophical of attribution to such A is as an of to or the for or most of can be for the of discrimination, any to attribution. To say that no discrimination is Block that due to the the discrimination system But there is no to that it about a in to a specific stimulus. Due to the the relevant sensory or are not in the same to the as they would the not So the information that the sensory or to will be different if they not been by the In short, not only do perceptual involve discrimination, one can the subject at One discriminates with Due to been by But they are discriminating. This about to at other of sensory can at sensory in to color to et and et et for the of that involve attribution, as Block if this the case, any involve discrimination as and they involve discrimination at a more fundamental the perceptual system needs to on some stimulus for the mental to as a perception. perception, Block the in and That said, it to attribution and discrimination, we that he the idea that there can be attribution discrimination does not Block's key So he would to that perception is a matter of discriminating. We that he Our for Block his discussion of singular content. Block the matter of whether perception is fundamentally a matter of or fundamentally a matter of in the of a discussion of whether perceptual is singular or It might be to how the two issues are argues that perceptual is singular perception is a matter of in the environment see So she argues that perception is by its in that perceptual is by the perceived This on the fact that perception is stimulus If is a the stimulus that her neural system from the Due to perceptual by the perceived perception on the of perception, and provides us with knowledge of the perceived knowledge of It is important to note that one can that perceptual is singular for other one may be to that perception is fundamentally a matter of for perceptual But if there is a between perception a matter of and perceptual then it would be along the Block's focus is on issues in philosophy of mind. It is on the that perception in and knowledge about in our environment on the it in and about singular So it is not that he has for the idea that perceptual is is much with which we agree in Block's discussion of singular content. For example, we agree that has no access to the distinction between and (p. this does not with that perceptual is at in by the on any to which phenomenal on or is in content, there can be aspects of perceptual that are not in phenomenal So the fact that we do not access to the of the perceived does not that the perceived do not perceptual content. In Block that his to what we access to is not of an the singular (p. be to she an that we see of is that we the shape of the the shape be we see what does not on us this her of as in are of the shape of the some of which of and others of which do One is that see the a shape to it. It is in of the shape of the that see it as that is in this is the a the and it is in the perception in of some of its properties but not is no to to on these (p. The Block to be that objects which in is as the idea that objects If that is the then it is just a of the an a shape is to say that the a which is a feature in the It is a feature that a and it is a feature that we can Block has it that features are in the environment to which we are and that features are But he does not that are in our has to Block with as to and for of them is a to and are to and would make this a the is not by that the same might a different (p. In the fact that the same could any does not that the is not by the it is that the to and of a is a It is in that it could be to and any This of is a key feature of It is to her that perception is Block from that are in this way, to that there is a and a of and But Block there is no and of and are on any understanding of If we a to and of the is the perceptual is by that of this argument, see We see no Block that perception has singular content. only is the idea that perceptual is singular with his key it would make his more to and of or with of how perception singular the of singular provides us with knowledge of in our environment, and singular about in our Our has been to but in we would to note that we are to most of Block's We are for the to with his