iCog 3
Sense and Space


Tactile Object and Embodied Subject
Tony H. Y. Cheng

This talk starts with an empir­ic­al vin­dic­a­tion of tact­ile field with the resources from vari­ous stud­ies done by Patrick Haggard’s lab. The basic idea is that the tact­ile field sup­ports com­pu­ta­tion of spa­tial rela­tions between indi­vidu­al stim­u­lus loc­a­tions, and thus under­lies tact­ile pat­tern per­cep­tion. Perception of spa­tial pat­terns across the field is linked to a struc­tur­al rep­res­ent­a­tion of one’s own body. Tactile pat­tern judge­ments depend on sec­ond­ary factors over and above loc­al tact­ile per­cep­tu­al abil­ity at the stim­u­lated loc­a­tions (sec­tion 1). While the nocicept­ive sense is neither extero­cept­ive nor object-directed, the thermal sense is extero­cept­ive but not object-directed, and the tact­ile sense is both extero­cept­ive and object-directed (Mancini et al. 2015, Marotta et al. 2015). The basic idea is that estim­at­ing dis­tances between stim­uli on the skin requires a met­ric rep­res­ent­a­tion of spa­tial rela­tions, which under­lies abil­it­ies of per­ceiv­ing the size of extern­al objects. Both the nocicept­ive sense and the thermal sense are poorly in estim­at­ing dis­tances like this, but the tact­ile sense is much bet­ter in this regard thanks to the oper­a­tion of tact­ile field (sec­tion 2). It will then be argued that con­tra O’Shaughnessy (1989) and Martin (1992), touch is intrins­ic­ally spa­tial thanks to the oper­a­tion of tact­ile field. More spe­cific­ally, exactly because tact­ile field brings us tact­ile objects as well as their spa­tial prop­er­ties such as size, the spa­ti­al­ity of touch is intrins­ic to the tact­ile sense itself (sec­tion 3). And finally, it will be dis­cussed how we are aware of our bod­ies as phys­ic­ally as well as physiolo­gic­al objects. In this sense, tact­ile pat­tern per­cep­tion pre­sup­poses a self that is an object embed­ded in the world, rather than simply a view­point on the world (Merleau-Ponty 1962, McDowell 1996).


Space and Phenomenal Presence
Mattia Riccardi

In visu­al per­cep­tion things are exper­i­enced as really being there, present before the eyes. I’ll call this fea­ture “phe­nom­en­al pres­ence” (PP). (Other labels are feel­ing (or sense) of pres­ence (Dokic 2012, Dokic/Martin 2012), feel­ing of real­ity (Farkas 2014), Leibhaftigkeit (Husserl 1907, Jaspers 1911).) Whereas imagery and pictori­al see­ing typ­ic­ally lack PP, many cases of hal­lu­cin­a­tion do dis­play it. So reflec­tion on these dif­fer­ent cases is argu­ably cru­cial for devel­op­ing a com­pre­hens­ive account of PP. Matthen (2005, 2010), who only focuses on see­ing and pictori­al see­ing, has argued that PP depends on the kind of spa­tial cog­ni­tion sub­serving (dorsal) motor-guiding vis­ion. Though I also believe that cer­tain spa­tial fea­tures of seen objects play an import­ant role in explain­ing PP in the visu­al case, I think that Matthen’s pro­pos­al is unten­able. The goal of this paper is to offer an altern­at­ive, attention-based explan­at­ory account. (The account is “explan­at­ory” because it remains neut­ral about the nature of PP, i.e. wheth­er it is a phe­nom­en­al prop­erty of exper­i­ence, some kind of doxast­ic atti­tude (Farkas 2014), a cog­nit­ive feel­ing (Dokic 2012, Dokic/Martin 2012), or some oth­er thing). First, I shall present decis­ive empir­ic­al evid­ence (regard­ing, in par­tic­u­lar, cases of dis­so­ci­ation between dorsal and vent­ral vis­ion and a cer­tain type of aud­it­ory hal­lu­cin­a­tions) against Matthen’s account. Second, I shall argue that PP is basic­ally a mat­ter of object-based atten­tion by con­sid­er­ing how per­cep­tion and hal­lu­cin­a­tion work both in the visu­al and in the aud­it­ory cases. Third, I shall argue that the case of pictori­al see­ing shows that, for the visu­al case, a purely atten­tion­al mod­el needs to be com­ple­men­ted by reflec­tion on spa­tial exper­i­ence. Against Matthen, I shall defend the claim that the crit­ic­al factor here is not motor-guiding spa­tial cog­ni­tion, but rather wheth­er visu­al objects are con­sciously exper­i­enced as volu­min­ous.


Acquaintance, Spatial Properties and the Explanatory Gap
Thomas Raleigh

Of the many pos­sible motiv­a­tions for Naïve-Realism, one that has received rel­at­ively little dis­cus­sion is the theory’s alleged abil­ity to help solve the ‘Explanatory Gap’ – e.g. Fish (2008, 2009), Langsam (2011). I provide a refor­mu­la­tion of this gen­er­al line of thought that makes clear­er how and when a Relational the­ory of per­cep­tu­al exper­i­ence could help to explain the spe­cif­ic phe­nom­en­al nature of such exper­i­ence. In par­tic­u­lar, I show how and why this form of explan­a­tion will work best for the case of visu­al shape phe­nomen­o­logy rather than col­our phe­nomen­o­logy. I also argue that the rela­tion­al the­ory can give a nat­ur­al explan­a­tion for why we should expect col­our phe­nomen­o­logy to remain less read­ily intel­li­gible than shape phe­nomen­o­logy.


Joint Experience, Demonstrative Reference, and Sense of Space
Axel Seemann

In this talk I con­sider the ques­tion of how to think about the kind of spa­tial exper­i­ence that is neces­sary to make and under­stand demon­strat­ive utter­ances in joint per­cep­tu­al con­stel­la­tions. I take this abil­ity to depend on the sub­jects’ com­mon know­ledge of which object they are demon­strat­ively identi­fy­ing. The prob­lem is to spell out what this know­ledge amounts to in joint cases. I argue that it pre­sup­poses the abil­ity to draw what I call the ‘found­a­tion­al dis­tinc­tion’ between sub­jects and objects, and sug­gest that this abil­ity informs the capa­city for perspective-taking. I con­sider some work in devel­op­ment­al psy­cho­logy to shed light on the con­nec­tion between the spa­tial sense of per­spect­ive and the social char­ac­ter of the envir­on­ment in which a range of per­spect­ives on a visu­ally giv­en object are avail­able.


Do We Experience Cross-Modal Spatial Illusions?
Alisa Mandrigin

Cross-modal illu­sions make up one of the main sources of evid­ence we have for per­cep­tu­al pro­cessing being multi­s­ens­ory. These illu­sions are taken to show that, at the very least, pro­cessing in a sens­ory sys­tem asso­ci­ated with exper­i­ence in one sense mod­al­ity can influ­ence what we exper­i­ence in anoth­er sense mod­al­ity. In this paper I query wheth­er we have evid­ence that this takes place, focus­ing on a par­tic­u­lar cross-modal illu­sion: the spa­tial vent­ri­lo­quism effect. The vent­ri­lo­quism effect is meas­ured by meas­ur­ing cross-modal bias­ing of loc­al­iz­a­tion. I sug­gest that we can explain the cross-modal bias­ing effect in terms of the recal­ib­ra­tion of the map­ping between sens­ory and motor ref­er­ence frames. This explains the bias­ing effect in terms of how one is liable to respond to what one hears, and not in terms of changes to what one exper­i­ences in vis­ion or audi­tion.


Oral Referral
Charles Spence

Oral refer­ral is cent­ral to multi­s­ens­ory fla­vour per­cep­tion. The phe­nomen­on, first described a little over a cen­tury ago, is char­ac­ter­ized by the mis­lo­c­al­iz­a­tion of food-related olfact­ory stim­uli to the oral cav­ity. Many research­ers believe that it con­trib­utes to the wide­spread con­fu­sion con­cern­ing which sense really provides the inform­a­tion that is bound togeth­er in fla­vour per­cepts. In this review, evid­ence sup­port­ing the role of a num­ber of factors that have been sug­ges­ted to mod­u­late oral refer­ral, includ­ing tact­ile cap­ture of olfac­tion, the rel­at­ive tim­ing of olfact­ory and gust­at­ory stim­uli, and gust­at­ory cap­ture (pos­sibly involving pri­or entry) is crit­ic­ally eval­u­ated. The latest find­ings now sup­port the view that the oral refer­ral of orthonas­al aroma (what some have chosen to call orthonas­al loc­a­tion bind­ing) is mod­u­lated by taste intens­ity, while for ret­ronas­al odours, it is the con­gru­ency between the ordour-taste(s) pair­ing that is key. Specifically, the more con­gru­ent a par­tic­u­lar com­bin­a­tion of olfact­ory and gust­at­ory stim­uli, the more likely the com­pon­ent uni­s­ens­ory stim­uli will be bound togeth­er as a fla­vour object (or Gestalt) and, as a res­ult, loc­al­ized togeth­er to the oral cav­ity. The pos­sible roles of atten­tion, atten­tion­al cap­ture, and the nutri­tion­al sig­ni­fic­ance of the taste in the phe­nomen­on of oral refer­ral are also reviewed. Ultimately, the sug­ges­tion is made that oral refer­ral may reflect a qual­it­at­ively dif­fer­ent kind of multi­s­ens­ory inter­ac­tion.


Gombrich and the Dual Content Theory of Pictorial Experience
Robert Briscoe

A num­ber of philo­soph­ers have main­tained that every­day visu­al exper­i­ence com­prises two dis­tinct ‘lay­ers’ or ‘folds’ of rep­res­ent­a­tion­al con­tent. The con­tent of the first lay­er rep­res­ents viewpoint-dependent appear­ances or ‘looks’, where these are typ­ic­ally taken to be phe­nom­en­ally 2D in char­ac­ter. The second lay­er of con­tent, in con­trast, is sup­posed to rep­res­ent intrins­ic, viewpoint-independent prop­er­ties as well as dis­tance in depth. Pictorial per­cep­tion is argu­ably the best case for this dual con­tent the­ory of visu­al exper­i­ence, since a pic­ture really is a flat patch­work of col­ors that, when present to sight, eli­cits the exper­i­ence of depth and 3D struc­ture. One of E.H. Gombrich’s main aims in Art and Illusion is to demon­strate that the the­ory fails even as an account of pictori­al exper­i­ence. In defend­ing this assess­ment, I show that two of the bet­ter known attempts to elim­in­ate the appear­ance of para­dox at the heart of the dual con­tent the­ory of pictori­al exper­i­ence are unsuc­cess­ful. The two putat­ive lay­ers of rep­res­ent­a­tion­al con­tent in pictori­al exper­i­ence, I con­clude, aren’t merely dif­fer­ent. As Gombrich insists, they are also incom­pat­ible.


Space and Spatial Direction in Olfactory Experience
Solveig Aasen

Accommodating the fact that the loc­a­tion of the nose puts a restric­tion on olfact­ory exper­i­ences need not involve suc­cumb­ing to the idea that we in olfac­tion only can be aware of the loc­a­tion where the nose is. I argue that in those of our olfact­ory exper­i­ences where the con­cen­tra­tion of the odours is sig­ni­fic­antly lower than usu­al, an aware­ness of a lar­ger space is integ­ral to our exper­i­ence. Moreover, I sug­gest that if olfact­ory exper­i­ences include exper­i­ences over time, as well as exper­i­ences at a time, there can be olfact­ory aware­ness of spa­tial dir­ec­tion.


Spatial Awareness and the Chemical Senses
Barry Smith

It’s widely assumed that the chem­ic­al senses are prox­im­al or con­tact senses, telling us about ourselves, not about the spa­tial world around us. That’s because exper­i­ences, like those of taste and smell, are con­ceived as mere sen­sa­tions rather than per­cep­tions of any­thing extern­al to us. This is a mis­taken view of the chem­ic­al senses, which also leaves out the role of spa­tial aware­ness in our exper­i­ences of taste, smell and fla­vour. I shall argue for the spa­tial dimen­sion of these sens­ory exper­i­ences by ref­er­ence to, among oth­er things, key spa­tial illu­sions they can give rise to.


Spatial Representations of Vision and Touch in Early Life
Jannath Begum Ali

When we feel a touch, we do not simply loc­ate that sen­sa­tion on the body sur­face, but also in extern­al space. The “crossed-hands effect”, a defi­cit in the loc­al­isa­tion of tact­ile stim­uli to the hands when they are placed in the crossed-hands pos­ture, has been used as a mark­er of the influ­ence of an extern­al frame of ref­er­ence for loc­al­ising touch; it is con­sidered to arise out of con­flict (when the hands are crossed) between the ana­tom­ic­al and extern­al (non-anatomical) frames of ref­er­ence with­in which touches can be per­ceived. Because early visu­al exper­i­ence is implic­ated in the devel­op­ment of an extern­al frame of ref­er­ence I will also report on a study in which the devel­op­ment of an abil­ity to per­ceive visu­al and tact­ile stim­uli in a com­mon spa­tial loc­a­tion was examined.


Unseen Colour Constancy and the Relationship between Sensation and Perception of the External World
Robert Kentridge

What are ‘sen­sa­tion’ and ‘per­cep­tion’? A per­vas­ive view, held by most since the Ancient Greeks, is (with some vari­ation in the details) that sen­sa­tions are con­scious exper­i­ences of the effects that stim­uli from the out­side world have on our sense organs and that, from these sen­sa­tions, we arrive a mod­el of the objects in the out­side world that are likely to have eli­cited those sen­sa­tions – our per­cepts of the extern­al world. Critically, per­cep­tion depends upon sen­sa­tion. Recent exper­i­ments on col­our per­cep­tion sug­gest that this is not the case. My col­leagues and I showed that per­cepts of col­our as a prop­erty of objects in the world influ­ence beha­viour des­pite being unseen. They do so in a con­sist­ent man­ner even under changes in light­ing sug­gest­ing that it is an uncon­scious rep­res­ent­a­tion of the col­our of an object, rather than the effect of the object on sense organs (which will change with changes of light­ing), that is driv­ing beha­viour. As sub­jects have no con­scious exper­i­ence of the object they must have con­struc­ted a per­cep­tu­al rep­res­ent­a­tion of it that is not derived from sen­sa­tion.


Looking before Seeing: Directing Visual Attention by the Primary Visual Cortex
Zhaoping Li

Visual atten­tion lim­its the amount of visu­al inputs for detailed pro­cessing since
the brain has lim­ited resources. We mainly dir­ect our atten­tion by dir­ect­ing our gaze to atten­ded spa­tial loc­a­tions. Traditional wis­dom pre­sumes that brain areas at or near the front of the brain are in charge of dir­ect­ing visu­al atten­tion.
I show that for exo­gen­ous atten­tion­al guid­ance, i.e., atten­tion guided by extern­al visu­al inputs rather than intern­al goal, the primary visu­al cor­tex is respons­ible. Hence, atten­tion can be dir­ec­ted to loc­a­tions before objects at those loc­a­tions are seen or recog­nized. I will show some data as well as a demo.


The Two-Visual-Systems Hypothesis and the Perspectival Intuition
Robert Foley

In a recent art­icle, Berit Brogaard (2012) argues that the two-visual-systems hypo­thes­is (TVSH) (Goodale & Milner, 1992) is incom­pat­ible with the per­spectiv­al claim. This raises a prob­lem for psy­cho­func­tion­al­ists, who hold that the best answer to the ques­tion of what it is to instan­ti­ate a par­tic­u­lar men­tal state will be provided by our best cog­nit­ive psy­cho­lo­gic­al the­or­ies.  Given that the per­spectiv­al claim seems to cap­ture a fun­da­ment­al fea­ture of our visu­al exper­i­ence, Brogaard con­cludes that we should reject a cent­ral claim of the TVSH that is incom­pat­ible with the per­spectiv­al claim.In this art­icle, I argue that there is no con­flict between the TVSH and the per­spectiv­al claim. As such, the psy­cho­func­tion­al­ist can main­tain both the TVSH and the per­spectiv­al claim without fear of incon­sist­ency. In what fol­lows, I first out­line Brogaard’s argu­ment. I then argue that the appar­ent ten­sion between the dis­so­ci­ation hypo­thes­is and the per­spectiv­al claim res­ults from an ambi­gu­ity in the use of the terms ‘allo­centric’ and ‘ego­centric’. Finally, I argue that dorsal stream pro­cessing is not the right kind of pro­cessing to explain ego­centric fea­tures of con­scious visu­al exper­i­ence.


A Sense of Body Ownership
Hong Yu Wong

Our bod­ies and our sense of embod­i­ment are crit­ic­al to our sense of ourselves as mater­i­al beings (Cassam 1997, Longuenesse 2006). One prom­in­ent strand of research on embod­i­ment con­cerns the sense of own­er­ship that we have over our bod­ies. The key ques­tions are how to under­stand this sense of own­er­ship, and what its func­tion is. I begin by char­ac­ter­ising the sense of body own­er­ship and its rela­tion to basic forms of bod­ily aware­ness, such as proprio­cep­tion (Martin 1995, de Vignemont 2007, Peacocke 2014). A major issue is the shape of a con­stitutive account and its status with respect to patho­lo­gies where it is com­prom­ised, such as soma­to­pa­raphrenia and ali­en limb syn­drome (Vallar and Ronchi 2009, de Vignemont 2007 and 2011). In this talk, I will sketch an account of body own­er­ship that diverges from the three major accounts.


Ebbinghaus Illusion Seems to Deceive Literates but not Non-literates
Inês Hipolito

Theory of Mind devel­op­ment is depend­ent on the mat­ur­a­tion of sev­er­al brain sys­tems and is shaped by par­ent­ing, social rela­tions, school­ing, and edu­ca­tion. The emer­ging field of social neur­os­cience has also begun to address this debate, by ima­ging humans while per­form­ing tasks demand­ing the under­stand­ing of an inten­tion, belief or oth­er men­tal state. Studies about the effects of con­text on size per­cep­tion have giv­en rise to extens­ive debates con­cern­ing the con­di­tions under which illu­sion effects occur and how they are to be inter­preted, under the Theory of Mind scope. In this essay I will assess the effect of an illus­ory sym­bol­ic sys­tem, the Ebbinghaus illu­sion, in chil­dren and in non-human prim­ates. My aim is to under­stand wheth­er the deceiv­ing induced by Ebbinghaus illu­sion could be cor­rel­ated with lit­er­acy. I will start by review­ing the lit­er­at­ure on the physiolo­gic­al and func­tion­al impact of lit­er­acy in the brain, and then I will assess object recog­ni­tion in Ebbinghaus illu­sion in non-human prim­ates and in non-literate chil­dren. These two groups are chosen due to its men­tal age and non-literacy. The hypo­thes­is is that this illu­sion deceives lit­er­ate but non-literate. I expect this to be pos­sible in vir­tue of Dahaene’s Neuronal Recycling Hypothesis. Moreover, this pro­cess could be sim­il­ar to the mir­ror invari­ance pro­cess.


ROBO-GUIDE: Autonomous Navigation of Dynamic Spatial and Social Environments
David Cameron, Jonathan M. Aitken, Luke Boorman, Adriel Chua, Emily C. Collins, Samuel Fernando, Uriel Martinez-Hernandez, Owen McAree, & James Law.

Assistive robots will need to seam­lessly integ­rate into envir­on­ments that have been expressly designed with human-use in mind. For mobile robots, this not only com­prises accur­ate per­cep­tion and reli­able nav­ig­a­tion of the spa­tial envir­on­ment but also effect­ive social com­mu­nic­a­tion in shared-space social envir­on­ments. ROBO-GUIDE (ROBOtic GUidance and Interaction DEvelopment) is an inter­dis­cip­lin­ary pro­ject, bring­ing togeth­er psy­cho­logy, nat­ur­al lan­guage, com­pu­ta­tion­al neur­os­cience, con­trol sys­tems, and form­al veri­fic­a­tion to address how mobile robots can suc­cess­fully per­ceive and oper­ate in human-focused spa­tial and social envir­on­ments.


Two-Sided Map: A Critical Study of Geography 2.0
Verbena Giambastiani

Traditionally, car­to­graphy was a mean to cre­ate maps for ori­ent­a­tion. Right now, in a map is included stat­ist­ic­al, demo­graph­ic and eco­nom­ic inform­a­tion. In the near future, maps will be gen­er­ated in a highly per­son­al­ized way, high­light­ing places fre­quen­ted by our social net­work friends, places that we have talked about in the emails or places we have looked for in the search engines. This up-to-date scenery could be inter­preted in two dif­fer­ent was. On the one hand, Google’s rep­res­ent­a­tion of space is deeply con­ser­vat­ive and backward-looking. Places that we have nev­er searched online – or for which we have nev­er shown interest – would be rather more dif­fi­cult to be visu­al­ized. On the oth­er hand, an online map shows us that space is not a pass­ive «out there» con­di­tion, a sort of sub­stance that exists only out­side. space is a human cre­ation, a kind of internal-external rela­tion­ship. The aim is to explore these dif­fer­ent approaches.


iCog 3 | Programme