Can neuroscience solve the mystery of consciousness ?

Iden­ti­fying what conscious­ness is has become the Holy Grail of neu­ros­cience. But can conscious­ness be redu­ced to what can be obser­ved in a labo­ra­to­ry ? And what ethics should be applied to this new dis­ci­pline, which raises the dual hopes of signi­fi­cant medi­cal advances and sub­stan­tial finan­cial gains ? We dis­cuss this with Laure Tabouy, a neu­ros­cien­tist and ethi­cist who is pur­suing a second PhD in the ethics of neu­ros­cience, digi­tal tech­no­lo­gy, neu­ro­tech­no­lo­gies and arti­fi­cial intel­li­gence at the Centre Gilles Gas­ton Gran­ger (CGGG UMR 7304) at Aix-Mar­seille Uni­ver­si­ty, during which she is conduc­ting a cri­ti­cal ana­ly­sis of the neu­roe­thics impo­sed by the deve­lop­ment of neu­ro­tech­no­lo­gies and the conver­gence of neu­ros­cience and AI.

Laure Tabouy. Conscious­ness is one of the most com­plex areas of stu­dy, fal­ling within the scope of both phi­lo­so­phy and neu­ros­cience, as it encom­passes various inter­t­wi­ned dimen­sions : awa­re­ness of the out­side world and of one­self, the abi­li­ty to reflect on and ana­lyse one’s own thoughts and actions, the source of free will, and the capa­ci­ty for moral jud­ge­ment. Since Socrates, it has been one of the major sub­jects of stu­dy for phi­lo­so­phers and has become a focus for neu­ros­cience since the 1960s. Since the 2000s, the conver­gence of nano­tech­no­lo­gy, bio­tech­no­lo­gy, com­pu­ter science and cog­ni­tive science has rai­sed hopes that increa­sin­gly sophis­ti­ca­ted tech­no­lo­gies will make it pos­sible to iden­ti­fy the phy­si­cal mar­kers of conscious­ness, and even of reflec­tive conscious­ness. In paral­lel with these expe­ri­ments on brain acti­vi­ty, various neu­ros­cien­ti­fic theo­ries known as “theo­ries of conscious­ness” have flou­ri­shed in recent decades.

It is a so-cal­led func­tio­nal theo­ry, which is very inter­es­ting from a concep­tual point of view and is not incon­sistent with cer­tain obser­va­tions made in the labo­ra­to­ry. It does not des­cribe what conscious­ness is, but what it does in a per­cep­tible way. Accor­ding to this theo­ry, the brain func­tions like a theatre : conscious thoughts occu­py the front of the stage, also known as the “glo­bal works­pace”, while in the back­ground, spe­cia­li­sed auto­ma­tic pro­cesses loca­ted in the brain are constant­ly at work, pro­ces­sing sen­so­ry sti­mu­li. At any given moment, only the results of some of these pro­cesses are brought into the spot­light, the­re­by beco­ming acces­sible to all the auto­ma­tic neu­ral pro­cesses ope­ra­ting behind the scenes.

Pro­po­nents of this theo­ry pos­tu­late that the glo­bal works­pace is made up of neu­rons with long axons, capable of trans­mit­ting infor­ma­tion to very dis­tant areas of the brain. In their view, the emer­gence of conscious­ness would the­re­fore mani­fest itself through the acti­va­tion of these vast cere­bral networks.

Around thir­ty of them ! Of course, not all of them have the same reso­nance within the scien­ti­fic com­mu­ni­ty – or in the media and among fun­ding bodies. One of the main com­pe­ti­tors to the glo­bal works­pace theo­ry is the “inte­gra­ted infor­ma­tion theo­ry”, pro­po­sed by Ita­lian neu­ros­cien­tist and psy­chia­trist Giu­lio Tono­ni in 2004. Rather than star­ting from brain acti­vi­ty to iso­late the mar­kers of conscious­ness, it sets out a glo­bal theo­re­ti­cal fra­me­work of what conscious­ness is, sup­por­ted by a mathe­ma­ti­cal model. This theo­ry defines conscious­ness as an emergent pro­per­ty of any phy­si­cal struc­ture capable of inte­gra­ting infor­ma­tion, and it applies this defi­ni­tion not only to the brain, but also to any infor­ma­tion-pro­ces­sing sys­tem. The degree of conscious­ness of a sys­tem would thus depend on the amount of infor­ma­tion it is capable of pro­ces­sing and its abi­li­ty to com­pare this infor­ma­tion at dif­ferent levels, both regio­nal and global.

Again, this is not incon­sistent with what can be obser­ved local­ly in the brain. But it leads pro­po­nents of this theo­ry to consi­der arti­fi­cial sys­tems such as ther­mo­stats or pho­to­diodes as “conscious” – an exten­sion of the concept of conscious­ness about which many neu­ros­cien­tists have serious objections.

This theo­ry has been tal­ked about as much for its novel­ty, even pro­vo­ca­ti­ve­ness, as for the fact that 124 resear­chers have label­led it pseu­dos­cience¹ in a pre­print (editor’s note : a ver­sion of a scien­ti­fic publi­ca­tion that pre­cedes its accep­tance by the edi­to­rial board of a scien­ti­fic jour­nal) publi­shed on the PsyArXiv plat­form. Howe­ver, this docu­ment remains rather light­weight from a scien­ti­fic point of view and is itself high­ly contro­ver­sial within the neu­ros­cience community.

No. A recent col­la­bo­ra­tive stu­dy² com­pa­red the glo­bal works­pace theo­ry and the inte­gra­ted infor­ma­tion theo­ry accor­ding to a pro­to­col esta­bli­shed by a consor­tium that claims to be neu­tral in theo­re­ti­cal terms. Their results confirm cer­tain pre­dic­tions by both theo­ries but also call into ques­tion some of their key prin­ciples. Ulti­ma­te­ly, this is not very sur­pri­sing. Neu­ros­cience is still a rela­ti­ve­ly new aca­de­mic dis­ci­pline. It has yet to find its inter­nal theo­re­ti­cal conver­gence, and this conver­gence will like­ly come about through the com­bi­na­tion of seve­ral theories.

Popu­lar neu­ros­cience often takes a reduc­tio­nist approach, mea­ning that it views its sub­ject (conscious­ness) as the result of sub­sys­tems (brain pro­cesses or the orga­ni­sa­tio­nal struc­ture of infor­ma­tion, for the two theo­ries men­tio­ned). Reduc­tio­nism is not a pro­blem in itself : it allows us to define a fra­me­work that makes the expe­riment pos­sible. This is easi­ly unders­tan­dable, given that it alrea­dy places a prism bet­ween us and rea­li­ty, through which cer­tain dimen­sions of conscious­ness may escape.

But the most wide­ly publi­ci­sed neu­ros­cien­ti­fic theo­ries – inclu­ding, above all, the glo­bal works­pace theo­ry and, to a cer­tain extent, the inte­gra­ted infor­ma­tion theo­ry – are also based on radi­cal mate­ria­lism : they assume that conscious­ness can be redu­ced to phy­si­cal pro­cesses. This is a major phi­lo­so­phi­cal assump­tion, which can – and must – be ques­tio­ned. It is not itself the result of scien­ti­fic consen­sus, and phi­lo­so­phy, from which it is deri­ved, offers a wealth of alter­na­tive models to explain what conscious­ness is : idea­lism, cer­tain forms of plu­ra­lism, dua­lism (par­ti­cu­lar­ly Car­te­sian dua­lism), spi­ri­tua­lism, etc. In fact, there is no rea­so­nable basis for defi­ni­ti­ve­ly favou­ring a mate­ria­list concep­tion of consciousness.

It seems to me to stem from the very condi­tions in which neu­ros­cience itself was born. It was the emer­gence of extre­me­ly power­ful tech­no­lo­gies for obser­ving brain acti­vi­ty that moti­va­ted its emer­gence. The incre­dible effec­ti­ve­ness of these devices has, in a way, blin­ded us : we have confu­sed what we are able to see now or will be able to see in the future – which is, indeed, incre­di­bly rich – with rea­li­ty as a whole. The search for fun­ding also plays a role in this stance : it is more mar­ke­table to say that we are going to get to grips with conscious­ness than to announce that we hope to make pro­gress in obser­ving some of the phy­si­cal phe­no­me­na lin­ked to the conscious acti­vi­ty of the brain…

It becomes pro­ble­ma­tic when it is consi­de­red the only way to access the truth of what conscious­ness is. In this case, it is an ideo­lo­gi­cal stance that pro­found­ly dis­torts ethi­cal debates by crea­ting a bia­sed cli­mate for reflec­tion. Based on this mate­ria­list pre­mise, some people are begin­ning to talk about the pos­si­bi­li­ty of down­loa­ding conscious­ness in the future, for example. This trans­hu­ma­nist ideo­lo­gy is tech­nos­cien­ti­fic in nature and is alrea­dy stee­ring research, poli­ti­cal and finan­cial choices in a direc­tion that is high­ly ques­tio­nable from an ethi­cal stand­point. Neu­ro­tech­no­lo­gy com­pa­nies are clear­ly adop­ting this bias by announ­cing that they can ‘read your brain’ or ‘deci­pher your brain waves to exploit your unsus­pec­ted abilities’.

The pros­pect of humans control­ling their own conscious­ness is so fas­ci­na­ting that it obs­cures other pres­sing issues. How far are we willing to go in arti­fi­cial­ly modi­fying the human brain ? How can we assess the impact of neu­ro­tech­no­lo­gy on human evo­lu­tion, and is this where we col­lec­ti­ve­ly want to go ? And the ques­tion that drives my the­sis : how can we free neu­roe­thics from the poli­ti­cal and eco­no­mic tech­no­so­lu­tio­nist stran­gle­hold impo­sed by the deve­lop­ment of neu­ro­tech­no­lo­gies and the conver­gence of neu­ros­cience and AI ? It is cru­cial that ethics be a part of this dis­cus­sion, which is not real­ly the case at the moment.

The ethics of neu­ros­cience is a recent deve­lop­ment : it can be said to have emer­ged as a branch of ethics in its own right only in 2002, at the San Fran­cis­co confe­rence on ‘neu­roe­thics’. The texts you refer to were writ­ten in the context of geo­po­li­ti­cal tur­moil asso­cia­ted with the launch of huge brain research pro­jects, such as the Human Brain Pro­ject⁵ under­ta­ken on the ini­tia­tive of the Euro­pean Com­mis­sion and the Brain Ini­tia­tive⁶ laun­ched by the Oba­ma admi­nis­tra­tion. These texts are less the result of genuine ethi­cal reflec­tion, ques­tio­ning the foun­da­tions and conse­quences of ongoing tech­no­lo­gi­cal deve­lop­ments, than an attempt to keep pace with these deve­lop­ments, dri­ven essen­tial­ly by mar­ket forces.

Dis­sen­ting voices exist among phi­lo­so­phers, ethi­cists and neu­ros­cien­tists them­selves, but they are cur­rent­ly being sti­fled… A genuine ethi­cal reflec­tion must be based on a real phi­lo­so­phi­cal, anthro­po­lo­gi­cal and cultu­ral contro­ver­sy. It must be borne in mind that Wes­tern mate­ria­list theo­ries are far from uni­ver­sal. Beyond the phi­lo­so­phi­cal cur­rents alrea­dy men­tio­ned, almost all spi­ri­tua­li­ties and reli­gions have a non-mate­ria­list view of conscious­ness. This should prompt us to shift our theo­re­ti­cal focus. Ethics must also remain inde­pendent of the finan­cial inter­ests lin­ked to the tech­no­lo­gi­cal deve­lop­ments it ques­tions. And final­ly, it must keep its focus on the one ques­tion that ulti­ma­te­ly mat­ters : are these tech­no­lo­gi­cal deve­lop­ments tru­ly ser­ving huma­ni­ty in all its dimensions ?

Interview by Anne Orliac