Category: Cognitive Models

I mentioned in a recent post Murray Shanahan’s model of global broadcast using spiking neurons. In brief, Murrays model contains distinct groups of neurons (which may for example represent different sensory modalities) each of which may encode for many different responses. These groups of neurons are all connected together via a comunications infurstructure which he calls the global workspace. Different cell assemblies in different groups fight for contol of the global workspace and once one particular set gains control its influence is broadcast through the workspace to the entire system.

What is most notable about this model is that the comunications inferstructure is generic and does not contain any specifically meaningfull connections between particular nodes or cell assemblies in the different groups that one may wish to communicate with each other or behave in some kind of complementary fashion. The question therefore arises as to how the different cell assemblies in seperated groups know what they are responding to and how to respond in the appropriate way? I recently put this question to Murray. In his response he was quick to point out that he does not wish to propose a model in which the signals that pass through the global workspace contain information in some kind of language of the brain. However, without meaningful connectivity it is necissary that the activations patterns being passed along this generic communications infurstucture do contain information that allows the recieving cell assemblies to respond appropriately. Murray suggest that the recieving cell assemblies adapt to repond to particular activation patterns in particular ways that are behaviourally beneficial. In addition the cell assemblies which send the signals will adapt in order to take advantage of responses from recieving cell assemblies that are also behaviourally beneficial.

Sounds pretty plausable to me, but pretty a complicated adaptation task as adaption will have to work from bottom up sensory inputs as well as top down global workspace inputs. Nevertheless working out how such sub-systems can adapt to comunicate with each other in this ways sounds like a really cool and interesting area of research.

Theoretical “blackboard” cognitive  architectures originate from AI and  have been around for decades. They propose that many specialized processes compete for space on a centralized blackboard on which information is shared and coordinated. I have never been keen on them because I never thought they particularly sat well with my preference for neural systems. Others such as Mumford disagree. In particular, he postulates that  the thalamus is the blackboard. His reasons for doing so are to explain the reciprocal connections between the thalamus and the cortex. Most input to the cortex passes through the thalamus, but the cortex also sends information back down to the thalamus. For Mumford, specialist areas in the cortex are seen to process data given them by the blackboard thalamus and then return their results. Others have suggested the thalamus as a mere relay, and some suggest the feedback is used in order to facilitate a gating mechanism on information being passed. For Mumford, gating in this way is overkill. However,  his concession that coordination between cortical areas can also occur via cortico-cortical pathways suggest that that the addition of a blackboard thalamus is itself overkill.

Further to this, the idea of a brain region being a blackboard suggests that high level conceptual information is registered and exchanged within the region. Interconnectivity within the thalamus is low and this concerns me as I would expect that a region in which high levels of exchange and voting occur would be quite densely connected.  I also feel that design suggestions such as these fall victim to a homunculi style problem, in which the thalamus becomes the object to which brain information is presented. If the brain is the organ in which sensory data is presented/conceptualised then why postulate a further area within the brain in which a final presentation occurs? For me it seems to contradict the principal of Occam’s razor. What do we suggest next, yet another area in which higher levels of processed blackboard data are presented, add infinitum? It seems that it is simpler to just postulate the specialized processing areas being the final presentation/conceptualization and try to figure out how they communicate with each other, bind, and form a unity of thought. Problems otherwise only deferred to within the thalamus.

I have never given much time to global workspace theory which owes its pedigree to blackboard architectures because I naively presumed them to be much the same. My concerns being those mentioned above as well as the presumption of discrete specialist processes when neural cell assemblies by and large do not encode in a discrete manner but often exhibit topologically fuzzier population coding. But about a year ago I came across Murray Shanahan’s spiking neuron model of global broadcast. In this he suggest recasting in terms of a communications infrastructure. In this model the specialist processes are not  discrete but more overlapping as they should be in a neural system. Most noticably recognition is given to the corona radiata which relays signals to cortex from the thalamus and back to the thalamus, but also the corpus callosum which carries traffic between the two cerebral hemispheres, as well as several fibre tracts directly connect remote parts of cortex to each other. The blackboard reformulated as a global workspace is the connectivity between specialist cortical areas, and most importantly for Shanahan this facilitates global broadcast of information from cell assemblies winning competition to exhibit influence over the entire system. The implementation details of his model raises a few questions (which I shall attempt to relate and resolve in a later post) but the principle behind the model is very attractive. His entire approach and philosophy behind it is discussed in his soon to be released book ‘Embodiment and the Inner Life’ which I have been fortunate enough to preview and recommend all to read.