Archive 16

ARCHIVE 16

25 January - 15 April 2012

Modern theories of consciousness are criticised for taking little account of olfaction (smell). The author thinks that the nature of the olfactory system could falsify some fashionable theories of consciousness because smell is the only sensory input which does not enter the cortex via the thalamus. This is a problem for those theories of consciousness that propose that consciousness arises exclusively from feedback loops between the thalamus and the cortex.

Recent research suggests that there are two olfactory routes to the cortex. The direct pathway leads from the initial receptors via the piriform cortex to the orbitofrontal cortex. A more indirect pathway also runs via the piriform cortex, but then goes to the thalamus before projecting to the orbitofrontal. This indirect pathway may be required for analysing or specifically identifying smells, and thinking about and deciding on relevant actions. However, it is not required for detecting odours or discriminating between different odours or their different intensities. These depend on the direct path that does not go via the thalamus.

The author suggests that some claims for the role of the indirect thalamic pathway may be overstated. Some tests that emphasise the role of the thalamus involve the conscious decision to sample particular smells. This is accepted as suggesting the importance of the thalamus when selective attention is required. However, the thalamus is not necessary for conscious awareness of smells in the subject's environment. The author considers and rejects the claim that the olfactory bulb functions in a manner equivalent to the thalamus. He considers that not only does the olfactory system not depend on the thalamus, but also that it does not contain anything that is functionally equivalent to the thalamus.

15 APRIL 2012
KAUFFMAN'S POISED REALM

Reinventing the Sacred

Stuart Kauffman (2008)

INTRODUCTION: From the point of view of consciousness studies, Kauffman's 'poised realm' is the most interesting aspect of his work. He identifies the poised realm with conditions in photosynthetic organisms where quantum coherence begins to decohere, but is forced back into partial coherence. His hypothesis is that consciousness is found in this border area of decoherence and recoherence, which is suggested to have the potential to supplement the deterministic algorithms of classical physics.

Kauffman is critical of the reductionist trend of science as practised during the last three centuries. He argues that reduction of the world to particles in motion does not by itself supply sufficient explanation in all instances. He holds that the organisation of complex areas such as evolution and biology cannot be deduced just from the existence of particles and their governing laws.

Kauffman further argues that given the basis of the physical laws, there is an infinite number of ways in which the quanta could be arranged, of which, the organisation of proteins and organisms is only a very limited subset. So the evolution of the present life forms is not a deterministic certainty, and the existing life forms and other aspects of the material world are only a small subset of the possible outcomes.

Origin of life:
Kauffman hypothesises that life originated from a simpler replicator than either DNA or the now popular idea of RNA. A helix of 32 amino acids, has been shown to be capable of binding together two smaller chains of 17 and 15 amino acids, and thus forming a copy of the initial sequence of 32 amino acids. If substantiated, this would demonstrate an emergent system of replication. Furthermore it is implied that many molecular systems could achieve this, so there are many paths to the origin of life that are allowed by the laws of physics, rather than a single deterministic path.

Life, even it is argued primitive life, involves agency which does not exist in physics. Agents are seen as emerging from the evolutionary process. Life is seen as introducing agency into the universe, and agency in its turn introduces meaning and value. The act of choosing between different types of behaviour is seen as requiring an agent. But in physics there are no such agents.

It should be stressed that Kauffman's position appears to be different from what might be termed the 'naïve emergent property theory', which has been a prominent theme in mainstream consciousness theory. This merely asserts that consciousness is an emergent property without showing how it emerges from a more fundamental level, as can be clearly demonstrated for other emergent properties such as the liquidity of water.

Far from thermal equilibrium:
Kauffman argues that to understand agency it is necessary to understand the Carnot work cycle in thermodynamics. The work-cycle system consists of a hot and a cold reservoir, plus a cylinder and piston with a compressible gas. Heat from the hot reservoir expands the gas and pushes the piston down the cylinder. The piston is pushed down the cylinder until the gas cools, and contracts again as a result of the cold reservoir at the other end of the cylinder. This power stroke is described as spontaneous because it does not require the injection of any energy into the system. A ball rolling down a hill is another example of spontaneous action. What does require an injection of energy is pushing the piston back up the cylinder, and this process is described as non-spontaneous. In other words, the system has to be reset. The usefulness of this type of steam engine is that it takes less energy to recompress the cooled gas than the amount of energy obtained from the hot gas.

The image of the Carnot steam engine is important in relation to the functioning of biological tissues, because these are both systems far from thermal equilibrium. The Carnot engine cannot work within a state of thermal equilibrium because it requires the heat on the cylinder to expand the cooler gas. It is stressed that in both the steam engine and organic processes, there is a need to reset the mechanism so that it can continue its work cycle.

Agency is seen as the biological feature that allows resetting of the work cycle. Kauffman examines the primitive example of a bacterium swimming up the glucose gradient. Receptors for glucose signal the glucose gradient. For the bacterium the meaning of the gradient signal is more glucose in a particular direction, and the bacterium interprets this meaning by swimming up the gradient.

Information:
Kauffman is critical of the concept of 'information' as applied to organisms, claiming that the concept is both restrictive and unclear. He points out that Shannon conceived of information as involving a source, a channel and a receiver. Information sent down the channel reduces the uncertainty of the receiver. It is pointed out that Shannon never defines information, but leaves the 'receiver' to decide this. A bacterium is seen as a receiver of information about the glucose gradient, and it responds to the information. Kauffman argues that the information approach ignores the physics of work and energy, and focuses narrowly on information transmission from DNA to RNA to protein. He regards this as being inadequate to explain the functioning of the cell.

Kauffman views the role of the cell as being to constrain activity so there are only limited degrees of freedom, and this constraint requires work to achieve it. In the steam engine, it requires work to assemble the piston and cylinder that provide constraints. Cells are thus seen as building their own constraints or boundary conditions. Organisms are viewed as an interwoven web of work and constraints.

It is further argued that information involves an agent, and that this agent is a constraint, so that the information is given meaning or interpretation. The advantage here is that there is not just one response to one stimulus as in an automation, but a choice of different responses according to context. The ability to discriminate between stimuli is argued to be a state poised between order and chaos, where order always gives the same answer for stimuli, despite varying outcomes from these stimuli in the past, and chaos gives a random outcome that is of no value. Kauffman's approach resonates with other recent research on brain areas such as the orbitofrontal, anterior cingulate, the basal ganglia and also the function of dopamine. These are important in providing the brain with a flexibility in its response to stimuli which could not be prestated because of the enormous number of possible outcomes.

The problem with computers:
In the foundational stage of computing, Alan Turing understood that he could write down symbols, modify then according to a set of rules, and eventually after perhaps a series of such modifications, come to an answer. Turing reconstituted this process as an idea for a machine that read symbols on a tape and made a rule-governed move, according to what the symbol was. It transpired that this machine could carry out all possible computations.

The computing/algorithmic systems, to which this approach gave rise, have had some successes at the level of robotic machinery. With defined objects in a defined setting, algorithmic systems can solve specific problems, such as a robot finding a source of electricity in a room. In this case, the solutions for the robot have been prestated, or predefined by the programmers, who know how it should respond to objects in the room. However, it is not possible to prestate all the conditions that may be faced by humans and other complex organisms. This would lead to an explosion in the number of possibilities that needed to be precalculated. The robot is provided with a frame limitation, but the explosion of options suggests that this is not feasible for humans which in turn suggests the brain is at least in part non-algorithmic.

Categorisation:
Kauffman also looks at the problem of categorisation. Humans place things in categories all the time, and two very dissimilar things can be consigned to the same category. Thus robins and penguins are both in the bird category. Plato was possibly first to discuss the problem of categories, suggesting that members of categories shared at least one essential feature. Wittgenstein, however, argued that categories might have no common features. This leaves us to look for similarities but it is not clear which similarities allow membership of which category. A bird may be the same size, weight and colour as a ball, but that does not put the ball into the bird category. Kauffman suggests that these difficulties may be overcome if categorisation is non-algorithmic.

Consciousness & the poised state:
In the final stage of his book, Kauffman argues that consciousness derives from a 'poised state' between quantum coherence and decoherence into classical states. He looks to the transition from a quantum world of persisting possibilities to a classical world of actual possibilities. The acausal nature of quantum mechanics is central to his thinking. The Schrödinger equation is solved for the amplitude of the electron at each point in space. These eigenfunctions square the amplitude at each point in space, and define the probability of finding an electron at each point in space. Nothing known causes the electron's choice of position, there are only probabilities at every point in space. For Kauffman, quantum mechanics breaks out of the causal closure of the reductionistic tradition. Amongst other things he suggests that this might resolve the problem of freewill, which cannot exist within deterministic physics.

Kauffman discusses the concept of phase information. The interference pattern seen in the two-slit experiment requires all the phase information on the final screen to add together to give the peaks and troughs of the interference pattern. Decoherence involves the loss of phase information as a result of interaction with the environment, often described as a heat bath of quantum oscillators. The interaction with the environment in seen as comparable to the interaction with the measuring device in the Copenhagen interpretation. However decoherence may not be as clear cut as the Copenhagen type measurement. In certain circumstances, only part of a system decoheres and some coherence remains.

Kauffman places consciousness at this 'poised state' where part of the system decoheres and part is coherent. The coherent state is suggested to influence the classical decoherent state. In looking for such a system, Kauffman examines the recent research on photosynthetic systems. In photosynthesis photons are captured by the chlorophyl molecule that is held by antenna protein. The chlorophyl molecule maintains quantum coherence for up to 750 femtoseconds. This is longer than the classical prediction, and is viewed as responsible for the higher than classically predicted efficiency of energy transfer. The antenna protein plays a role in preventing more rapid decoherence, or in inducing recoherence in decohering parts of the chlorophyll molecule. Part of the quantum system may start to decohere, but be forced back into coherence, sometimes described as quantum error correction.

Within the chlorophyll molecule the superposition of the Schrodinger solutions allows the simultaneous exploration of all the possible pathways. This is more efficient than the serial or one-path-at-a-time exploration, and is taken as an explanation for the mid 90 percentage efficiency of the system, in contrast with the 60-70% predicted for a classical system.

Kauffman thinks that the system seen in the chlorophyll molecule raises the possibility that webs of quantum coherence or partial coherence can extend across a large part of a neuron, and can remain poised between coherence and decoherence. Kauffman's discussion refers to coherent electron transport, but he recognises that other forms of coherence such as phonons and electron spin could be relevant.

The 'poised state' is supposed to span states that are between being mainly coherent and partly decoherent. Information injected into the system can induce recoherence. The flow of information into cells is seen as a means by which recoherence could be induced and coherence maintained. In other writing, Kauffman suggests a two-way flow of influence, with quantum possibilities effecting classical systems, while classical systems could influence recohering quantum systems.

In relating quantum coherence to consciousness, Kauffman assumes like Hameroff that coherence would have to be sustained for the milliseconds timescales associated with neural processing, rather than the femto and picosecond timecales associated with quantum coherence in photosynthetic organisms. It might be debatable if a direct one-to-one correlation between processing activity and conscious episodes is necessary.

2 APRIL 2012
SUBJECTIVE VALUES IN THE VENTROMEDIAL

Contributions of ventromedial prefrontal and frontal polar cortex to reinforcement learning and value-based choice

Erie D. Boorman & MaryAnn Noonan

In:- Neural Basis of Motivational and Cognitive Control

INTRODUCTION: This chapter more than once stresses the importance of both expected and experienced subjective value of reward, thus implicitly conflicting with mainstream consciousness theories that assign no role to conscious experience.

This chapter consider how agents select options for behaviour. The values that decide the selection of options are often learnt. The authors consider the role of the ventromedial prefrontal and the orbitofrontal. The orbitofrontal is viewed as being particularly involved in flexible behaviour. Its main role is argued to be making predictions about the outcome of behaviour relative to stimuli in the environment, and also evaluating errors in reward prediction. This is seen as relating to learning and to future choice of stimuli, but not to the immediate selection of actions. The orbitofrontal is further involved in encoding a reward as having been the result of a particular choice of behaviour.

The orbitofrontal is strongly connected to areas of the sensory cortex responsible for assembling representations, such as the inferior temporal, somatosensory and olfactory. There are also strong connections to the anterior cingulate. A distinction is made between the lateral orbitofrontal that is seen to encode the expected subjective value of stimuli, assigning credit for particular rewards to particular action and contributing to learning, and the medial orbitofrontal that is more related to the comparisons of different values.

Subjective value:
The decrease in the firing rate of orbitofrontal neurons following satiety in feeding indicates that the orbitofrontal encodes the subjectively experienced value of internal states in relation to rewards, according to the level of such internal states as hunger. Studies suggest that the orbitofrontal encodes the probability and magnitude of rewards and the preferences of the subject. The region appears to code for both the expectation and the subsequent experience of rewards, and is seen as encoding rewards of current relevance. P. Activity in the ventromedial prefrontal, adjacent to the orbitofrontal cortex, also correlates with the subjective value of both expected and experienced rewards. The ventromedial has been shown to encode the subjective value of experienced rewards for a range of sensory stimuli. Its correlation with subjective values is seen as confirming the response to satiety in feeding already detected in the orbitofrontal. Parts of the ventromedial correlate with expected values, including abstractions such as monetary values. Activity in this area is thought to match the value of an option during decision making. It can, for instance, reflect trust in a third party's opinion relative to the subject's own experience and also the value of different choices of action. This area may code for such features as delay in receiving, probability of receiving and also size of reward. In trials conducted by the authors, the ventromedial has been seen to encode the expected value of a chosen option and an option that was not chosen. This could apply to rewards that were not directly comparable and to the gain or loss of such rewards. However, the cost in effort needed to achieve a reward is seen to involve the anterior cingulate rather than the ventromedial. Such, evaluation of cost is not found in the orbitofrontal or the ventromedial.

28 MARCH 2012
FREE WON'T

Fronto-basal-ganglia circuits for stopping action

Ian Greenhouse, Nicole Swann & Adam Aron

In:- Neural Basis of Motivational and Cognitive Control

INTRODUCTION: This chapter describes a neural network that can stop initiated actions. This qualifies the naïve interpretation of the Libet experiments that ascribes all choice to readiness potentials that proceed conscious awareness of the decision to make (trivial) actions. Libet himself suggested there might be a 'free won't' that could override the automaton type readiness potentials and recent neuroscience looks to support this suggestion.

The authors consider situations in which a subject has already begun an action, but stops in response to an external stimuli. Many studies suggest that a particular fronto-basal-ganglia circuit is responsible for this control. It is suggested that in this situation sensory information relative to external stimuli is projected to the prefrontal, and particularly the areas of the right inferior frontal and the presupplementary motor area, and that these areas send a 'stop' command via the basal ganglia. Part of the inferior frontal may relate to attention, and a part to inhibition. The subthalamic nucleus is seen as a region of the basal ganglia that is suited for a 'stop' signal. It is well placed to increase inhibition of thalamo-cortical output. It is thought capable of inhibiting both basal ganglia and motor system output. The subthalamic nucleus receives inputs from the inferior frontal and the presupplementary. The inferior frontal, the presupplementary and the subthalamic are viewed as a connected functional network sending a fast inhibitory signal to the motor system. Stopping an initiated action is also seen to involve increased GABA activity. A specific oscillatory frequency in the beta band also appears to be involved. The 'stop' process appears to require a degree of preparation and the targeting of a specific action.

17 MARCH 2012
MEMORY IN MICROTUBULES
Molecular match for CaMKII phosphorylation encoding of microtubule lattices

Stuart Hameroff, Travis Craddock & J.A. Tuszynski

Journal of Integrative Neuroscience, vol 9, no. 3 (2010) 253-267 - Imperial College Press DOI: 10.1142/S0219635210002482

Learning and the formation of memories are supported by the post-synaptic flux of calcium ions which activates the hexagonal calcium-calmodulin kinase enzyme. Two groups of six kinase domains can be phosphorylated and these are able to phosphorylate other proteins. This is viewed by the authors as potentially encoding memory within neurons.

Microtubules are regarded as the best candidates for intraneuronal memory storage. The authors use molecular modelling to show that the spatial dimensions and geometry of the kinase domains exactly match those of hexagonal lattice neighbourhoods on microtubules. They suggest that this arrangement creates a phosphorylation mechanism.

Phosphorylation prolongs calcium-calmodulin kinase (CaMKII) activity, and the authors suggest that this allows the memory of synaptic activity to be encoded in the CaMKII. The calcium-calmodulin kinase in shown in the author's modelling to overly both the microtubule A and B lattices. The enzyme interacts with six tubulins in the microtubular lattice. The encoding here is according to whether or not the CaMKII is phosphorylated or not, as a form of binary system.

The authors suggest that neither the original flux of calcium ions nor the subsequent activation of calcium-calmodulin is sufficiently stable to comprise memory storage that may last a life time. Microtubules are here suggested as a possible end point beyond both the calcium flux and CaMKII, where permanent storage can occur. The authors appear to consider the microtubule lattice to be the most likely site for memory storage.

It is further suggested that memory-related information in microtubules could subsequently determine neuronal and synaptic structures, and by means of stored information regulate the initial firing of axons.

CONCLUSION: The authors do not discuss the implications for consciousness in this paper. However the connection to axonal regulation and the similarity of the hexagonal lattice system to that proposed for consciousness in dendrites must suggest the possibility of conscious activity here as well.

11 MARCH 2012
DECISION MAKING IN THE FRONTAL CORTEX
Decision making in frontal cortex: From single units to fMRI

Steven W. Kennersley & Philippe N. Tobler

In:- Neural Basis of Motivational and Cognitive Control

INTRODUCTION: The authors discuss the role of three frontal brain regions, the orbitofrontal, the anterior cingulate and the lateral prefrontal. Although not mentioned as such, the work is interesting in relation to the involvement of subjective conscious experience in determining the decision making discussed here, and its apparent relationship to variations in the firing rate of single neurons.

Decisions by humans and other animals require the consideration of multiple influences and possibilities. The decision of which of a choice of foods to eat, or how to reach a required destination is often influenced by a number of potentially conflicting factors. In the simple case of decision to eat, the outcome is at least influenced by the internal state (how hungry), longer-term goals such as future health, and cost factors such as the effort etc. needed to obtain the food. The brain then has to decide which of a number of options best meet the various needs and goals.

A particular problem is to assign values for, or have a common neural currency for, experiences that do not appear to have a common currency, such as the taste of a food and the energy required to obtain it. The currency also needs to be able to compare the very different outcomes of the various possible courses of action.

Evidence from the research of recent years combines to suggest that the frontal cortex has a role in decision making. The authors emphasise the role of the anterior cingulate, the lateral prefrontal, the orbitofrontal and also the ventromedial prefrontal. These are suggested to be involved in representing internal states, representing external variables, assigning values to actions and selection of options based on these action values.

Studies show, for instance, that activity in the orbitofrontal and ventromedial varies according to whether there is hunger/thirst or satiety. Internal reward representation also changes in response to experience. A food substance that used to be associated with a good experience can be devalued and avoided by subjects after earlier rewards fail to be repeated. However, this change does not happen with subjects that have damage to the orbitofrontal. Lateral and central areas of the orbitofrontal receive projections mainly from the sensory cortex, which could explain the sensitivity to devalued rewards. The medial orbitofrontal is linked more to regions with a motor function such as the anterior cingulate.

Single neurons in the orbitofrontal, anterior cingulate and lateral prefrontal are sensitive to the size and/or probability of reward and the time until reward. There was considerable variability in the function of single neurons. Given three decision variables, some neurons coded for the value of a single option, some for two-out-of- three possibilities and some for all three possibilities. The authors found no evidence that any of the three frontal areas studied were specialised for dealing with particular decisions. However, single neurons in the anterior cingulate could encode up to at least three decision variables, suggesting that this brain region might integrate the values of different components of a decision. The lateral prefrontal increased activation for reward probability and magnitude and their combined value, and also integrated the levels of value and risk attached to an action. The medial orbitofrontal processes reward probability and the lateral orbitofrontal processes risk. The anterior cingulate appears to be particularly related to assessing the cost as opposed to the potential value of actions.

The authors see the orbitofrontal as relating to expectancies for the outcomes of particular sensory stimuli, while the anterior cingulate relates to the value of actions. The anterior cingulate and the lateral prefrontal project to the pre-motor area, while the orbitofrontal is most notable for inputs from the sensory areas.

Some single-neurons adapt their rate of firing to the range of possible outcomes and also to the type of outcome being assessed. These neurons have been identified in the orbitofrontal and to striatal and dopamine producing regions, and there is also some evidence for this type of firing in the anterior cingulate. The authors speculate as to whether this type of range sensitivity could be more widespread.

29 FEBRUARY 2012
RESETTING ENTANGLEMENT AT HIGH TEMPERATURES

Steady state entanglement in open and noisy quantum systems at high temperature

L. Hartmann, W. Dur, & H.J. Briegel, Innsbruck University

Phys Rev A, vol. 74, issue 5 (dated May 15 2011)

INTRODUCTION: This paper is significant in moving away from a Tegmark type orthodoxy of rapid decoherence in high temperature systems, towards a recognition that systems far from thermal equilibrium, such as biomolecules, are capable of resetting entanglement by drawing new particles from the environment.

This paper demonstrates how quantum entanglement can be sustained in open and noisy environments that are far from thermal equilibrium, despite the tendency of such systems to decoherence. Such a system has a large number of interacting particles, and can also interact with and exchange particles with the environment. The impact of decoherence is counteracted by resetting of some of these particles to their initial state.

The situation of entanglement in macroscopic solids and fluids with large numbers of particles interacting with one another and the environment has been unclear. Studies have shown that entanglement is present in such systems, but this only referred to very low temperatures at which particles were close to their ground state. The main question which the authors address here is what happens in systems that are far from thermal equilibrium and exchange particles with the environment. This involves systems where individual particles are subject to decoherence, and particles are exchanged with the environment. Such systems include biomolecular processing within cells. The past expectation, as with the well-known Tegmark (2000) paper, is that decoherence would quickly destroy entanglement in such a system.

The authors identify a mechanism which can allow entanglement in systems that are not close to their ground state. This involves particles from the environment replacing particles in the system. In combination with particles already in the system, the 'fresh' particle is able to create entanglement. The system is described as being coupled to two reservoirs, a high temperature reservoir creating decoherence, and a second low temperature reservoir from which 'fresh' particles are drawn. This described a system far from thermal equilibrium.

All the authors results, which also involved simulating the system on a computer, produced the same conclusion that entanglement could persist in systems that are far from thermal equilibrium. Thus there is a sharp distinction made by the authors between systems at thermal equilibrium where entanglement can only occur at very low temperatures, and those that are far from equilibrium where entanglement can be sustained at high temperatures. Entanglement is suggested to persist in the middle ground where it has time to build up, but decoherence is not too fast. In the case of biomolecules that are far from thermal equilibrium, and where there are fluctuations in the number of particles in the system, 'fresh' particles from the environment are seen as likely to be responsible for the reset mechanism.

27 FEBRUARY 2012
ANTERIOR CINGULATE

An integrative theory of anterior cingulate cortex function: Option selection in hierarchical reinforcement learning

Clay Holroyd & Nick Yeung

In:- Neural Basis of Motivational and Cognitive Control

This chapter discusses the role of the anterior cingulate cortex. This brain region can be seen as part of a circuit involving the orbitofrontal, the ventral striatum, the dopamine neurons and finally the dorsolateral prefrontal. From the point of view of subjective consciousness, the orbitofrontal can be seen as injecting a subjective consciousness element into the final selection of extended behaviours by the anterior cingulate, which in turn acts on the executive area of the dorsolateral prefrontal.

The anterior cingulate cortex (ACC) is regarded as an important brain area for cognitive control. The ACC receives inputs from the limbic areas, the orbitofrontal and the midbrain dopamine neurons, and has dense projections to the motor cortex. It is seen as part of an 'executive network' involving other regions of the sensory and frontal cortex. There are four main theories, which to a good degree overlap, concerning the function of the anterior cingulate. Firstly there is a concept of monitoring behaviour, error detection and response to this. Secondly action selection focusing on the wilful generation of behaviour. Thirdly, reinforcement learning focused on learning to select particular behaviours. Finally, some theories focus on the ACC's role in determining the cost of behaviours.

A unified theory for the role of the anterior cingulate has yet to be developed, but the authors try to move towards one in this chapter. One hypothesis is that the ACC monitors for when more cognitive control of actions is required, although activity in the region can increase even without errors. It is variously suggested that the ACC could predict the likelihood of errors or conflicts between actions, and signal the need for the increased involvement of the dorsolateral prefrontal.

Conflict-related ACC activity in one experimental trial is predictive of more dorsolateral activity in the next trial, which in turn leads to improved performance suggestive of a feedback loop between the anterior cingulate and the dorsal prefrontal. However, this is not the extent of the ACC's role as shown by neural imaging, which also points to sustained activity related to task preparation and execution. The anterior cingulate is also considered to learn about the consequences of internally generated actions through dopamine projections.

The anterior cingulate has for a long time been seen as part of the limbic circuit. It is seen as integrating presumably conscious and subjective hedonic value into action plans, and may produce emotional responses to events as they occur. There is evidence that ACC lesions encourage the selection of less costly actions, and that dopamine input to the ACC is necessary for the selection of more costly actions.

The authors offer the hypothesis that the ACC supports the selection and execution of complex behaviours over time, such as the decision to run up a mountain rather than remain home on the sofa. Lesions to the ACC result in more immediate responses rather than the ability to carry out extended and often costly behaviours. Options supported by the ACC may provide excitatory input to the dorsolateral. The orbitofrontal is seen as being involved in this circuit by providing positive or negative evaluations for the options for actions found in the ACC. The ventral striatum of the basal ganglia is seen as supporting individual steps towards the longer term options selected by the ACC.

20 FEBRUARY 2012
CONSCIOUSNESS, THE SELF & ALTERED STATES

Neural correlates of the psychedelic state as determined by fMRI studies with psilocybin

www.pnas.org/cgi/doi/10.1073/pnas.1119598109

Keywords: psilocybin, altered states of consciousness, thalamus, cingulate cortex

INTRODUCTION: This paper further undermines the persistent claim within mainstream consciousness studies that all that needs to be done is to deconstruct the self (a relatively easy process) and then declare the consciousness problem solved. Altered states of consciousness have always appeared to contradict this claim, but the evidence of this was at a rather anecdotal level. Compiled by several prominent universities, this paper demonstrates that it is likely that the self can be deactivated by a drug while the subject continues to have conscious experiences.

In this paper, the authors use psilocybin, the active compound in so-called 'magic mushrooms' to study the transition from a normal to an altered state of consciousness. The authors were surprised that a markedly altered state correlated with a reduction in blood flow and BOLD signal, rather than their expectation of increased neural activity. The biggest reductions in blood flow and BOLD were observed in the thalamus and the cingulate cortex. The larger the decrease observed, the greater was the reported strength of the subjective experiences. In particular, psilocybin caused a large decrease in interaction between the medial prefrontal cortex and the posterior cingulate cortex. This is taken by the authors to imply that the altered state correlates with decreased connectivity between hubs involved in connecting and organising the brain.

In this study, subjects receiving psilocybin were compared to a control group that received a placebo. The psilocybin group demonstrated a significant decrease in cerebral blood flow in parts of the thalamus, the posterior and anterior cingulate cortex, the medial prefrontal cortex, the orbitofrontal, the frontal operculum and a number of other brain regions. The decreases were localised in associative regions or hub/connector regions such as the thalamus. In each area of the brain that was studied the decrease in blood flow correlated to the reported intensity of the subjective effect. A separate study based on the BOLD signal showed regional decreases in the same areas as those that saw a decrease in cerebral blood flow. The cingulate cortex and the medial prefrontal are seen as being particularly implicated in the action of psilocybin. A study with rats showed a decrease in local field potentials after receiving psilocybin.

The authors consider the results of their study unexpected and therefore in need of some explanation. Previous studies had shown an increase in brain activity in terms of glucose metabolism, and there has been as assumed connection between psychedelics and increased neural activity. The authors suggest that stimulation of serotonin transmission by the drug, leading to increases in GABA transmission, could in turn lead to the inhibition of pyramidal cells, and the observed deactivation in some brain areas.

The posterior cingulate and the medial prefrontal showed the most consistent deactivation under psilocybin and are also areas that have a 20% higher rate of metabolism than the rest of the brain. The study showed a decrease in interaction between the posterior cingulate, and some theorists have suggested that the posterior cingulate and the associated default-mode network have a role in the experience of the 'self' or self-consciousness. The default network of which the posterior cingulate is part also involves the largest concentration of cortico-cortical connections in the brain. Deactivation of such connections may relate to alterations in conscious states. This idea is seen by the authors as being consistent with Aldous Huxley's idea of the brain as a reducing valve. One possibility suggested is that deactivation in frontal areas such as the posterior cingulate leads to enhanced influence from sensory areas such as the parietal cortex.

From the point of view of consciousness studies, this paper further undermines the persistent claim in mainstream works that all that needs to be done was to deconstruct the self (a relatively easy process), and then declare the consciousness problem solved. Altered states of consciousness have always appeared to contradict this claim, but the evidence of this was at a rather anecdotal level. This paper based on research in prominent universities including Imperial College London indicates that it is likely that the self can be wholly or partly deactivated by a drug while the subject continues to have conscious experiences.

13 February 2012
REWARD AND DECISION CIRCUITS
Neural circuits of reward and decision making: Integrative networks across corticobasal ganglia loops

Suzanne Haber

In:- Neural Basis of Motivational and Cognitive Control – MIT Press (2011)

Keywords: reward circuit, dopamine, orbitofrontal, anterior cingulate, ventral striatum

INTRODUCTION: This chapter emphasises the degree of interface and cross-talk between the different segments of the reward circuit and also between the reward circuit and the cognitive and motor areas.

The authors see the reward circuit as a basis for decision making. This is supported by tests showing that rats would work for electrical stimulation in particular brain areas. The orbitofrontal, anterior cingulate, and nucleus accumbens are important components within this circuit, which also relates to dopamine producing neurons in the ventral tegmental area (VTA). Recent research extends involvement in the reward circuit to the whole of the ventral striatum in which the nucleus accumbens is located and also to the separate dopamine neurons of the substantia nigra.

The ventral striatum receives its main input from the orbitofrontal and the anterior cingulate plus dopamine from the VTA in the midbrain. The ventral striatum feeds back to the prefrontal areas via the thalamus. This whole arrangement is referred to as the corticobasal ganglia system and is central to the reward circuit.

The concept of the corticobasal ganglia system is relatively new in neuroscience. In the past the basal ganglia were seen only as part of the sensory-motor system, and were not related to the reward circuit. It is now, however, thought that the basal ganglia have separate loops for limbic, associative and sensorimotor functions. Adaptive behaviour in the form of action plans and inhibition of harmful behaviour is thought to depend on cross-talk between these loops. At each stage of the reward circuit there is communication between different parts of the reward circuit and also between the reward circuit and the associative circuit. Inputs from the ventromedial prefrontal cortex, orbitofrontal cortex and anterior cingulate terminate in sub-regions of the ventral striatum. The dorsal and lateral regions of the ventral striatum receive inputs from the orbitofrontal.

There is increasing evidence of interfacing between terminals of different cortical areas within this system. Thus projections from the orbitofrontal, ventromedial prefrontal and anterior cingulate converge within regions of the ventral striatum. It is suggested that coordinated activity of the orbitofrontal, ventromedial and the anterior cingulate produced by this convergence could generate the reward-based incentives for selecting particular options. The ventral striatum projects strongly to the dopamine neurons in the midbrain, and this is thought to play a part in the evaluation of rewards and the modification of responses to stimuli over time.

The reward network does not operate in isolation, but also interfaces with the cognitive and motor circuits. This integration occurs in convergence zones. In this way reward information can influence cognitive activity and motor action, once again facilitated by dopamine pathways.

7 FEBRUARY 2012
DOPAMINE AND MOTIVATION

The influence of dopamine in generating action from motivation

Mark Walton, Jerylin Gan & Paul Phillips

Neural Basis of Motivational and Cognitive Control - MIT Press (2011)

Keywords: dopamine, midbrain, basal ganglia, nucleus accumbens, opioid neurotransmitters

INTRODUCTION: The release of dopamine into the striatum and particularly the nucleus accumbens is closely related to the subjective evaluation of sensory inputs, and to the subsequent selection of behaviour and actions.

The authors start by referring to a distinction between the evaluation of reward, and the process of deciding to obtain, and then acting to obtain a reward. It is suggested that much twentieth century research fell short in not paying attention to the internal motivation of subjects. The authors acknowledge that several regions of the brain may be implicated; their emphasis here is concentrated on the striatum, particularly the nucleus accumbens and also dopamine projections.

The dopamine projection to the nucleus accumbens come from the ventral tegmental area (VTA) in the midbrain. Dopamine is a modulatory neurotransmitter often associated with the modulation of the excitatory neurotransmitter, glutamine. There is particularly dense innervation of the striatum by dopamine. Release of dopamine and availability of dopamine receptors in the nucleus accumbens is associated with drug addiction and also with compulsive shopping, eating and gambling.

A good deal of past research has concentrate on the role of dopamine in the selection of isolated rewards, rather than the more realistic situation of subjects assessing competing rewards and associated uncertainty as to the costs and probabilities of obtaining particular rewards. Recent studies, however, point to a correlation between the firing of dopamine neurons and the size and probability of particular rewards. Some studies also suggest a connection between dopamine activity and the timing of future rewards. Dopamine is seen as important in allowing the subject to exert the effort needed to obtain a particular reward. Dopamine release is viewed by the authors as facilitating, but not controlling, responses that seek potentially costly rewards. It is seen as a motivation to seek novel options and potential future rewards.

Evidence suggests that dopamine is involved in signalling the availability of reward. This is partly related to the prediction of reward, but also to actions directed towards gaining rewards. Additionally, the release of dopamine from the VTA can increase the probability of a reward being sought. In situations where there is conditioning, dopamine release can change from being directly related to the arrival of the reward, to being merely something that predicts the future probability of the reward. Dopamine activity can also increase where a reward is either above or below the predicted level, being thus an indicator for error predictions. The authors see dopamine in the nucleus accumbens as being important in making reward predictions when the subject is encountering an uncertain environment. However, this is viewed as only one influence on the subject's actions.

30 JANUARY 2012
ARGUMENT FOR THE PHYSICAL NATURE OF FREEWILL

Where has your willpower gone

Roy Baumeister, Florida State University

New Scientist, 28 January 2012

Keywords: Free will, will power, self control, emotions, neurotransmitters

INTRODUCTION: It is quite encouraging to find Baumeister writing on free will or self control in a popular science magazine, given that as a psychologist, he is a long way from the mainstream's reliance on a simplistic interpretation of the Libet experiments. Although the article is given a rather reductionist spin, stressing that will power is driven by glucose based energy, its arguments are fatal for the deterministic establishment view as to the non-existence of free will.

In contrast to the mainstream view that there is no such thing as free will, with unconscious and deterministic computations responsible for all human actions and behaviour, Baumeister argues that free will or self control requires energy, and is therefore part of the physical processing of the brain, rather than an illusion as the mainstream would have it.

Baumeister states that research demonstrates that when subjects have had to exert self control, they perform poorly on a subsequent test of self control. It is argued that energy is depleted by the first exercise of self control leaving less available for the second attempt.

In one such test, subjects were left next to a table with chocolate biscuits, which they were not supposed to eat. Some of the subjects succumbed to temptation and ate the biscuits. Subsequently, both the subjects who had succumbed and those who had resisted attempted a puzzle, which unbeknown to them was unsolvable. Those who had resisted the biscuit temptation gave up sooner on the puzzle, suggesting that their mental energy had been depleted by the effort of resisting temptation.

In this context, will power is compared to a muscle that can tire, although its full energy can return after a period of recuperation. Baumeister proposes that the energy driving will power is ultimately based on glucose that is the basis of neurotransmitters instructing axons to fire. A meta-analysis performed in 2010 showed that as in the tests mentioned earlier, subjects' performance deteriorated between a first and second self control test. However subjects dosed with glucose after the first test performed well on the second test.

26 JANUARY 2012
MATHEMATICAL PROBLEMS WITH INFLATION THEORY

Death of the eternal cosmos

Lisa Grossman/based mainly on Alexander Vilenkin

New Scientist, 14 January 2012

The fashionable theory of eternal inflation at the beginning of the universe has been used to explain both the fine tuning of the laws of nature and to allow for string theory having 10⁵⁰⁰ solutions.

Alexander Vilenkin of Tufts University has examined the equations relating eternal inflation to the Hubbke constant describing the expansion of the universe in Physical Review Letters, DOI: 10,1103/physrevlett.90.151301. The conclusion is that it is impossible to have a spacetime with this property, as the constant has a limit that prevents inflation. The same sort of constraint applies to the idea of cyclic universes going through endless Big Crunches followed by Big Bangs. From this Vilenkin concludes that there is no possibility of a universe that did not have a definite beginning.

25 JANUARY 2012
DESPAIR WITH POPULAR SCIENCE TREATMENT OF CONSCIOUSNESS
Perhaps we should despair of modern consciousness, studies or at least ban it from appearing in popular magazines. A quarter century on from the lifting of the complete taboo on mentioning consciousness in scientific circles, a popular article in a popular science magazine (which often does good stuff on other subjects) can come over as a mixture of error, misrepresentation in the early stages, followed by a move to peripheral topics which could never by themselves explain consciousness.

The first part of the article takes the familiar route of looking at Penrose's ideas, and then quickly demonstrating how they are wrong. Except unfortunately that these arguments are also in error. The ancient argument that microtubules can't support consciousness because they are present in all cells in the body and not just neurons is trundled out. While much of consciousness studies is painfully complex, here there is a simple answer, to the effect that microtubules are denser and more stable in neurons, making them more suitable for information processing than the cells in the rest of the body.

Not content with this, the article goes on to make a double misrepresentation as to why quantum consciousness theories are unpopular. The article correctly states that 'almost everyone researching consciousness rejects the quantum computing theory' but misrepresents the reason for this. It is claimed to be because invoking one mystery (quantum theory) to explain another (consciousness) gets you nowhere. This lets down the less well informed reader, for the reason that Penrose never proposed that because quantum theory was a mystery, it was a good basis for explaining another mystery. This was a criticism, or rather just a piece of ridicule coined by the philosopher, David Chalmers, in the 1990s, but tirelessly repeated by the more superficial critics of quantum consciousness. In fairness to mainstream consciousness studies, there are more serious reasons for arguing against quantum consciousness, but these are not touched on in this article.

Another misunderstanding here is the suggestion that Penrose's theory was proposing neural correlates of consciousness, whereas it was proposing the basis of mathematical understanding later extended to cover consciousness itself rather than a correlate, which is simply a feature found in the same times or places as consciousness. There is, an in itself interesting, section on building up knowledge of which brain areas react to which images an actual image of what the brain is looking at. This is a tremendous tour de force technically, but tells us precisely nothing about why these brain activities produce consciousness.

Having whiled away a good deal of space discussing potential correlates of consciousness, the article ends on a curious note. It is suggested that scientists will never find the correlates of consciousness. Why? Because, we are told, there is no difference between conscious and unconscious processing, and as a result the whole idea of consciousness, or possibly just the idea of consciousness being a problem (which of these isn't really clear) is a function of muddled thinking. This seems wrong on first principles, because we know that some systems of activity in the brain are necessary for consciousness, (for instance the global gamma synchrony) and systems of activity lacking this are not conscious. Being conscious is a different for the subject from not being conscious and is produced by different systems of activity (as opposed to specific brain areas referred to in this article), and on this basis it is impossible to refer to the two as being the same.