• increasing the fitness of the group.
Altruism obviously increases the fitness of the group, but decreases the fitness of individuals what
at first glance conflicts with the theory of evolution and natural selection. But there are three attempts
to explain why individuals decrease their fitness for the fitness of a group, namely
1. group selection,
2. kin selection and
3. reciprocal behaviour
which will be explained in more detail in chapter Evolutionary Perspective on Social Cognitions.
We will focus here on reciprocal behaviour with regard to problem solving.
Reciprocal Behaviour
Why should an individual behave altruistic if it cannot be sure whether its recipient will also
behave altruistic or not? Reciprocity is one explanation for these phenomena. That is, an altruistic
individual will only offer an altruistic act to an individual which is known to be altruistic and will
withhold altruistic behaviour to individuals which only act selfish. This exception prevents altruists
from extinction and allows them to spread in population, but it presupposes that both individuals
interact more than once and that they are able to recognize each other.
We can distinguish two types of reciprocal behaviour: direct and indirect reciprocity. The direct
one is an exchange of altruistic behaviour between the same two individuals ("I scratch your back and
you'll scratch mine") whereas the indirect one is between different individuals ("I scratch your back and
someone will scratch mine"). The latter is even more complicated to explain, but it is a fundamental
trait in our contemporary society. The basic idea to explain these phenomena is the development of
reputation in society. That is, altruists decide whether or not to interact with someone according to the
reputation of an individual.
(Iterative) Prisoner's Dilemma
The problem of cooperation is also topic in game theory a branch of applied mathematics where
players try to maximize their winnings. There are many convergences between the theory of reciprocity
and game theory, one famous example is the prisoner's dilemma. Two people A and B have been
captured by the police, they committed a crime but the police is not able to proof that they are guilty,
but they have enough evidence to arrest them for six months. Before A and B have been captured by
the police they both agreed to keep silent. At the police department they were questioned in separate
rooms and both have the choice to cooperate with his partner or with the police. If one betrays the other
he will get free and his partner will have to serve for ten years. If they both betray each other they will
both have to serve for two years. But if both keep silent, they only have to serve for six months.
Prisoner B
Wikibooks | 15
Chapter 2
Stays Silent
Betrays
Prisoner A serves ten
Stays Silent Both serve six months
years
Prisoner
Prisoner B goes free
A
Prisoner A goes free
Betrays
Prisoner B serves ten
Both serve two years
years
The dilemma is that both accused people do not know how the other has decided or will decide.
Regardless how the other will decide, confessing the crime will improve the outcome. If A betrays B, A
will get free or he will stay in prison for two years. If he does not betray B, A will stay in prison for six
months or ten years depending on however B decides. So obviously the best choice is to betray the
other. There is an interesting extension of the dilemma called iterative prisoner's dilemma. The game is
played again and again so it is possible to punish selfish behaviour in order to support altruism. One
good strategy for the iterative prisoner's dilemma is tit-for-tat. At the first round this strategy suggests
to cooperate with the partner. All other rounds one will do whatever the partner did in the round before.
If someone betrays his partner, he will get punished next round. On the other hand, if someone always
acts altruistic he will get paid back. This strategy is nothing but reciprocity.
Consciousness
When bringing Problem Solving and evolution together, explaining consciousness is an important
point to understand how we have come this far. The answer shall be given in three steps: (1) The
advantages that consciousness gave us during the evolutionary process. (2) The observations, through
which neuropsychology has approached consciousness. Observations of various kinds of impairment
like blindsight, commissurotomy, hemineglect, anosognosia, and also another approach called “binding
problem” which tries to explain how distributed activities of neurons make up conscious perception by
means of EEG monitoring. (3) Finally, what is probably the most controversial step, dealing with some
suggestions of how consciousness is involved in Problem Solving, namely (psycho-)functionalism,
metacognition and situation models.
Evolution of Consciousness
When trying to explain consciousness from an evolutionary perspective, there are two possible
options of approach. Either you specify the function of consciousness and thus give reasons for an
evolutionary progress or you explain how our abilities we gained through evolution made it inevitable
to make us conscious. Furthermore, it has to be considered at what time consciousness may have
appeared, that is where we can find consciousness in animals. While the first two theories presented
here will give reasons for why the function of consciousness has some benefits, the third theory is more
about the development of the brain that was not caused by any benefits of cognition but nevertheless
enabled the emergence of consciousness.
( direct source [1])
As a pioneer in this field, William James (1890)[2] argued that evolution pushes the behaviour of an organism into a direction that is of interest for it. The brain was seen as an instrument to make
predictions and therefore also having the ability to choose among many possibilities. So consciousness
16 | Cognitive Psychology and Neuroscience
Problem Solving from an Evolutionary Perspective
is involved in reinforcing the favourable possibilities while repressing the unfavourable. James assumes
that the evolution of consciousness happened at the same time at which the cerebrum had evolved. It
allowed to selectively guide the nervous system in an environment that became more and more
complex throughout evolution (1890/1891, p. 147)[2].
James distinguishes three classes of animal consciousness. The first class contains bilateral
invertebrates (earthworms, leeches, spiders, and insects) that show a centralisation of the nervous
system. The main criterion for this class is the differentiation between having a sensation and not
having that particular sensation. Although this can only be considered as a primitive mental state, the
detection of stimuli is thought to be a condition for consciousness. An example of scientific
investigation was done by Keunzi and Carew (1991)[3] showing that the marine snail Aplysia californica reacted differently to light from various directions and could also be trained to behave in a
certain way through conditioning.
The second class contains animals that do not only remember previous experiences but are also
capable of equate them with present experiences. They are able to copy a model or in other words
imitate a behaviour which is regarded as the beginning of conceptual thought. Here the animal class of
cephalopods should be mentioned. Octopus vulgaris which belongs to this class was examined by
Fiorito and Scotto (1992)[4]. They separated the octopuses into the “demonstrator” and “observer”
group. The “demonstrator” group was trained with conditioning techniques to attack either a white or a
red ball when offered both. Then an octopus of the “observer” group watched an octopus from the
“demonstrator” group attacking a ball with the specific colour. The “observers” imitated the attacking
behaviour rapidly, however they sometimes chose the ball with the respective other colour. Because of
this choice, it is assumed that a primitive form of consciousness is involved.
The third class entails humans as well as great-apes (gorillas, orang-utans, chimpanzees) and
cetaceans (whales, dolphins). In comparison to other animals they all possess a larger surface area of
the cerebrum, the neocortex. The main feature of this class is the capability of self-consciousness which
is according to James a more complex form. Gallup (1970)[5] introduced an experiment to find out whether animals or infants are able to distinguish “me” from “not me”. Red dye is put on the forehead
that can not be recognized except with the aid of a mirror. However, it is in question whether this
method can be seen as a test for self-consciousness in the sense that there is an awareness of one's own
thoughts.
These three classes were shortly presented because later, in the third part of the subtopic of
consciousness, we will introduce a general definition for consciousness which will give us a different
but quite similar classification and a suggestion of how to overcome this last problem.
( Blackmore, S. J., 2004 [6])
Another theory that contributes to the idea of consciousness having a function is held by Nicholas
Humphrey. The main thesis is that consciousness has a social function. While animals like
chimpanzees live in a social environment, humans are highly specialized to social skills. Humphrey
believes that skills like understanding, predicting and manipulating the behaviour of others became
necessary for our ancestors because, in a group, they were facing situations like deciding whether a
group member is a friend or an enemy or when they should form alliances etc. He calls our ancestors
having evolved this way “natural scientists”.
How consciousness could have had influenced this can be shown by considering the following
Wikibooks | 17
Chapter 2
scenario. Imagine an early hominid Suzy seeing Mick with a large piece of food, with him sits her
friend Sally. Suzy thinks about distracting Mick so that Sally could grab the food. But first she needs to
ask herself various questions like “Is Sally going to share the food with her”. In other words, she needs
to put herself into the position of another person. Thus our ancestors developed a self-reflexive insight
or an “inner eye” that gives us a perception like other sense organs, however, not of the outer world but
of one's own brain activities (1986[7]; 2002[8]). According to this theory consciousness can be attributed to social creatures like great-apes, elephants, wolves and dolphins. But it would also claim
that most creatures are not conscious.
The third theory which gains support from Robert Ornstein (1991)[9] is mainly about the necessity to withstand heat mainly for purposes of our head which means that the growth of the brain about 2
million years ago was caused neither by social nor cultural but physiological adaptation. The reason for
this assumption is that the increasing size of the brain happened in advance of human characteristics.
First, the importance of cooling the head shall be emphasized. Human beings are sensitive to high
temperatures because a rise of 1 or 2ºC above normal can disturb brain functions and a rise of 4ºC can
cause a heat stroke. In addition, human beings lack a cooling system like dogs for example, they have a
special blood circuitry which cools the blood when they start panting. One way to achieve cooling is an
upright posture. It is assumed that bipedalism was influenced by a climate change that due to its dry
conditions thinned the forests. Thus, hunting became a more reliable source for food than plants. But it
also caused the temperature to rise in our body. An upright posture had the effect that at noon the sun
hit a much smaller surface and more of our body mass is above the hot ground that had a vegetation of
50 cm (adopted from Peter Wheeler). After this evolutionary step it took about one million years until
the brain started to grow. Anthropologist Dean Falk assumes that during this time our ancestors
developed the net of emissary veins that lead in and out of the brain. In normal circulation these veins
carry heated blood from the brain out to the surface of the skull in order to cool it down. If the brain
temperature rises due to exercise like conditions, the blood flow in these veins reverses.
The second method our ancestors developed to withstand heat is increasing the cortical size
(Konrad Fialkowski, 1986)[10]. On the one hand this enhances the cooling effect of the emissary veins mentioned before. On the other hand the abundance of neurons made it possible that other areas of the
brain could take over tasks when there was a loss of neurons. An indication for such a development is
that a small piece of the cortex looks like any other piece and except for the size it is even similar to
that of other species. Also the density of the neurons is almost the same as for example in chimpanzees.
This development could explain how the human brain gained a parallel organization which is
preliminary for complex thought. The oversupply of supporting cells (glia) might have allowed more
interconnections between neurons due to the space they fill up.
Compared to the previous theories consciousness has no function but is a result of the evolution of
the brain. It is important to see that evolution as such is independent of intellect. We will come back to
this later in part three because this hint is essential for the definition of consciousness and will be
discussed in the short remarks on functionalism.
Neuropsychology and Consciousness
In philosophy there are many notions of consciousness. The topics of the second part are most
likely ascribed to phenomenal consciousness which is about our subjective experiences. However,
philosophers like John Searle and Thomas Nagel list three features of consciousness that seem to be
essential: subjectivity, unity and intentionality. First we will deal with unity also referred to as the
18 | Cognitive Psychology and Neuroscience
Problem Solving from an Evolutionary Perspective
“binding problem”.
The model of Wolf Singer[11] tries to explain the binding problem with a concept other than the classical one. The classical concept says that cognitive operation is the generation of explicit neuronal
representations. These representations are realized by individual neurons that are tuned to particular
constellations of input activity. Specificity is gained through selective convergence of input
connections in hierarchically structured feed-forward architectures. But according to Singer this view
has several disadvantages. It requires a high number of neurons and seems to be inappropriate for the
encoding of syntactical structure and hierarchical relations of elements composed in a perceived object
(Roelfsema, P.R., Engel, A.K., Koenig, P. & W. Singer, 1996)[12].
So the concept suggested by Singer is a distributed dynamical process which relies on self-
organization. It is assumed that neurons are associated with so called functionally coherent assemblies
that represent objects. The advantage is that one neuron can participate in different assemblies. Each
neuron is tuned only to a subset of elementary features (colour, movement, orientation). This concept
may be strong enough to explain phenomenal consciousness due to the combinatorial complexity and
flexibility. In contrast single neurons show little difference in sleeping or anesthetized animals.
Now two important questions arise. What is the mechanism of selection that dynamically separates
one assembly into two and how is an assembly labelled in order to be recognized by subsequent
processes. To give an answer to the first question there are three possibilities. First, the inhibition of
non-grouped responses, second, the selected response can be amplified and third, the selected cells fire
in synchrony. However, it is unlikely that the modulation of discharge rates (action potential) is
involved for the following reasons. An explanation of this type leads to ambiguities when considering
the second question of how neurons are labelled. In addition, the processing time would be too high
because for evaluation the action potentials first need to be integrated. Also different assemblies can
not co-exist in time if they share the same neurons. Otherwise they would not be indistinguishable. This
would only allow a sequential processing.
The main hypothesis is that selection and labelling is achieved through the synchronization which
comes in with several advantages. It is independent of the firing rate of single neurons and can be used
in parallel. Assemblies can follow one another much faster (Singer, 1999/2000)[13] and output activity has a high precision because of minimal latency jitter (Abeles, M. 1982[14]; Softky, W. 1994[15];
Koenig, P., A.K. Engel & W. Singer, 1996[16]). The processing speed increases because synchronized
EPSPs trigger action potentials with a minimal delay.
With this hypothesis some preliminaries for selection have to be considered. Neurons must be
sensible to detect coincident synaptic input. Further, they have to be able to coordinate rapidly in a
context dependent way. One example of neurons working with high precision is the auditory nucleus
where delays in the sub millisecond range are evaluated. Another example is the oscillatory responses
of retinal ganglion cells which are transmitted to cortical neurons (Castelo-Branco, M., S.
Neuenschwander & W. Singer, 1998)[17]. When awake these oscillatory patterns are in the gamma frequency range of 30-60 Hz (also see Crick, F. & Koch, C., 1990)[18]. In many experiments rapid synchronization has been observed in the visual cortex of cats. Fluctuations of the local field potential
shifted the response latency accordingly to the polarization of the potential. In other words, these sub
threshold oscillations can cause a delay in the response and thus are responsible for synchronization
tasks.
Coming back to the binding problem, we can examine it with the study of attention. Attention can
Wikibooks | 19
Chapter 2
facilitate synchronization. In one experiment cats were trained to react to visual stimuli with a motor
response. When they focused their attention, cortical areas that are involved in the execution of the task
synchronized their activity. Immediately after the stimulus was shown, synchronization further
increased. Thus, attention has the functional role of expectancy. It acts like a dynamic filter which in
advance selects neurons that participate in the execution and therefore accomplishes binding.
Now we will have a look an various brain damages which reveal additional insights in the subject
of consciousness.
Blindsight[6]
Lawrence Weiskrantz (1986[19]; 1997[20]) had been studying a patient called D.B. who lost vision in a large part of his left visual field due to the removal of a tumor that was in an area of his
visual cortex. In an experiment a circle containing stripes was shown to him in the blind field. He said
that he could not see anything within this area, for he was blind there. However, when he was asked to
guess whether the orientation of the stripes is either vertical or horizontal, he answers correctly in 90-
95 % of the time. (compare with hemineglect below)
Commissurotomy (split-brain)[6]
In the 1960s operations severing the corpus callosum had been carried out. This should prevent
epileptic seizures spreading from one hemisphere to the other as the corpus callosum is the primary
root where both hemispheres can interact. When the patients recovered they performed equally well on
problem solving tasks and language as before. When considering the visual pathway, a cut through the
corpus callosum prevents information from going from one side to the other (see picture). This a