Cognitive Psychology by K. Gilhooly, F. Lyddy, and F. Pollick (first edition) – Summary chapter 1

Cognitive psychology is the scientific study of how people and animals process information. It studies how we acquire information, store information in memory, retrieve information and work with information to reach goals. In all these cases, individuals deal with internal or mental representations.

There is a long history of cognitive psychology:

1. Ancient Greeks (500 B.C)
   Plato stated that there is ultimate knowledge and we only see representations of that. The ancient Greeks are also the founders of epistemology.
2. Empiricism and associationism (17th – 19th century)
   Empiricists stated that all knowledge comes from experience and the followers of the associationism stated that ideas and memories were linked by associations (e.g: cow and milk are associated and thus easier remembered together).
3. Introspectionism
   The followers of Introspectionism stated that all cognitive processes could be consciously reported by using introspection. The disadvantages of this are that it required a lot of training and could not be used with a lot of people, with children and with people with reduced mental capacities. The idea that all cognitive processes could be consciously reported was later debunked because a person is not able to report how that person perceives visual illusions. Besides that, reporting a process also has the potential to slow down the process.
4. Behaviourism
   This approach states that it is impossible to know what cognitive processes are active. The only observable things are the input and the output. It also states that all behaviour can be explained by reinforcement and punishment; all mental phenomena could be traced to behavioural activity. Behaviourists also state that language is learned by reinforcement and punishment as well.

Tolman was a behaviourist but was the beginning of the end for behaviourism. He stated that rats have mental maps or mental representations of a spatial layout. Tolman partially debunked the behaviourist approach of describing basic behaviour by his experiment showing latent learning; animals learning even though there was no reward. The behaviourist approach had many successes in accounting for basic animal learning, it was less applicable to complex mental phenomena such as reasoning, problem-solving, decision making and language.

An information processing approach is an approach for understanding mental activity, based on computing. Computer programs to solve suitable problems could be seen as comparable strategies that humans might use to solve the same problems. Strategies are systematic ways to carry out a cognitive task. A simulation program is a program which expresses a model of human thinking and involves programming computers to solve problems in a similar way to humans. The difference between a simulation program and artificial intelligence is that A.I tries to solve the problem as effectively as possible, without any attempt at mimicking human strategies.

Theorists using the information processing approach try to explain performance in cognitive tasks by using concepts of internal representations, which are transformed by mental operations. Internal representations are the mental representations of external objects and events. Mental operations are inner actions manipulating mental representations. Connectionism is an approach to cognition in terms of networks of simple neuron-like units that pass activation and inhibition through the receptor, hidden and output units. The link strength is modified through learning rules such as backward propagation, which modifies the weight on the links between units in a connectionist network, in response to error, to obtain the desired output. The basic components of a connectionist network are:

1. A set of processing units
2. Weighted connections between units
3. A learning strategy

Units affect each other by excitation or inhibition. Functionalists state that we don’t need to know the processes of the brain, as long as we know the functions and the functional properties.

The brain is divided into two hemispheres, connected by the corpus callosum. The outer layer consists of the frontal lobe, the primary motor cortex, the somatosensory cortex, the parietal lobe and the occipital lobe. Deeper in the brain are the thalamus, the hippocampus and the amygdala. At the base of the brain is the cerebellum. There are specific terms for determining the location of something in the brain:

1. Dorsal and ventral
   Dorsal means towards the top and ventral means towards the bottom.
2. Anterior and posterior
   Anterior means toward the front and posterior means towards the back.
3. Lateral and medial
   Lateral means toward the side and medial means toward the middle.

Broca’s area is vital for speech production. Language functions are strongly localized in the left hemisphere. Neuropsychology states that all specific cognitive functions are linked to a specific brain area. Phrenology attempted to localize psychological functions to bumps in the skull taken to reflect the growth of the brain in specific areas. Cases of double disassociation are cases in which patients can be found with opposite patterns of impairment in two functions (e.g: poor short-term memory but good long-term memory). This is particularly useful for localization. There is a difference between structural imaging and functional imaging. Structural imaging is methods showing brain anatomy and functional imaging are methods showing brain activity.

There are several structural imaging methods:

1. Magnetic resonance imaging (MRI)
   A magnetic field and provides high-resolution anatomical images.
2. Functional magnetic resonance imaging (fMRI)
   This measures the degree to which oxygen in the blood flow goes to certain brain areas. The brain exists of voxels, some sort of pixels of hundreds of thousands of neurons.

There are also several functional imaging methods:

1. Electroencephalography (EEG)
   This gives a record of function as a summary of electrical activity over a wide area of cortex, measured using sensors.
2. Positron emission tomography (PET)
   This is very good for localization, but it uses radioactive fluid.

fMRI uses reverse inference to draw conclusions, but this only works in cases where ‘if and only if’ applies. The default mode network reflects internal tasks and this is more active in rest than when focussing on external visual signals. In the infant's brain, there is limited evidence for the existence of the default mode network, but the activity in the default mode network is more consistent in children as they age.