How does cognitive neuroscience differ from cognitive psychology?

Introduction

Cognitive neuroscience is a discipline that attempts to determine the neural mechanisms underlying cognitive processes. Specifically, cognitive neuroscientists test hypotheses about brain-behavior relationships that can be organized along two conceptual domains: functional specialization—the idea that areas of the cerebral cortex represent functional modules that are specialized for a specific cognitive process—and functional integration—the idea that a cognitive process can be an emergent property of interactions among a network of brain regions, and thus that a brain region can play a different role across many functions.

Early investigations of brain–behavior relationships consisted of careful observation of individuals with neurological injury resulting in focal brain damage. The idea of functional specialization evolved from hypotheses that damage to a particular brain region was responsible for a given behavioral syndrome characterized by a precise neurological examination. For instance, the association of nonfluent aphasia with right-sided limb weakness implicated the left hemisphere as the site of language abilities. Moreover, upon the death of a patient with a neurological disorder, clinicopathological correlations provided information confirming the site of damage that caused a specific neurobehavioral syndrome. For example, in 1861 Paul Broca's observations of nonfluent aphasia in the setting of a damaged left inferior frontal gyrus cemented the belief that this brain region was critical for speech output. The introduction of structural brain imaging more than 100 years after Broca's observations, first with computerized tomography and later with magnetic resonance imaging (MRI), paved the way for more precise anatomical localization in the living patient of the cognitive deficits that develop after brain injury. The superb spatial resolution of structural neuroimaging has reduced the reliance on the infrequently obtained autopsy for making brain–behavior correlations.

Functional neuroimaging methodologies, broadly defined as techniques that measure brain activity, have expanded our ability to study the neural basis of cognitive processes. As technology has advanced, a number of these techniques have arisen, including single photon emission computed tomography (SPECT), positron emission tomography (PET), functional MRI (fMRI), and magnetoencephalography (MEG). Using these techniques, researchers can measure regional brain activity in healthy subjects while they perform cognitive tasks, and thereby link localized brain activity with specific behaviors. For example, functional neuroimaging studies have demonstrated that the left inferior frontal gyrus is consistently activated during the performance of speech production tasks in healthy individuals. Such findings from functional neuroimaging complement findings derived from observations of patients with focal brain damage. In order to provide the reader with the necessary background for understanding functional neuroimaging data, this article focuses on the principles underlying each of these techniques, as well as their relative strengths and weaknesses. Particular attention will be paid to fMRI, as it is perhaps the most widely-employed neuroimaging method. The article concludes with some of the novel ways in which these techniques are being combined in order to harness their complementary strengths to answer questions in cognitive neuroscience.

7.6 Unsupervised Learning Algorithms

Unsupervised learning algorithms are used to group cases based on similar attributes. These models also are referred to as self-organizing maps. Unsupervised models include clustering techniques and neural networks. Different algorithms use different strategies for dividing data into groups. Some methods are relatively straightforward, quickly dividing the cases into groups based on common attributes or some other similarity. The Two-Step clustering method differs somewhat in that an optimal number of clusters is determined in an initial pass through the data, based on certain statistical criteria. Group assignment is then made on a second pass through the data; hence the name “Two-Step.” Neural networks are more complicated than some of the other unsupervised learning algorithms and can yield results that are difficult to interpret.