Chapter 2 Summary

The technical approaches that have made it increasingly possible to link the inferences and concepts of cognitive psychology to their neural underpinnings are diverse and continually improving. The approaches that have been most effective to date fall into two major categories: brain perturbation and neuromonitoring.
Methods that perturb the brain in some way allow inferences to be made about the role of specific brain areas or brain systems in particular cognitive functions. Approaches that entail perturbation include the natural disturbances of brain function that arise from trauma, stroke, or disease; perturbations induced pharmacologically; and perturbations induced by electrical stimulation of relevant brain regions. The latter methods include direct intracranial electrical stimulation, mostly in animals, as well as stimulation of brain tissue through the skull using transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS).
Methods that measure neural activity during cognitive tasks provide information about the specific neural activity patterns that are engaged during the processing of a specific type of stimulus or the performance of a specific cognitive task. The major activity-measuring techniques associated with cognitive processes include invasive electrical recording in experimental animals, noninvasive electrical or magnetic recording in humans, and both noninvasive and invasive imaging methods that depend on altered metabolism and/or blood flow in active brain regions.
In single-unit electrophysiological recording, metal electrodes are inserted into the brain structures of experimental animals to measure the action potentials produced by individual neurons. Animal researchers are also increasingly investigating the nonspike fluctuations in the dendritic local field potentials.
Electroencephalography (EEG) is a noninvasive method for recording the brain’s electrical signals from the scalp, which reflects the volume-conducted dendritic field potentials. EEG and the event-related potentials (ERPs) that can be extracted from EEG data via time-locked averaging have been widely used in the study of human brain activity.
EEG and ERPs have counterparts in magnetoencephalography (MEG) and event-related field responses (ERFs), which measure magnetic-field fluctuations due to neuronal currents rather than the associated voltage fluctuations. These methods all have high temporal resolution but coarse spatial resolution.
Techniques of three-dimensional functional brain imaging have revolutionized cognitive neuroscience with their ability to visualize brain activity during cognitive task performance. The first of these to be widely used, positron emission tomography (PET), can localize brain activity during extended task blocks, but it requires the use of radioactive isotopes and has essentially no temporal resolution. Accordingly, it has been largely supplanted in cognitive research by functional magnetic resonance imaging (fMRI), which has much higher temporal resolution (although it is still much lower than that of electrophysiological methods) and can be performed in an event-related way.
Much work has been devoted to advancing analytic methods for fMRI in order to enhance the ability to make inferences from the brain activity patterns. Such methods include using multivoxel pattern analysis to examine consistent variations in the spatial patterns of activity locally within an area and using repetition suppression to examine the activations of intermingled neural populations within an area. Considerable research has also been devoted to developing methods to analyze the functional interactions between different active areas in the brain and how they work together to accomplish cognitive functions.
Optical brain imaging techniques are based on differences in light absorption and transmission in active brain areas that can be detected and imaged using optical recording devices.
With both brain perturbation studies and brain activity studies, the approach is to establish which cognitive functions are associated with certain neural structures or neural activity, and which are dissociated. This linkage of neural and cognitive processes is made even stronger by establishing double dissociations—that is, establishing that Task A is associated with neural region 1 but not with region 2 and that Task B is associated with region 2 but not with region 1.
Increasingly, different methodologies are being combined in multimethodological approaches that together can provide greater insight into the neural mechanisms underlying cognitive processes.