Neuropsychological Basis of Neuropsychiatry
L. Clark
B. J. Sahakian
T. W. Robbins
Introduction
Neuropsychology makes an essential contribution to neuropsychiatry. It seeks objectively to characterize mental competence in component cognitive functions such as perception, attention, spatial cognition, memory, learning, language, thinking, and ‘executive’ function. Executive function is often associated with the functions of the frontal lobes, although these are not at all synonymous; we will pay special attention to this domain below, as it may be crucial to the understanding of several neuropsychiatric disorders. Neuropsychology is often conveniently divided into clinical neuropsychology and cognitive neuropsychology. The former is primarily concerned with the methodology and psychometric theory that lies behind the selection, administration, and interpretation of standardized psychological tests aimed at assessing deviation from the norm and an individual patient’s profile of strengths and weaknesses with a view to optimizing functional outcome and quality of life. Cognitive neuropsychology, by contrast is more concerned with the elucidation of cognitive processes through the study of patients, using both classical and newly devised tasks.(1) Neuropsychology also forms part of cognitive neuroscience, which has as its major goal the understanding of normal, as well as abnormal cognitive function, not only through the neuropsychological study of patients and healthy controls, but also using other techniques, including functional neuroimaging and the use of transcranial magnetic stimulation or psychopharmacology. In practical
terms, a neuropsychological assessment is often made together with a psychiatric examination; in addition to contributing to diagnosis it also helps to define the functional status of the patient.
terms, a neuropsychological assessment is often made together with a psychiatric examination; in addition to contributing to diagnosis it also helps to define the functional status of the patient.
Functions of neuropsychological assessment
These are perhaps easiest to define when there has been an organic brain injury causing a lesion; for example, due to a stroke, removal of a tumour, or a closed or penetrating head injury. It is clearly vital to have an accurate evaluation of a patient’s cognitive status so that his or her care can be optimized. This applies equally in neuropsychiatric disorders. In conditions like attention-deficit/hyperactivity disorder (ADHD) or prodromal Alzheimer’s disease, the cognitive examination provides essential information for making a diagnosis. Whilst acquired brain damage can lead to neuropsychiatric symptoms (e.g. depression or apathy), it may also be associated with specific patterns of cognitive deficit. However, most neuropsychiatric disorders are not associated with clearly defined brain injuries, but are instead hypothesized to result from the cumulative effects of neurodegenerative disease, neurotransmitter malfunctions, developmental hypoplasias, diffuse white matter lesions, or brain volume gains and losses. This is an important point, as it may require more sensitive new methodologies to characterize deficits produced by regional overactivity, or deficits associated with changes in regional connectivity, as distinct from brain lesions per se. In these examples, neuropsychology offers the opportunity to examine underlying pathophysiological mechanisms, in combination with other methods such as structural and functional brain imaging and evoked brain potentials.
Cognitive deficits (e.g. of memory) can exist in parallel to, and independently of, psychiatric symptoms (e.g. of melancholia) but they may also be intrinsic to them. For example, part of the psychiatric description of anxiety may emphasize abnormal attentional biases paid to threatening stimuli, which can be objectively assessed using cognitive testing.(2) Furthermore, some psychological factors may influence performance across a number of distinct cognitive domains, producing a broad profile of impairments from a relatively specific form of deficit. For example, patients with depression may show a ‘catastrophic response’ to receiving error feedback during testing, such that they are then more likely to respond incorrectly on the subsequent trial.(3) This interaction of emotional and social factors with cognitive processes forms an especially important part of neuropsychology as applied to psychiatry. Neuropsychology also enables the impact of neuropsychiatric symptoms to be assessed on functional status—whether the patient will be able to function in everyday life, and return to paid employment, and how rehabilitation may be best achieved. It is becoming clear that the effective treatment of certain symptoms (e.g. psychotic symptoms) can sometimes unmask profound cognitive impairments that are actually the main barrier to rehabilitation, as has recently been shown in schizophrenia.(4,5) Taking into account the corollary finding that cognitive status may be the best predictor of functional outcome and return to paid employment, this has made cognitive disabilities a new target for pharmaceutical treatment. Of course, with schizophrenia again in mind, it is also necessary to ascertain that medication is not associated with significant cognitive toxicity, for example in terms of sedation, distractibility, or impaired judgement. Neuropsychology must play a major role in providing such evidence. Sometimes neuropsychology may substantially contribute to the diagnosis itself of a psychiatric state, for example in dissociative disorders and in the study of fatigue disorders and malingering.
Principles of neuropsychological testing
Neuropsychological scores on most tests are standardized with respect to overall age, IQ, and ideally gender. Parallel forms of tests exist in different languages although cultural and ethnic factors are still difficult to take account of adequately. Testing is generally done in a quiet room without distraction by an experienced clinical neuropsychologist. Most tests have generally been shown to give consistent results when given by different testers and on different occasions to the same patient, when using standardized instructions. These ‘inter-tester’ and ‘test-retest’ forms of reliability are often critical factors in situations where testing has to be repeated, for example, in drug trials, or epidemiological studies.(6) Good test-retest reliability (i.e. r > 0.8) can be hindered by practice effects that markedly change how subjects approach the tasks. Such factors typically affect measures of executive function, where the subject may evolve strategies for dealing with the task over repeated test sessions. The various cognitive domains are usually tested in one or two sessions that contain an assortment of tests drawn from the types described above to provide a cognitive profile of the patient. The duration of the test sessions should be carefully considered in view of the concentration span and distractibility, or apathy of many patients. Neuropsychological test batteries normally comprise a selection of paper and pencil tests which are being supplemented increasingly with computerized elements. Computerized tasks benefit from the ease and accuracy of recording and analysing complex data (e.g. reaction times), and of standardizing the presentation of the test materials and trial-by-trial feedback. The Cambridge Neuropsychological Test Automated Battery (CANTAB), for example, utilizes a touch sensitive screen that allows the subject to interact directly with the test materials and obviates the need for divided attention between a video monitor screen and keyboard or desktop. Other possible advantages of computerized testing include the removal of the ‘confrontational’ or ‘interrogative’ elements of conventional testing, which may be especially advantageous when testing, for example, schizophrenic patients. The construction of batteries may also be affected by other factors such as the need to translate findings, presumably through the use of non-verbal tests, across species, based on the extensive knowledge of underlying neurobiological mechanisms gained through studies of non-human primates. Alternatively, there is a trend to customize neuropsychological batteries so as to focus on typical deficits in a given disorder, as illustrated by the recent derivation of the MATRICS battery for cognitive deficits in schizophrenia.(7)
Domains and neuropsychological tests of cognitive function
Neuropsychological assessment is made generally with respect to the overall profile of cognitive performance. For example, it is difficult to interpret an apparent memory deficit if the subject has a profound impairment in perception or attention. An important index of overall performance is the intelligence quotient, or IQ. The structure of intelligence is still being debated; whether there is a distinction for example between Cattell’s fluid and concrete
intelligence, and the possible existence of a general factor, Spearman’s g versus Thurstone’s more specific components.(8) Regardless of these theoretical issues, it is still useful to classify an individual in terms of their overall intelligence with a mean scaled score of 100 and a standard deviation of 15. IQ is often measured using sub-tests from the Wechsler Adult Intelligence Scale (or the child equivalent, the WISC), which broadly subdivides into verbal and non-verbal (‘performance’) components. The individual subtests include such categories as vocabulary, information, comprehension, arithmetic, digit span, similarities, block design, picture arrangement, picture completion, object assembly, and digit symbol, which thus probe a range of abilities from general knowledge and basic language skills to memory span, working memory, visuospatial construction, and psychomotor speed. As the time taken to make all these assessments can be prohibitive, a ‘prorated’ score based on a smaller selection of the tests is often employed, justified by the relatively high inter-correlation of performance among the 12 sub-tests. Intelligence is also measured effectively by the Raven’s Matrices, a set of visuo-spatial problems based on analogies.
intelligence, and the possible existence of a general factor, Spearman’s g versus Thurstone’s more specific components.(8) Regardless of these theoretical issues, it is still useful to classify an individual in terms of their overall intelligence with a mean scaled score of 100 and a standard deviation of 15. IQ is often measured using sub-tests from the Wechsler Adult Intelligence Scale (or the child equivalent, the WISC), which broadly subdivides into verbal and non-verbal (‘performance’) components. The individual subtests include such categories as vocabulary, information, comprehension, arithmetic, digit span, similarities, block design, picture arrangement, picture completion, object assembly, and digit symbol, which thus probe a range of abilities from general knowledge and basic language skills to memory span, working memory, visuospatial construction, and psychomotor speed. As the time taken to make all these assessments can be prohibitive, a ‘prorated’ score based on a smaller selection of the tests is often employed, justified by the relatively high inter-correlation of performance among the 12 sub-tests. Intelligence is also measured effectively by the Raven’s Matrices, a set of visuo-spatial problems based on analogies.
Frequently, it is useful to be able to gauge the patient’s premorbid intelligence level, before the onset of psychiatric symptoms. One way of estimating this is from years of education, as frequently employed in the United States. A second method depends on the National Adult Reading Test (NART), an instrument that depends on the subject’s ability to pronounce infrequent words; this correlates with educational level, and its utility was realized when it became apparent that patients with Alzheimer’s disease showed relative sparing of reading abilities, thus enabling their premorbid IQ to be captured by this test.(9) Alongside its US analogue, the Wechsler Test of Adult Reading (WTAR), the NART is now widely used to estimate premorbid intelligence in neuropsychiatric disorders including schizophrenia.
Superimposed on this general assessment of IQ is performance on tests of more specific abilities. The Wechsler Memory scale provides a method by which memory can be assessed in the context of overall intelligence. Some of its components, such as the Logical Memory test are still much used. However, with our burgeoning theoretical understanding of the components of cognition has come the introduction of ever more sophisticated instruments for measuring sub-components of cognitive performance. This brief chapter can but summarize some of the consequences of this, and encyclopaedic compilations of the various tests are now available, together with details of their mode of administration and interpretation.(10) However, the main domains of function that are generally evaluated are now outlined.
Over and above basic clinical sensory testing, measures of higher order perception in either the auditory or the visual modalities are available—for example, the Visual Object and Space Perception Battery.(11) Memory function is a controversial and complicated area of assessment. Most rapid batteries include tests of recognition memory, simply deciding whether or not a stimulus is familiar or not. Recent evidence, for example from the animal literature, suggests that such memory is relatively independent of hippocampal function, and depends instead on such regions as the perirhinal cortex.(12) Verbal recall, particularly a free recall, is considered to be a more demanding form of memory, thought to require the coordinated functioning of medial temporal lobe and frontal lobe structures. The frontal lobes are also implicated in retrieval of the temporal sequence or order of events, or the source of a particular memory in the past. Tulving’s distinction between episodic and semantic memory is still influential.(13) Episodic memory generally refers to the subjective reminiscence of an event that has ‘what, where, and when’ qualities, almost invariably associated with a person’s autobiographical memory. By contrast, semantic memory reflects memory for facts that may well have a different form of representation within the temporal lobe and is importantly influenced by verbal processes. In addition to these various modalities of memory material, are specific memory processes that have to be evaluated, such as encoding and retrieval. The latter is often assessed efficiently by the so-called verbal fluency tests, in which subjects have to retrieve words from a semantic category (e.g. animals) or beginning with a specific letter (e.g. F) in a set time period, usually of 60 s. Standardized tests of learning of verbal or non-verbal forms of material are provided by the California Verbal Learning Test (CVLT), the Rey Auditory Verbal Learning Test (RAVLT), and the CANTAB Paired Associate Learning (PAL) test. Emotional memory, which almost certainly implicates such structures as the amygdala, is another area of memory research that requires urgent development for application to psychiatry, for such conditions as Post-Traumatic Stress Disorder (PTSD). However, it is difficult to see how the current tests of traditional neuropsychology can be of much assistance here when the problem is to quantify the disruptive effect of a specific memory that has become over-salient rather than inaccessible.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
