Cognitive Functioning

Conclusion

Cognitive functioning and HRQOL have become important patient-oriented outcome measures in patients with brain tumors, next to traditional measures such as (progression-free) survival. New treatment options and combinations of treatment for brain tumors warrant further use of these outcome measures. An important rationale for the selection of a given brain tumor treatment may be related to cognitive side-effects and the impact on HRQOL. In long-term surviving patients with brain tumors, such as low-grade glioma, assessment of cognitive function has also become crucial.

Apart from its use in clinical trials, cognitive testing and assessment of HRQOL of the individual patient with a brain tumor may guide the physician in clinical decision-making. These measures may not only have prognostic significance but also, during follow-up, may give the first indication of recurrent tumor growth. The neurological handicap and the burden of epilepsy in patients with brain tumors should be considered as well.

Comprehensive questionnaires for HRQOL and comprehensive test batteries for cognitive function are needed for daily clinical practice. In addition, computer-adapted assessment of cognition and HRQOL are increasingly being developed towards this goal.

The prevention of side-effects of brain tumor treatment is becoming another challenge now that (subgroups of) these patients are living longer.

8 Conclusion and future directions

Cognitive functioning and cognitive performance are heterogenous among all individuals and, similarly, cognitive performance and cognitive decline are heterogenous among individuals with PD. The hypothesis underlying the studies summarized here is that genetics may explain some of this heterogeneity.

Studies focusing on PD genetics may be crudely categorized as focusing on known mutation carriers, i.e., genetic PD, or focusing on more common variants in idiopathic PD. Among carriers of pathogenic mutations in genetic PD, genotype may explain some componet of the different phenotypes and patterns of progressions, with regards to congition and cognitive phenotype, which have been observed and are described above.

Specifically, dementia is rare in PRKN-PD, less frequent in LRRK2-PD than idiopathic PD, and likely more common in GBA-PD and SNCA-PD. Interestingly, the stronger the association between genotype and α-synuclein pathology, the greater the risk for cognitive decline. In non-genetic PD, the most convincing data support APOE variants relationship to poorer cognitive outcomes.

We anticipate that as research advances, additional variants will be associated with a range of cognitive phenotypes. Specically, as investigations into polygenic risk scores expand further into correlating risk scores with clinical characteristics, we hypothesize that additional genetic contributions to cognitve progression in PD will be uncovered.

Cognitive domains and neuropsychological tests

Cognitive functioning comprises multiple cognitive domains, such as memory, language, visuoconstruction, perception, attention, and executive functions (Lezak et al., 2004). In addition, mental speed is an essential feature of cognitive functioning. These cognitive domains can be impacted selectively or can be affected by other noncognitive factors that contribute to task performance, such as motivation or mood. Detailed neuropsychological tests are available to assess performance on these cognitive domains. It is important to acknowledge, however, that most tests depend on multiple cognitive processes and therefore do not tap a single cognitive domain. For example, in a memory test where words must be remembered, attention is also needed to be able to perform the test. By combining multiple tests, the pattern of disturbances often reveals which domains are most likely to be affected.

The next paragraphs briefly address the three domains that are affected most often in patients with diabetes, namely memory, information-processing speed, and executive functions (Brands et al., 2005; Reijmer et al., 2010). For a more detailed description of cognitive domains and cognitive test procedures the reader is referred to neuropsychological textbooks (Lezak et al., 2004).

Memory is usually assessed with tests that require patients to recall a list of unrelated words that is presented to them repeatedly. They are then asked to recall these words immediately after presentation and again after a 30 minute delay, to assess immediate or short-term verbal memory and long-term verbal memory, respectively. “Working memory” is the ability to maintain and manipulate information for a period of several seconds. This is measured by the digit span task, where patients are asked to repeat a list of digits of increasing length in both the same way as presented (forward) and in reverse order (backward) (Wechsler, 1997). There are also several tests available that assess memory in a nonverbal manner.

Information-processing speed is a measure of the ability to process information within a limited amount of time. It is essential for all other cognitive processes and depends on the integrity of the cerebral network as a whole. Information-processing speed is tested using reaction time tasks, such as the Digit Symbol Substitution Test of the Wechsler Adult Intelligence Scale – 3rd edition (WAIS-III) (Wechsler, 1997). In this test the patient has to copy as many symbols as possible that match the digits on top of the paper in a period of two minutes. Information-processing speed declines naturally with age, but is also sensitive for cognitive dysfunction. The grooved pegboard test is a test of psychomotor speed that is particularly sensitive to the modest cognitive decrements that are associated with type 1 diabetes (Ryan et al., 2003; Brands et al., 2005).

Executive functions are needed to plan and organize all aspects of daily life. They are necessary to control our behavior, to set and monitor goals, and to inhibit automatic responses. They are thereby closely related to more complex attention processes. The domain is also described as “mental flexibility.” This domain can be measured by the Stroop Color Word Test (Stroop, 1935) for complex, selective attention and the Trail Making Test part B (Corrigan and Hinkeldey, 1987) for divided attention. The main item in these tests is that patients need to switch efficiently between multiple task components. Another widely used test is the Verbal Fluency Test (Deelman et al., 1981), where patients must generate as many words as possible starting with a specified letter of the alphabet or belonging to a specified category (for example, animals) within one minute.

Is depression related to cognitive functioning and structural and functional brain findings?

Cognitive functioning has also been shown to be associated with depression in PwMS in a number of studies now, especially the domains of complex attention and information processing speed, as well as executive functioning (Arnett et al., 2008, 2021). Not surprisingly, PwMS who have evidence of more structural brain damage report greater depression (Gold et al., 2014). In fact, one study found that nearly 50% of depression in MS was predicted by various measures of structural brain integrity, including lesions, brain atrophy, and the integrity of white matter tracts (Feinstein et al., 2010; Bakshi et al., 2000). The brain regions most often found to be compromised in MS depression are temporal and frontal areas (Arnett et al., 2008; Feinstein et al., 2010; Kiy et al., 2011; Gold et al., 2010). It is unclear, however, how such structural damage actually leads to depression in MS, that is, the mechanism by which this might occur. It may be that structural changes result in functional brain changes that in turn predict depression in MS.

Passamonti et al. (2009) examined functional brain activity in relation to depression in MS. They found that relapsing-remitting PwMS who reported greater depression than a control group showed decreased functional connectivity between the amygdala and the prefrontal cortex (PFC) when they performed an emotional matching task. It may be that such reduced functional connectivity reflects a disruption in an important affective processing system in the brains of PwMS early in the disease process that might ultimately put them at risk for emotional difficulties such as depression. Clearly more research is necessary to supplement this intriguing finding. Feinstein has speculated that “a combination of inflammation, demyelination (hyperintense lesions), and atrophy within medial inferior frontal areas would disconnect neural connectivity” in frontal-subcortical circuits and lead to depression. The Passamonti et al. (2009) study would seem to provide some support for decreased neural connectivity in such frontal-subcortical circuits in PwMS who report higher levels of depression than healthy controls.

An fMRI study examining depressed PwMS showed that those who were depressed had lower levels of connectivity between the dorsolateral and ventrolateral prefrontal cortices and also the hippocampus and amygdala (Riccelli et al., 2016). Also of interest, these investigators found that, when depressed PwMS were shown faces depicting sadness and anger, their lower connectivity found between the above regions was associated with increased activation in the prefrontal cortex.

Social engagement and cognitive decline

Preserved cognitive functioning is an integral component of maintaining a healthy, active, and independent lifestyle for older adults. Whether it is managing multiple medications, learning new skills and hobbies, or managing finances and paying bills, many everyday activities require complex cognitive processes. As a result, a better understanding of the protective effect that social engagement has on cognitive functioning is vital for designing solutions or interventions targeting declines in cognitive ability in older adulthood.

The most severe and debilitating form of cognitive decline is the development of dementia, with Alzheimer’s disease being the most common cause (Burns & Iliffe, 2009). Due to the high personal and societal costs associated with the condition, numerous studies have investigated the effect of social engagement on the occurrence and onset of dementia (e.g., Sörman, Rönnlund, Sundström, Adolfsson, & Nilsson, 2015; Wang, Karp, Winblad, & Fratiglioni, 2002).

Fratiglioni et al. (2000) conducted a study of 1,203 community-dwelling older adults (aged 75 and above). Social engagement was assessed by a measure that included both structural and qualitative aspects of individuals’ social lives. This included items about marital status, living arrangement, having children, as well as contact frequency and relationship satisfaction with various social ties (e.g., children, relatives, close friends). During a 3-year follow up after the baseline interview, it was found that poor or limited social engagement increased the risk of dementia by 60%.

A recent meta-analysis was carried out on the effect of social relationship factors on dementia risk in longitudinal cohort studies (Kuiper et al., 2015). The results of the meta-analysis were that individuals with lower levels of social participation, individuals with lower frequency of social contacts, and individuals with higher levels of loneliness were, respectively, 1.41, 1.57, and 1.58 times more likely to have a higher risk to develop dementia than their more socially engaged counterparts.

Even though dementia has been a main research focus, the benefits of social engagement for cognitive functioning in older adults are not solely confined to the onset of dementia. Social engagement has been found to associate with cognitive functions more broadly (Barnes, de Leon, Wilson, Bienias, & Evans, 2004; Seeman, Lusignolo, Albert, & Berkman, 2001). For instance, in a sample of 838 older adults without dementia, measures of social activity and social support related to higher cognitive functioning (Krueger et al., 2009). In this particular study, cognitive functioning was assessed using multiple measures of a variety of cognitive processes, including episodic memory, semantic memory, working memory, perceptual speed, and visuospatial ability. These results demonstrate that a high degree of participation in social activities and the maintenance of social connections serve to preserve an array of cognitive functions in late adulthood.

The relationship between social engagement and cognitive functioning in older adulthood could be bidirectional. That is, cognitive decline may cause a reduction in social engagement, because successfully maintaining interpersonal social relationships requires cognitive processing. Moreover, older adults experiencing cognitive decline could face barriers in participating in social activities, which would result in lower reported levels of social engagement. However, evidence suggests that enhanced social engagement exerts a protective effect on cognitive functioning for older adults. In one longitudinal study that used data from the Health and Retirement Study (Ertel, Glymour, & Berkman, 2008), linear growth curve models were used to determine if social engagement predicted subsequent memory decline. In fact, higher levels of social engagement did predict a reduced rate of memory decline over a 6-year period. However, no reverse causality was found (i.e., earlier memory scores did not predict subsequent social engagement), suggesting a causal path from degrees of social engagement to levels of cognitive functioning in older adulthood.

Mini-Dictionary of Terms

5.4 Cognitive Functioning

General cognitive functioning in individuals with PWS comprises a normal IQ distribution centered over a mean shifted downward by about 40 IQ points, meaning that most individuals have a mild to moderate ID (Whittington et al., 2004). Mathematical skills and short-term memory appear to be specific areas of cognitive deficit (Bertella et al., 2005; Stauder, Brinkman, & Curfs, 2002). Deficits in EF have also been demonstrated (Jauregi et al., 2007; Woodcock, Oliver, & Humphreys, 2009b).

Certain aspects of the PWS cognitive phenotype vary across different genetic subgroups. Those with a deletion show strength in performance relative to verbal IQ, whereas those with UPD show a relative strength in verbal IQ (Roof et al., 2000). The relative strength in visual processing in deletion PWS (Dykens, 2002) may be specific to the ventral processing stream (processing surface properties), while there remains relative impairment in the dorsal processing stream (processing location, Woodcock, Humphreys, & Oliver, 2009). In addition, verbal IQ is lower in those with a Type I compared to a Type II deletion (Milner et al., 2005; Whittington et al., 2004; Zarcone et al., 2007). These subtype comparisons have the potential to inform on more refined genotype–phenotype associations.

40-6 NEUROCHEMICAL THRESHOLDS AND COGNITIVE PROCESSING LIMITATIONS

Adequate cognitive functioning requires a relatively high level of neurochemical capacity. Functional (metabolic) “lesions” may represent depressed levels of neurotransmitters and/or receptor populations and thus form the biological basis of many processing limitations. Because the quality and quantity of receptor populations is a dynamic property, long-standing functional lesions may signal the potential restitution of impairments.

Processing limitations may be a primary cause of impairments at central levels of ongoing language activity (Friederici, 1995). Linebarger, Schwartz, and Saffran (1983) showed that agrammatic patients were able to detect syntactic anomalies, but were not able to use this knowledge on-line to produce well-formed sentences. Broca’s aphasic patients did detect syntactic violations in spoken language if sentences were simple, but not if they were complex (Haarman & Kolk, 1994). Martin, Dell, Saffran, and Schwartz (1994) studied a patient over a period of 6 years post-onset. His overall improvement in naming performance as well as change in error pattern was interpreted as a normalization of the decay time of activated lexical items. This demonstration of a restitution of function after long-standing specific processing limitations is as elegant as it is rare. Adequate lexical retrieval no doubt requires sufficient processing resources to occur. A restitution of processing capacity very likely indicates a change in neurochemical thresholds. This change may have been brought about indirectly, by brain reorganizations unrelated to the function under study, or more directly by specific behavioral and/or pharmacological treatments.

Luria et al. (1969) were among the first to appreciate the dynamic nature of cognitive deficits. They used a combination of pharmacological and behavioral treatment for seemingly permanent disorders and showed significant behavioral improvements by boosting levels of the neurotransmitter acetylcholine. It was further reported that the drug effects were stronger if combined with behavioral training. This beneficial effect of drugs as potential adjuvants to behavioral treatment was recently confirmed in a study reporting significant improvements in chronic aphasic patients receiving systematic language training in combination with a “cognitive enhancer” such as piracetam (Huber, Willmes, Poeck, Van Vleymen, & Deberdt, 1997).

The concept of biochemical thresholds for the recovery of different functions was investigated by Russell, Smith, Booth, Jenden, and Waite (1986). They injected into rodents a compound binding irreversibly to cholinergic receptors in the brain and reducing the receptor population to approximately 10% of its normal volume. The results showed that recovery, from an almost complete neurochemical lesion, occurred in a strict hierarchical and temporal order: physiological functions reappeared first and cognitive functions last. This hierarchical recovery matched the increase of a newly synthesized receptor population. Cognitive functions were observed only after the receptor population had reached 90% of its normal level. These studies indicate that biochemical thresholds and balances in damaged areas and their connections may constitute key concepts for understanding (a) cognitive processing impairments in the absence of extensive structural damage, (b) improvements without extensive structural changes, and (c) late recovery phenomena. Training programs may sometimes not be successful because they inappropriately tax the available processing capacity. Pharmacological and/or stimulation treatment preceding structured training may sufficiently alter neurochemical activity levels to make improvements possible and enhance the learning process. It has recently been shown that it is possible to significantly improve verbal memory of aged adults, if the subjects were administered a drug, sensitive to memory encoding, prior to the learning task (Lynch et al., 1997). These drugs (ampakines) promote the induction of long-term potentiation (Staubli et al., 1994) and thus change functional biochemical thresholds. The authors suggest that the drugs used may be particularly effective in the context of reduced memory processing. There probably is more potential for recovery than is often assumed in the field of cognitive rehabilitation.