DeutschClear Cookie - decide language by browser settings
Redel, Petra (2009): Visual attentional assessment in mild cognitive impairment and Alzheimer’s disease based on a theory of visual attention. Dissertation, LMU München: Faculty of Psychology and Educational Sciences



In the following sections, English summaries of the three studies presented in this dissertation are given. For a detailed German synopsis of the present work, see chapter 8 (pp. 118 et seqq.). Research in the field of aging and dementia is a main concern as the population of elderly people is growing continuously due to increasing life expectancy and thus, an accumulative number of people who live well beyond 65 years of age run a risk of developing age-associated neurodegenerative disorders of cognitive function, such as Alzheimer’s disease (AD), emerging as a major health problem. The present work is based on growing evidence that deficits in visual selective attention occur early in the progression to AD (Foldi et al., 2002) and therefore might be present as the first non-memory deficits (Perry & Hodges, 1999) at the early prodromal stage of mild cognitive impairment (MCI; Petersen et al., 1999). The present dissertation was performed to contribute to the still ongoing debate as to whether certain aspects of visual selective attention are par-ticularly vulnerable or preserved, especially at the stage of MCI, and whether attentional func-tioning might be qualitatively and/ or quantitatively different from attentional performance at the AD stage on the one hand or normal functioning on the other hand. As theoretical basis, Bundesen’s theory of visual attention (TVA; Bundesen, 1990, 1998) was employed to assess several latent, mathematically independent and quantitative parameter es-timates which are derived from two highly comparable paradigms – computerized whole re-port and partial report of briefly presented visual letter arrays. Central conclusions arising out of TVA-based investigations (e.g., Bublak et al., 2005; Bublak, Redel, & Finke, 2006; Duncan et al., 1999; Duncan et al., 2003; Finke et al., 2006; Gerlach, Marstrand, Habekost, & Gade, 2005; Habekost & Bundesen, 2003; Habekost & Rostrup, 2006; Peers et al., 2005) point at four central strengths of this tool for attentional assessment – the quality criteria sensitivity, specificity, reliability and validity. Study 1: In AD, the amyloid cascade hypothesis (Hardy & Selkoe, 2002) assumes that rising plaque and tangle burden invokes loss of nerve cells through direct and indirect effects on synaptic, neuronal and neuritic function (see e.g. Cirrito et al., 2005), resulting in progressive intellectual decline. Thus, sensitive biomarkers loading functionally on the neural alterations invoked by AD from early on, might improve the possibility to identify at risk subjects in time, providing a chance for effective treatment (Shah et al., 2008). The first study (see chapter 4, pp. 31 et seqq.) examined whether cognitive parameters for estimating the capacity of visual attention might serve that purpose. Based on Bundesen’s (1990) TVA, visual information uptake was analyzed in 18 subjects with probable AD, 18 subjects with amnestic MCI, and 18 healthy elderly control subjects. Groups were matched for gender, age, and education. From a whole report task requiring ver-bal report of briefly presented letters, four parameters were derived, characterizing different aspects of visual processing capacity: perceptual threshold t0, iconic memory μ, processing speed C, and visual short-term memory (VSTM) storage capacity K. Comparison of these attentional parameters between groups revealed an elevation of the per-ceptual threshold already in MCI subjects, while processing speed and VSTM storage capacity showed a significant decline for AD patients, only. AD patients on medication with acetyl-choline esterase inhibitors had higher processing speed, but were still below the level of MCI patients. Perceptual threshold values were significantly correlated with disease duration, but not with cognitive measures. Conversely, speed and VSTM were significantly related to cog-nitive scores, but not to disease duration. In particular, VSTM storage was related to neurop-sychological tasks applying visual material (picture naming and visuo-construction), while speed showed an additional relationship also to measures of verbal memory. These results indicate a staged pattern of deficits affecting pre-attentive visual processing in MCI, and attentive processing in AD. They fit into the amyloid cascade hypothesis according to which the neuropathology of AD is characterized by a net accumulation and deposition of β-amyloid (Aβ) in the initial phase, giving rise to neuronal and neuritic dysfunction. Later, gradual neuronal loss and transmitter disturbances finally cause the increasing intellectual de-cline during further progression of the disease. A threshold elevation may thus be considered as a possible index of impaired neuronal functioning prior to cell death, while speed and VSTM deficits may be indicative already of a substantial loss of neuronal cell assemblies and a degeneration of neurotransmitter systems. Study 2: AD is the most frequent form of dementia which appears both as a familial and a sporadic va-riant. In the by far more frequent sporadic form, a genetic risk factor is also implicated, in that carriers of the apolipoprotein E ε4 allele (ApoE4) have a 3 to 15 times higher risk of develop-ing the disease, compared to non-carriers (Blennow, de Leon, & Zetterberg, 2006). Using an identical TVA-based partial report paradigm as in the present study, Finke et al. (2006) had found a close relationship between the severity of the underlying genetic pathology in another neurodegenerative, namely Huntington’s, disease and the direction and degree of spatial attentional weighting. Sensitive tools for assessing selective visual attention might serve as early cognitive markers in the course of AD and therefore enhance the identification rate of at-risk subjects at the MCI stage (Shah et al., 2008) as well as of subjects with underlying genetic risk (ApoE4). The second study (see chapter 5, pp. 60 et seqq.) aimed at examining whether attentional parameters of visuospatial and task-related selection are appropriate means for that purpose. Visual selective attention was investigated in 32 patients with amnestic MCI, 16 patients with probable AD, and 36 healthy elderly control subjects. Groups were matched for age, gender and educational level. In combination with Bundesen’s (1990) TVA, two mathematically in-dependent and quantitative parameter estimates were derived from a partial report of briefly presented letter arrays: top-down control of attentional selection, representing task-related at-tentional weighting for prioritizing relevant visual objects, and spatial distribution of atten-tional weights across the left and right visual hemifield. Compared to controls, MCI patients showed significantly reduced top-down controlled selec-tion which further deteriorated in AD subjects. Moreover, attentional weighting was signifi-cantly unbalanced across hemifields in MCI and tended to be more lateralized in AD. The ma-jority of patients was biased to the left. Across MCI and AD patients, ApoE4 carriers revealed a leftward spatial bias. The leftward bias was the more pronounced the younger the ApoE4-positive patients and the earlier disease onset. ApoE4-negative subjects showed balanced attentional weighting. These results indicate that impaired top-down control may be linked to early dysfunction of cortico-cortical networks connecting parietal and frontal lobes. Accompanying, an early inter-hemispheric asymmetry in temporo-parietal cortical interactions might cause a pathological spatial bias. As the inheritance of ApoE4 is associated with an interhemispheric imbalance in parietal cortical interactions, a pathological spatial bias may function as early cognitive marker for detecting subjects at risk for probable AD. Study 3: In the latter study, the TVA-based partial report paradigm proved to be a sensitive tool for ve-rifying that both, deficits in task-related selection and a pathological attentional imbalance, are already present at the early stage of amnestic MCI and increase further at the AD stage (see second study, chapter 5, pp. 60 et seqq.). It was hypothesized that these deficiencies in selective attention may result from an early disconnection syndrome and an interhemispheric imbalance in cortical interactions, respectively, in the fronto-parietal attention network, before gradual neuronal loss leads to further decline in selective attentional and intellectual functions at later stages. In the third study (see chapter 6, pp. 93 et seqq.), these hypotheses were tested by investigat-ing the relationship of both partial report parameters, top-down control α and especially the laterality index of attentional weighting wλ, to regional glucose metabolism measured by rest-ing-state positron emission tomography (PET) in a sample of 30 amnestic MCI or mild AD patients. Hypometabolism across all patients was slightly increased in the left hemisphere. Interestingly, the more reduced the metabolism in the left temporo-parietal junction (TPJ) the more pro-nounced was the top-down control deficit. Accordingly, hypometabolism in the left TPJ pre-dicted the magnitude of the spatial bias. Furthermore, relative hypometabolism in the left TPJ and left inferior parietal lobe (IPL) as compared to the right TPJ and right IPL, respectively, was correlated with direction and degree of spatial bias. Taken together, PET imaging results support the hypotheses that, one the one hand, early def-icits in task-related weighting may result from a fronto-parietal disconnection syndrome al-ready at the stage of MCI. On the other hand, very early AD seems to be also associated with an interhemispheric imbalance of metabolism, particularly in the temporo-parietal cortices, resulting in a correspondingly directed and distinctive visuo-spatial attentional bias. Conclusions and outlook: This dissertation intended to investigate the probable valuable contribution of the whole and partial report of briefly presented letter arrays based on Bundesen’s theory of visual attention (TVA; Bundesen, 1990, 1998; Bundesen, Habekost, & Kyllingsbaek, 2005) in assessing am-nestic MCI and AD patients in comparison to healthy elderly control subjects. The results of the three presented studies suggest a staging model of visual selective atten-tional impairments in MCI and AD. Deficits of pre-attentive processing (perceptual threshold t0), task-related (top-down control α) and spatial weighting (laterality index of attentional weighting wλ) were already detectable in MCI patients, while aspects of processing capacity (perceptual processing speed C and VSTM storage capacity K) were still intact. At a later stage of the disease, further deterioration of top-down control α and increasing lateralization of spatial weighting wλ accompanied impairments in perceptual processing speed C and VSTM storage capacity K. In conclusion, the TVA-based assessment of selective visual attention proved to be a sensitive diagnostic tool for revealing subtle deficits already at the stage of MCI which might exhibit the capability of an early cognitive marker for the identification of subjects at risk of AD. To address this question, this survey needs to be complemented by longitudinal studies.