A small study shows how those on the road to Alzheimer’s show early semantic problems long before memory problems arise, and that such problems can affect daily life.

The study compared 25 patients with amnestic MCI, 27 patients with mild-to-moderate Alzheimer's and 70 cognitively fit older adults (aged 55-90), on a non-verbal task involving size differences (for example, “What is bigger: a key or a house?”; “What is bigger: a key or an ant?”). The comparisons were presented in three different ways: as words; as images reflecting real-world differences; as incongruent images (e.g., a big ant and a small house).

Both those with MCI and those with AD were significantly less accurate, and significantly slower, in all three conditions compared to healthy controls, and they had disproportionately more difficulty on those comparisons where the size distance was smaller. But MCI and AD patients experienced their biggest problems when the images were incongruent – the ant bigger than the house. Those with MCI performed at a level between that of healthy controls and those with AD.

This suggests that perceptual information is having undue influence in a judgment task that requires conceptual knowledge.

Because semantic memory is organized according to relatedness, and because this sort of basic information has been acquired a long time ago, this simple test is quite a good way to test semantic knowledge. As previous research has indicated, the problem doesn’t seem to be a memory (retrieval) one, but one reflecting an actual loss or corruption of semantic knowledge. But perhaps, rather than a loss of data, it reflects a failure of selective attention/inhibition — an inability to inhibit immediate perceptual information in favor of more relevant conceptual information.

How much does this matter? Poor performance on the semantic distance task correlated with impaired ability to perform everyday tasks, accounting (together with delayed recall) for some 35% of the variance in scores on this task — while other cognitive abilities such as processing speed, executive function, verbal fluency, naming, did not have a significant effect. Everyday functional capacity was assessed using a short form of the UCSD Skills Performance Assessment scale (a tool generally used to identify everyday problems in patients with schizophrenia), which presents scenarios such as planning a trip to the beach, determining a route, dialing a telephone number, and writing a check.

The finding indicates that semantic memory problems are starting to occur early in the deterioration, and may be affecting general cognitive decline. However, if the problems reflect an access difficulty rather than data loss, it may be possible to strengthen these semantic processing connections through training — and thus improve general cognitive processing (and ability to perform everyday tasks).