Verbal paired associates and the hippocampus: The role of scenes

Abstract

It is widely agreed that patients with bilateral hippocampal damage are impaired at binding pairs of words together. Consequently, the verbal paired associates (VPA) task has become emblematic of hippocampal function. This VPA deficit is not well understood, and is particularly difficult for hippocampal theories with a visuospatial bias to explain (e.g., cognitive map and scene construction theories). Resolving the tension among hippocampal theories concerning the VPA could be important for leveraging a fuller understanding of hippocampal function. Notably, VPA tasks typically use high imagery concrete words and so conflate imagery and binding. To determine why VPA engages the hippocampus, we devised an fMRI encoding task involving closely matched pairs of scene words, pairs of object words and pairs of very low imagery abstract words. We found that the anterior hippocampus was engaged during processing of both scene and object word pairs in comparison to abstract word pairs, despite binding occurring in all conditions. This was also the case when just subsequently remembered stimuli were considered. Moreover, for object word pairs, fMRI activity patterns in anterior hippocampus were more similar to those for scene imagery than object imagery. This was especially evident in participants who were high imagery users, and not in mid and low imagery users. Overall, our results show that hippocampal engagement during VPA, even when object word pairs are involved, seems to be evoked by scene imagery rather than binding. This may help to resolve the issue that visuospatial hippocampal theories have in accounting for verbal memory.

Identifying the cognitive processes underpinning hippocampal-dependent tasks

“…Naming of the components was determined by the tasks that most strongly loaded on to each (see Table 2 for the proportion of variance explained by each component, Table 3 (Wechsler, 2008(Wechsler, , 2011, the Symbol Span and Digit Span tests measure executive function, working memory and attention (Wechsler, 2008(Wechsler, , 2009), and the F-A-S is reported as an executive function task (Strauss, Sherman, & Spreen, 2006). Finally, while the Abstract Verbal Paired Associates test is a verbal memory task (and aligns also with component 2), it is a more challenging task than the other verbal memory tasksas shown by the performance scores in Table 1 and has also been found to require the recruitment of frontal 'executive' brain regions (Clark et al 2018), suggesting that processing of Abstract Verbal Paired Associates may reflect general IQ as well as verbal memory. Component 4 involved three of our tasks of primary interest -scene construction, autobiographical memory (internal details) and future thinking, and the inclusion of the simple scene description task (which also loaded onto the perceptual component).…”

“…Recent work using functional neuroimaging lends credence to this idea by finding that high imagery concrete word pairs evoked hippocampal activity due to the use of scene imagery, while low imagery abstract word pairs did not (Clark et al, 2018). Another way to test this in the future would be to interrogate the explicit strategies that people use to perform different verbal recall tasks.…”

“…Verbal recall was assessed using the Rey Auditory Verbal Learning Test (RAVLT; see Strauss et al, 2006), and the Logical Memory and Verbal Paired Associates subtests of the WMS-IV (Wechsler, 2009). Two additional verbal recall tasks were also included (Clark, Kim, & Maguire, 2018). A limitation of the WMS Verbal Paired Associates task is its reliance on concrete, imageable words (Clark & Maguire, 2016;Maguire & Mullally, 2013).…”

Identifying the cognitive processes underpinning hippocampal-dependent tasks.

“…Matrix Reasoning and the estimate of FSIQ from the TOPF are designed to be measures of general IQ (Wechsler, 2008, 2011), the Symbol Span and Digit Span Tests measure executive function, working memory, and attention (Wechsler, 2008, 2009), and the F-A-S is reported as an executive function task (Strauss et al, 2006). Finally, although the Abstract Verbal Paired Associates test is a verbal memory task (and aligns also with Component 2), it is a more challenging task than the other verbal memory tasks—as shown by the performance scores in Table 1—and has also been found to require the recruitment of frontal “executive” brain regions (Clark et al, 2018), suggesting that processing of abstract verbal paired associates may reflect general IQ as well as verbal memory. Component 4 involved three of our tasks of primary interest: scene construction, autobiographical memory (internal details) and future thinking, and the inclusion of the simple scene description task (which also loaded onto the perceptual component).…”

“…Verbal recall was assessed using the Rey Auditory Verbal Learning Test (RAVLT; see Strauss et al, 2006), and the Logical Memory and Verbal Paired Associates subtests of the Wechsler Memory Scale IV (Wechsler, 2009). Two additional verbal recall tasks were also included (Clark, Kim, & Maguire, 2018). A limitation of the Wechsler Memory Scale Verbal Paired Associates task is its reliance on concrete, imageable words (Clark & Maguire, 2016; Maguire & Mullally, 2013).…”