Eye movements reveal a dissociation between memory encoding and retrieval in adults with autism

Abstract

People with Autism Spectrum Disorder (ASD) exhibit subtle deficits in recollection, which have been proposed to arise from encoding impairments, though a direct link has yet to be demonstrated. In the current study, we used eye-tracking to obtain trial-specific measures of encoding (eye movement patterns) during incidental (natural viewing) and intentional (strategic) encoding conditions in adults with ASD and typical controls. Using this approach, we tested the degree to which differences in encoding might contribute to recollection impairments, or whether group differences in memory primarily emerge at retrieval. Following encoding of scenes, participants were asked to distinguish between old and similar lure scenes and provide 'remember'/'familiar' responses. Intentional encoding increased eye movements and subsequent recollection in both groups to a similar degree, but the ASD group were impaired overall at the memory task and used recollection less frequently. In controls, eye movements at encoding predicted subsequent correct responses and subsequent recollection on a trial-by-trial basis, as expected. In contrast, despite a similar pattern of eye movements during encoding in the two groups, eye movements did not predict trial-by-trial subsequent memory in ASD. Furthermore, recollection was associated with lower similarity between encoding- and retrieval-related eye movements in the ASD group compared to the control group. The eye-tracking results therefore provide novel evidence for a dissociation between encoding and recollection-based retrieval in ASD.

Reduced Hippocampal Functional Connectivity During Episodic Memory Retrieval in Autism

“…Notably, however, activity in LPFC predicted subsequent retrieval success only in the control group, not in the ASD group, a finding that was also apparent in inferior temporal cortex following a whole-brain analysis. This finding partially supports that of previous fMRI research (Gaigg et al 2015) as well as our recent observation of a dissociation between encoding-related eye movements and retrieval success in ASD (Cooper et al 2017). Based on the lack of difference in overall LPFC activity during encoding, a reduced relationship between encoding activity (presumably a marker of encoding processes) and memory could suggest that operations at encoding are less likely to determine retrieval success in ASD compared with typical controls, implying that retrieval processes may have a more pronounced influence on the episodic memory deficits observed here.…”

“…Research in ASD has yet to examine retrieval precision directly, but there are hints of a reduction in memory quality in terms of reduced specificity of autobiographical event details (Crane et al 2012), reduced memory salience (Lind et al 2014), and reduced confidence in correct memories (Grainger et al 2014). Furthermore, individuals with ASD show a reduction in the number of eye movements to previously studied regions of scenes during successful recollection (Cooper et al 2017), suggesting less reconstruction of event details. Findings of impaired source memory and a reduced ability to discriminate similar items could be partially driven by impoverished precision of memories, leading 2 sources or items to appear more similar in memory.…”

Spatial navigation from same and different directions: The role of executive functions, memory and attention in adults with autism spectrum disorder

“…Indeed, when looking at the relevance of executive functions for egocentric and allocentric navigation performance in the current study, perseverative errors in the IED task were a significant predictor, again emphasizing the relevance of the influence of executive functions in memory and navigation performance in ASD [see Maister et al, 2013]. A third important factor to consider is attention at encoding, as indicated by a number of previous studies [see Cooper, Plaisted-Grant, Baron-Cohen, & Simons, 2017;Gaigg, Bowler, Ecker, Calvo-Merino, & Murphy, 2015;Loth, G omez, & Happ e, 2011;Ring, Bowler, & Gaigg, 2017]. In the current study, ASD compared to TD individuals looked for a significantly shorter time and less often at the two landmarks marking the intersections, which may in part explain their later difficulties in remembering the landmarks on the item tests.…”

“…A third important factor to consider is attention at encoding, as indicated by a number of previous studies [see Cooper, Plaisted‐Grant, Baron‐Cohen, & Simons, ; Gaigg, Gardiner, & Bowler, ; Gaigg, Bowler, Ecker, Calvo‐Merino, & Murphy, 2015; Loth, Gómez, & Happé, ; Ring, Bowler, & Gaigg, ]. In the current study, ASD compared to TD individuals looked for a significantly shorter time and less often at the two landmarks marking the intersections, which may in part explain their later difficulties in remembering the landmarks on the item tests.…”

A Physiological Marker of Recognition Memory in Adults with Autism Spectrum Disorder? ‐ The Pupil Old/New Effect

“…Episodic and semantic memory can be compared directly using the so‐called Remember/Know (R/K) recognition procedure [Tulving, ], which asks participants whether they Remember (R—episodic memory) or Know (K—semantic memory) that they have encountered particular information in the past. ASD individuals show specific quantitative reductions in Remembering but not Knowing compared to typically developing (TD) individuals [Bowler, Gardiner, & Grice, ; Bowler, Gardiner, Grice, & Saavalainen, ; Bowler et al, ; Cooper, Plaisted‐Grant, Baron‐Cohen, & Simons, ; Gaigg, Bowler, Ecker, Calvo‐Merino, & Murphy, ; Souchay, Wojcik, Williams, Crathern, & Clarke, ; Tanweer, Rathbone, & Souchay, ]. When looking at this literature, it is, however, important to note, that the majority of R/K studies in ASD has utilized verbal materials (12 out of 15; e.g., Bowler, Gardiner, & Grice, ; Bowler, Gardiner, Grice, & Saavalainen, ; Bowler et al, ; Gaigg et al, ), one has used pictures [Souchay et al, ], one autobiographical memories [Tanweer et al, ], and one non‐meaningful kaleidoscope images [Massand, Experiment 4].…”